Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: oatmeal769 on October 23, 2009, 10:41:36 am
-
Hello friends,
I have a 35 year old Alembic F2-B Pre-Amp. I play bass through it, and nothing else sounds as good to me. I've been using it so long, I don't know that anything ever will.
The last few years I've noticed the sound seems to have progressively dulled and lost it's life. The unit also has a little hum, and is pretty noisy. I was wondering if anyone could take a look at a picture and / or Schematic (I have both) and suggest the items I can replace which will restore it, maybe even make it sound better than new. All the resistors, etc. came with it originally, and it may have been re-capped back in the 80's but nothing has been done to it since then.
Here are the things I'd like to add or update with it:
It has an old 2 prong electrical cord with no safety ground - That can't be good when beer gets spilled on it, especially while I have my lips stuck to a metal microphone.
I'm feeding the input with active EMG pickups. They have 200 ohms output impedance. Should I put a resistor on the input? I've read the F2-B really only likes passive pickups.
I'd REALLY like to make the main output a balanced 3 conductor connection instead of the unbalanced 1/4" one currently there, is that possible? All of the other gear in my rack is balanced in/out, and I use the rig in some really noisy environments!
I'm feeding a balanced TRS 10K ohms input (a compressor) from this unit. I don't really think I need more gain, I'd just like to ensure no hum from lighting, electrical, etc.. As well as giving the compressor the drive it wants to operate well.
=====================================
* Here is some advice I've been given so far elsewhere, on another forum. Can anyone add to or suggest changes? I'd really like balanced but it's starting to sound beyond my abilities.: *
" ... Tubes and caps are the most likely for hum.
The orange ones are not electrolytics and do not need replacing. All cracked resistors obviously need replacing
Every one of the electrolytics needs replacing. The yellow ones and the silver ones.
If you want to quiet the noise, replace one of the resistors in the PS with a cheap choke -- it'll do wonders. - Replace the first 1K5 resistor with something like a hammond 154G
Replace the power cord! Particularly since you plug an instrument that you touch in to this, that should be fixed asap.
If you want a balanced output, use an output transformer -- Edcor would be good here, something like the XSM10K/10K would likely work well and could just be bolted to the output.
there is an issue which is that the Z out of the preamp is like 50K and won't likely do a good job driving the transformer's inductance. I think you should either work another tube in there (6T4 perhaps?) to lower the Z out so it can actually drive something like a cable, or the next stage, or you should switch from the 12ax7 to something with a lower rp -- maybe a 12ay7 or 12at7. The former will increase the signal a bit -- a 0.1u coupling cap will be too small in either case.
Assuming you do lower the impedance, on the primary you would connect out to + and - to ground (or the cathode if you like), and leave the input ct unconnected. For out, the "right" way to do it is to connect the + and - transformer taps to hot and cold, and connect the chassis (now grounded to safety ground) to the "ground" pin on the xlr.. ... "
Thanks for any input!
-
(http://www.geofex.com/FX_images/alembpre.gif)
Sure does look like a Fender front end don't it?
Replace all the electrolytic caps. I'd go ahead and replace the diodes too with 1N4007. That would be a dollar well spent, especially if they're old germanium diodes which I'm fairly certain they are. That should fix the hum. A grounded power cord is always a good idea.
Zout is roughly 37K. It's also coming off a plate. Not exactly ideal for driving a loZ output but you might have luck simply hanging a transformer off the back end. Adding a cathode follower stage between the existing output and the loZ transformer would be the right way to do it. The Cinemag CM-2810 is the bomb when it comes to loZ output transformers. Don't forget a polarity switch and a ground lift switch if you tread into these waters.
-Richard
-
...It has an old 2 prong electrical cord with no safety ground
...it may have been re-capped back in the 80's
Like buttery says, cap job and grounded cord
The last few years I've noticed the sound seems to have progressively dulled and lost it's life.
and... have you ever replaced the tubes?
The unit also has a little hum
hopefully that will go away with a cap job, wait and see
and is pretty noisy.
replace it all with metal film resistors will cure it (you may loose some mojo though...)
That can't be good when beer gets spilled on it,
My god man, that's beer your spilling! Lick it up off the amp before you kiss the mic
-
I wanna toss out that I question the correctness of the schematic I posted. I've never seen, much less been inside of a F2-B, but I have protoboarded this circuit when developing the Panna Maria. At the input, the 1M grid leak would be found in front of the grid stopper in all Fender amps. The 68K grid stopper produces a f-3dB of roughly 30kHz. During my experiments, I changed this to a 33k which brightened thangs up, but my customer quickly dismissed the circuit for having excessive gain and for sounding like a Fender. Another thang that bothers me is the output coming off the plate. Not good for driving a cable more than 2 feet long. None the less, the F2-B has good collector's value. Personally I wouldn't hack on it. Give it a cap job, new diodes, and put a grounded cord on it and see where she stands. You might need to change a plate resistor or 2 if it's crackling. Check the power supply resistors too and make sure they are within tolerance.
Please post pictures.
Oh yeah... welcome to the forum.
-Richard
-
Sure does look like a Fender front end don't it?
Yep, it was adapted straight from the front end of a Fender Showman Blackface.
Replace all the electrolytic caps. I'd go ahead and replace the diodes too with 1N4007. That would be a dollar well spent, especially if they're old germanium diodes which I'm fairly certain they are. That should fix the hum. A grounded power cord is always a good idea.
Which caps are electrolytic caps, and what should I replace them with? I was told "Replace Everything ORANGE" (see pictures)
Oh, I forgot to mention - I'm really good with a soldering iron, but I know next to nothing about circuits and how they work. I know how to point out certain components, and how not to get electrocuted but that's about it. I'm a tinkerer, but not all that knowledgeable about the subject.you gotta dumb it down for me somethimes...Zout is roughly 37K. It's also coming off a plate. Not exactly ideal for driving a loZ output but you might have luck simply hanging a transformer off the back end. Adding a cathode follower stage between the existing output and the loZ transformer would be the right way to do it. The Cinemag CM-2810 is the bomb when it comes to loZ output transformers. Don't forget a polarity switch and a ground lift switch if you tread into these waters.
I'm driving a balanced TRS 10K ohms input (a compressor) from this unit. The cord goes from the unbalanced Alembic main output to the balanced input of the compressor. The cord I'm using is an unbalanced shielded cable about 1 foot long... Do I really even NEED to balance? It sounds like more trouble than it's worth, and as you say, it does have collector value. But I do want to be sure I'm feeding a quiet signal at the exact right amount that the compressor would like to see.I wanna toss out that I question the correctness of the schematic I posted. I've never seen, much less been inside of a F2-B, but I have protoboarded this circuit when developing the Panna Maria. At the input, the 1M grid leak would be found in front of the grid stopper in all Fender amps. The 68K grid stopper produces a f-3dB of roughly 30kHz. During my experiments, I changed this to a 33k which brightened thangs up, but my customer quickly dismissed the circuit for having excessive gain and for sounding like a Fender.
It's known to some that the commonly available schematics on the web contain an error or two, the 68k thing sounds familiar... But the schematic looks like a lot of chicken scratch to me. Again, I can kinda identify components, and see the routes but schematics give me a headache.Another thang that bothers me is the output coming off the plate. Not good for driving a cable more than 2 feet long. None the less, the F2-B has good collector's value. Personally I wouldn't hack on it. Give it a cap job, new diodes, and put a grounded cord on it and see where she stands. You might need to change a plate resistor or 2 if it's crackling. Check the power supply resistors too and make sure they are within tolerance.
If I replace one resistor, I may as well do all, right? They aren't expensive, even the fancy ones, right?Please post pictures.
I'll post some now for analysis, and then after to compare.Oh yeah... welcome to the forum.
Thanks guys for a great welcome! I look forward to learning a lot.
One other question:
I'm feeding the input with active EMG pickups. They have 200 ohms output impedance. Should I put a resistor on the input? I've read the F2-B really only likes passive pickups.
.
-
have you ever replaced the tubes?
Yes, I have a stable of some old Mullards and Telefunkens, as well as some 70's RCA's that sound great.hopefully that will go away with a cap job, wait and see
I was always told that hum could likely be Caps. replace it all with metal film resistors will cure it (you may loose some mojo though...)
Have you a link? I don't understand the difference, I just know the ones in there now look ancient. I certainly don't want to lose mojo, my playing needs all the help it can get! My god man, that's beer your spilling! Lick it up off the amp before you kiss the mic
Sir, I don't know you but are you accusing me of a spillin' BEER?? I'd never dump perfectly good beer into a useless piece of metal and silicon - even if it was foreign beer. Heck, I'd lick it offa the circuits even if I'da done it. Them's wrasslin' words. Besides, who said it was MY beer that done got spilt??
J/K, I'm a wise ass!
-
Your 4th picture, Caps.jpg are the caps you want to replace. I notice 3 thangs:
These caps are 20uF at 450V and the schematic I posted shows 47uF.
The schematic shows 4 caps but yours has 5, one of which is mounted in an "after thought" fashion.
Seems like these caps should be radial leaded (both leads coming out 1 end) instead of axial leaded (1 lead out each end).
Apparently Alembic still makes this unit. Wouldn't surprise me 1 bit if the later units have 47uF caps instead of 20uF caps. Yours looks like a 70's vintage unit to me. I can't really tell from the pictures why there's a 5th cap. At a guess, I think somebody has done some mods. That unit isn't supposed to have a loZ out, but yours does (last pic). The loZ stuff looks after-market to me whereas the XLR jack is screw-n-nut and everythang else is pop rivet. Looks like the input grid stoppers are in fact 68K and that they are in fact in front of the 1M grid leak resistors. Hmm.... dual input jacks per channel. Is there a 68K on each jack or only 1 per pair of input jacks?
There shouldn't be a problem using active pickups with this unit unless your pickups have DC at the outputs (which they shouldn't). You could put a series bypass cap on the input, but that's gonna totally jack with the tone.
The diodes do not look like they're old germanium so don't worry about replacing them. It wouldn't hurt to measure the 3 resistors (red violet brown and 2 brown green red) in the cap picture. You'll need to disconnect 1 end of each to accurately measure these. I doub't there's a problem here since there is very little load on the power supply but.... it is the power supply and it's always a good practice to verify the power supply is healthy. Look at the 20uF cap connections for any tell tale signs of replacement (ie the solder joints look messy compared to the rest of the board).
Personally, I'd replace as little as possible and make every attempt to keep it looking as original as possible. A cap job and a grounded cord might offend some collectors but screw them. A safe working unit is more important. That's a cool box.
-Richard
-
The schematic shows 4 caps but yours has 5, one of which is mounted in an "after thought" fashion.
the schematic seems to be something someone drew up, other than alembic - it is possibly a drawing of a later production unit, or unit someone already modified.
Seems like these caps should be radial leaded (both leads coming out 1 end) instead of axial leaded (1 lead out each end).
or perhaps not - look closely, there is a trace where the (-) of the caps lands on the PCB - in early 70's, axial units were much more common in HV. the dates codes are certainly early 70's - having seen how other alembic equipment was constructed around that time, id be willing to bet that unit is stock, other than perhaps the lo-Z out - still, it is possible that it's a factory hack too. i'm betting it is.
don't add a choke - leave it stock, but as buttery says, do check PS resistors and the plate resistors for drift - CC resistors will drift up in value, never down.
you do have a hum balance pot - check it as well.
before you start, be aware that 60-70s PCBs are usually really crappy, in that it's easy to lift the foil off the board and you could lose a pad or two - use a solder sucker, keep the heat time on the pads to a minimum, and DON'T roll the iron tip on the pads. use a quality low watt soldering iron suitable for PCB work, or soldering station. e.g. weller.
note the polarity of caps and diodes BEFORE you pull them. you already have pics for orientation of the caps and one of the diodes, so make sure at least you have a good pic of the other diode. you could also do it old fashioned way - mark the PCB or draw.
at the very least, check the the PS diodes. personally, i'd replace them w/ 1N4007s. you'll need to lift one end of the diodes to check them. while you have the ends up of (or pulled) diodes, pull all the PS caps and measure the PS resistor chain, (the two 1.5k) if it's more than 3k (>10%) then assume one or both of the resistors has/have drifted - check the 270 ohm resistor as well.
while you have all the PS caps out, pull the tubes (moot) and measure the plate resistors, again if >10% they should be replaced. next, pull the bypass electrolytic caps, measure the cathode resistors.
now tack in all the new parts, double check your work.
the thing that will probably irk you the most: it is likely you'll have to open up the hole for the AC power cord entry, if you upgrade your unit to a 3 wire grounded power cord. suggest a uni-bit and clamp the unit down on a bench before drilling.
oh yeah, welcome! great bunch of folks here and a very gracious host.
PS. doug sells sprague atom-lytic caps (use the 500V caps) and the diodes - click here (http://www.hoffmanamps.com/MyStore/perlshop.cgi?ACTION=enter&thispage=parts4.htm&ORDER_ID=!ORDERID!)
-
Holy smokes! Richard and ISOTone, I am SO SORRY for wasting your time! Those are the wrong pictures. :embarrassed: I opened the wrong folder, and haven't seen the inside of mine in a while.... The next post will be pictures I just took today.
Again, BIG apologies, I really value the suggestions you're making.
The blank output jack in my new pictures is a jury rigged jack for a footswitch on the outputs, (obviously added later) from which I removed the wiring to the outputs.
The caps have been replaced once before, as evidenced by different solder work.
I think there's been some tinkering with resistors on inputs as well.
The ones on the output have been replaced, but I don't know if they are correct or even belong there.
The 1/4" out jacks were all replaced by me - (late one night on a bus after a gig and a few beers if I remember correctly, LOL) because the stock ones had worn out.
Would there be any benefit to replacing the power transformer with a newer torroidal unit, instead of inserting a choke? I think I read that torroids are much quieter?
Is the 'hum balanced pot' adjusted by a meter, or just 'by ear'?
Since it's already been modded, I don't mind modding it further as the goal is better quieter sound. It's been with me on almost every gig and studio situation for 20 years, that's not gonna change, it's a work horse, not a showpiece. It could likely be restored to original fairly easily as well.
Someone (not me) also had pulled out the grommet and pulled the AC cord through a few times forcefully, enlarging the hole. I have the right tool - a panel punch - to make a clean one and replace it with a modern round cable.
(http://farm3.static.flickr.com/2227/4043992882_161e026a22_b.jpg)
(http://farm3.static.flickr.com/2749/4043246531_f63c3a4175_b.jpg)
(http://farm3.static.flickr.com/2755/4043246091_14f541fefb_b.jpg)
(http://farm3.static.flickr.com/2427/4043247877_d5f72410a6_b.jpg)
(http://farm3.static.flickr.com/2542/4043245089_139e7e14cd_b.jpg)
-
No problem. (1)Notice that the power caps are magically 40uF now. My guess is that this unit is a little newer but still 70's vintage. From your pictures, it does not appear that the power caps have been replaced. (2)The diodes look like 1N4007 so they're probably fine. If you get in there to replace the caps, there's no harm in replacing those diodes. They're cheap. Be sure you pay close attention to polarity of the diodes and caps. (3)These are the power supply voltage dropping resistors. I'd flat out replace these with metal film instead of carbon comp. They're in the power supply so flame proof is a plus and you won't lose any carbon comp mojo since they're not in the audio circuit. (4)This is the hum balance pot. It sets the point at which the filaments are grounded. You set it by ear. Typically centered is the place to be. (5)Please fix this with a proper strain relief and grounded cord. Take the green wire to the chassis right by the power transformer. Now... the existing wiring is neutral fused which works but is not preferred. I'd rewire as follows. Take the white wire to 1 line lead of the power transformer (doesn't matter which one). Take the black lead to the tip of the fuse. From the side of the fuse, run a wire to the power switch. From the other terminal of the power switch, take a wire to the remaining transformer primary lead. (6)There is the possibility that the cathode bypass caps are failing. They are electrolytic caps too. The one furthest to the right looks suspicious on the + end. This probably wouldn't cause hum but it certainly would cause a decrease in gain. (7)Not that there is a 68K resistor on each jack which with 1 input will put them in parallel effectively creating a 34K grid stopper. RG Keen's schematic is erroneous.
As ISO noted, you can get all the parts you need right here from Hoffman. Quickest shipper in the amp parts biz. Another very important point he mentioned is the fact that it's super easy to accidentally pull up a trace on these old 70's PC boards.
I see no reason to change out that Stancor PT. If you do, send me your old one. :angel
-
This is excellent, I'm going to do exactly what you've specified, thanks very much! I'm going to order all at once, so it might be a week or two until you hear from me again, need to wait for my next check.
Yes, this is a vintage '70's unit, about 1973-74 I think. Serial number 712.
1.) If I read your post correctly, the only way the input could be the 68k as posted in the schematic would be if both inputs were used at the same time. when either is used singularly, the actual resistance is 34k? How does the second input reduce signal by 6dB as it says in the spec's?
2.) Right now, my caps are 40uf 450v, the schematics say use 47's... What is the difference / benefit of either? I have another picture of a re-cap job done supposedly by Alembic which shows 47's (below). The pic probably isn't real recent though, because as you see, the power cord was not switched out.
3.) Power Supply diodes will be replaced, at least one of them is a 4007, I looked today. I'll do the power supply resistors, and cathode bypass caps with metal films too. as you say they aren't in the audio loop, and they're also only a buck or so each, so why not. Can I just go through and replace the orange drop caps too while I'm in there? They're also a buck each, and will take a couple minutes each to do.
4.) I'm assuming that by leaving the carbon comp resistors alone, that will help retain MoJo? Can those drift also, or do they simply 'work or not.'
5.) So between the Sprague Atom, and the Ill. Power Supply, that Doug supplies, the Spragues are better? Is it true or possible that these caps can have effect on sound? In 'audiophile' circles, they all seem to like Rubycon, etc.
6.) Last Q., is a balanced out beneficial or even necessary? I'm driving 8 inches of cable into a balanced TRS 10Kohms input. At this point though I'm starting to think it would be cool to keep it as 'stock' as possible.
Picture is of a re-cap job on someone else's F2-B, supposedly done by Alembic.
-
gotta love the foam under the caps to keep them rattling... :grin: i'd have used goop - that way the next guy would cuss at me - a lot! j/k
a telefunken and a mullard... life is good. :glasses9:
if you like the way it sounded, and it was quiet enough for you once upon a time; why mess with it? the toriod (or any mod other than for safety) is probably not a wise idea for that vintage piece of equipment. just my 2 cents.
happy hacking! :smiley:
--pete
-
1.) If I read your post correctly, the only way the input could be the 68k as posted in the schematic would be if both inputs were used at the same time. when either is used singularly, the actual resistance is 34k? How does the second input reduce signal by 6dB as it says in the spec's?
2.) Right now, my caps are 40uf 450v, the schematics say use 47's... What is the difference / benefit of either? I have another picture of a re-cap job done supposedly by Alembic which shows 47's (below). The pic probably isn't real recent though, because as you see, the power cord was not switched out.
3.) Power Supply diodes will be replaced, at least one of them is a 4007, I looked today. I'll do the power supply resistors, and cathode bypass caps with metal films too. as you say they aren't in the audio loop, and they're also only a buck or so each, so why not. Can I just go through and replace the orange drop caps too while I'm in there? They're also a buck each, and will take a couple minutes each to do.
4.) I'm assuming that by leaving the carbon comp resistors alone, that will help retain MoJo? Can those drift also, or do they simply 'work or not.'
5.) So between the Sprague Atom, and the Ill. Power Supply, that Doug supplies, the Spragues are better? Is it true or possible that these caps can have effect on sound? In 'audiophile' circles, they all seem to like Rubycon, etc.
6.) Last Q., is a balanced out beneficial or even necessary? I'm driving 8 inches of cable into a balanced TRS 10Kohms input. At this point though I'm starting to think it would be cool to keep it as 'stock' as possible.
1.) Look at this schematic. Pay special attention to how the jacks are switched.
http://www.schematicheaven.com/fenderamps/twin_reverb_ab763_schem.pdf
2.) 40...47uF... it's all good. Back in the olden days, caps were expensive & bulky so they skimped. A little more filtration is a good thang with ss rectification.
3.) No reason to change the orange drops. The likelihood of them being bad is slim to none. Not worth risking damaging the PCB.
4.) Carbon comps can & will drift. That's why you either love them or hate them. I'm a CC fan myself. Leave em be.
5.) I've used the cheap Illinois caps w/o issue a zillion times. Sprague is a better cap. For this unit, I'd do Sprague just cuz.
6.) A balanced out would be beneficial if you intend to hit the snake at this point. In your case (jumping to the next box in the rack), it's a total waste of time & money.
-
I've used the cheap Illinois caps w/o issue a zillion times. Sprague is a better cap. For this unit, I'd do Sprague just cuz.
Please read what has been published here and elsewhere re. today's Sprague Atom caps: they are bog standard elcos fitted in big cans (hence lower reliability because of thermal problems) and resold for two or three times the price of better performing modern caps.
-
Butterylicious,
Awesome, thanks man! Before I submit my parts order to Doug, I'll Post to be sure I didn't forget anything.1.) Look at this schematic. Pay special attention to how the jacks are switched.
Not sure I understand the schematics, inputs, but I do see that the original uses two, rather than one 68k, I'm sure that has something to do with it...2.) 40...47uF... it's all good. Back in the olden days, caps were expensive & bulky so they skimped. A little more filtration is a good thang with ss rectification.
Okay, I'll go for the 47's. Maybe even 50's?6.) A balanced out would be beneficial if you intend to hit the snake at this point. In your case (jumping to the next box in the rack), it's a total waste of time & money.
Makes total sense RE: Balancing. I never have any intention of running this any more than a foot away to the next box. I'm just hoping to keep the signal as quiet and noise free as possible. I also want to be sure I'm feeding the 'optimal' level to the balanced 10K input. It seems though from what I've been reading that as long as it's over 600 ohms, it'll be fine.
Please read what has been published here and elsewhere re. today's Sprague Atom caps: they are bog standard elcos fitted in big cans (hence lower reliability because of thermal problems) and resold for two or three times the price of better performing modern caps.
FYL,
What would you use or recommend?? Some Hi-Fi audio dudes elsewhere speak highly of Rubycon, but I know Doug doesn't sell them. I have a high dollar 'boutique' headphone amp, all point to point using the 'best' components. It uses some big oil, or P.I.O. caps. It's been said those were the best 'sounding' ever. Would any of these (Rubycon, P.I.O., or cheapies) even make a difference, since none are in the audio loop?
gotta love the foam under the caps to keep them rattling... i'd have used goop - that way the next guy would cuss at me - a lot! j/k
ISoTone,
LOL, seems like someone already had that idea in mine, I'll be cleaning all that stuff out during my restoration...a telefunken and a mullard... life is good.
Not just any Telefunken, it's a '63 smooth plate. * Sounds of applause, fireworks, Ooohs and Ahhhs * The Mullard is a '61 short ribbed plate. I've recorded, level matched and compared. The difference is ever so slight, but it's there. The Tele is a little 'darker', the Mullard is a little bit more 'sparkly' I still have the original RCA's that came with it, which sound markedly better than a couple new Russians I tried several years back.if you like the way it sounded, and it was quiet enough for you once upon a time; why mess with it? the toriod (or any mod other than for safety) is probably not a wise idea for that vintage piece of equipment. just my 2 cents.
That's a good two cents. Based on how Butterylicious was salivating for the original Stancor that is in there, I think I'll keep it. I was also going to install an IEC AC jack, but I think I'll go with a grommet / strain relief, and a plain black cable, about a foot long. I think I will go ahead and keep the collector's value with it. Someday, I'll die, and it'll be my legacy, LOL
Thanks for all the help again!
-
What would you use or recommend??
Illinois, F&T, Nichicon, MIEC, CDE/Mallory, TCI and many others produce really good HV elcos. If you want 500V+, go for F&T or TCI; if you're looking for compact caps MIEC and Nichicon will fit the bill, if you're looking for big cans, CDE/Mallory has some really nice stuff, etc.
I have a high dollar 'boutique' headphone amp, all point to point using the 'best' components. It uses some big oil, or P.I.O. caps. It's been said those were the best 'sounding' ever.
Caps have no sound of their own. They can show defects when improperly used and thus impose their sonic signature, but no one - I repeat no one - has ever been able to find any subjective differences between two properly chosen and used capacitors in a double blind test.
Would any of these (Rubycon, P.I.O., or cheapies) even make a difference, since none are in the audio loop?
The power supply caps *are* in the signal path.
-
Illinois, F&T, Nichicon, MIEC, CDE/Mallory, TCI and many others produce really good HV elcos. If you want 500V+, go for F&T or TCI; if you're looking for compact caps MIEC and Nichicon will fit the bill, if you're looking for big cans, CDE/Mallory has some really nice stuff, etc.
I just need to keep them about the same size and shape, they need to fit just as they're shown in the pictures.
Caps have no sound of their own. They can show defects when improperly used and thus impose their sonic signature, but no one - I repeat no one - has ever been able to find any subjective differences between two properly chosen and used capacitors in a double blind test.
I think you mean OBjectively, right? SUBjectively is based on listener opinion... I'm mostly a fan of DBT's too, so point taken. I'll get some 500V 47 or 50mf. The Illinois one's are fine then, I take it...
The power supply caps *are* in the signal path.
But what you're saying is that it's irrelevant because the brand or type doesn't matter as long as the volts and farads are matched correctly, right?
.
-
The power supply caps *are* in the signal path.
But what you're saying is that it's irrelevant because the brand or type doesn't matter as long as the volts and farads are matched correctly, right?
As long as the caps are properly chosen - voltage ratings, capacitance, ESR, etc. - and used, yes.
We're looking at a small power supply for a couple of low-bandwith tube stages, basic Illinois caps will be fine.
-
Awesome! This project really isn't even going to cost much!!
-
Just a thought- all this talk of caps and it's not going to make a huge difference. Correct value Xicons will work just fine.
If it were mine and I were driving a compressor I'd add a mosfet source follower to lower the output impedance just a little bit. I'm sure it could be done in a minimally invasive way- mosfet gate (thru a small stopper) would connect where the big coupling cap connects, 47-100k to ground from the source and drain to B+...up the value of the coupling cap for full effect.
That setup will even drive a pretty low Z out, should you want to.
Don't assume that cleaner is better- sometimes the imperfections are ideal for a given instrument.
I once built a bass DI for a guy that I'm sure wasn't 100% clean but it sure did sound good. It was a single 12ax7 in a pedal operating as a gain stage into a direct coupled cathode follower connected to a DI box transformer through a large value film cap. A second unbalanced out was taken at the cathode follower with another film cap. I made the 1st stage cathode bypass cap switchable for a "gain change" if needed. I'm sure it wouldn't have held up well on hi-end test gear but it made his bass sound great.
jamie
-
If it were mine and I were driving a compressor I'd add a mosfet source follower to lower the output impedance just a little bit. I'm sure it could be done in a minimally invasive way- mosfet gate (thru a small stopper) would connect where the big coupling cap connects, 47-100k to ground from the source and drain to B+...up the value of the coupling cap for full effect... That setup will even drive a pretty low Z out, should you want to....
Uhh... Will that make it go gooder? I have NO idea what you just said, my knowledge is extremely limited beyond how to solder, and how to point out components. You'd have to draw me a picture, LOLDon't assume that cleaner is better- sometimes the imperfections are ideal for a given instrument.
Oh, not at all. If I wanted ultra clean I'd go solid state. I want that little bit of tube fuzz, and soft compression. - And the cool analog tone stack... But only that stuff, not hum and noise etc. introduced by worn components or incorrect usage.
-
Great project. Is it possible that you could post the B+ voltage and the AC voltage output of the transformer? Thanks
-
Hey guys,
I'm ready to make a purchase of my components, but I have some big gaps in my electronics knowledge.
I can identify components, and more or less how a circuit flows. I know how to test with a multi-meter, and I know how to solder but as far as which resistors, and capacitor sizing, schematics, etc. I'm clueless. I need to be sure I do this right, so could you guys help me make a "shopping list" here on Hoffman?
For the moment, let's assume that all the CC resistors in the channels will be fine, I can always go to the electronics store and pick up 1 or two that I need. I'd like to keep the mojo there if I can...
The big PS caps I already know are going to be 500v 50uF but the rest of the stuff I'm not sure how to select...
To replace the two silver looking electrolytic capacitors in each channel, I can see that they say Sprague 300V 50-160 TC-1160, but no idea what that means, or what to get. I'm assuming they are the electrolytic cathode bypass caps.
Then I want to replace all the carbon comp. resistors in the PS with new metal film resistors. I know which ones they are, but the voltages, watts, etc.?
-
(2x) 1.5K resistor (metal Film)
(1X) 270 resistor (metal Film)
(1x) 250 resistor (metal Film)
(1x) 68 resistor (metal Film)
Then I need 4 diodes (I think) for the PS as well.
(4x) 1N4007 Diode - again, which ones?
Any thing else I should add to the list? I have a power cord, extra wire, solder equipment, meter, etc.
-
Yes.. the differences in tone caps CAN be heard.. IF your playing solo.
-
Parts list:
4 - 47uf 450V Electrolytic Caps
2 - 50uf 25V Electrolytic Caps
2 - 1N4007 diodes
2 - 1k5 1/2 watt metal film resistors
1 - 270 ohm 1/2 watt metal film resistor
1 - 250 ohm 1 watt trimmer
The other diode & resistor only feed the LED. Not worth messing with.
For the big caps: 40, 47, or 50 uF is fine, 450 or 525V is also fine
For the little caps: 50uF at 25 to 50 volts is fine. Even 100 volts is fine but would be a pain to install due to their size. Doug sells a 50/50 Sprague, use that.
The plain ol resistors are 1/2 watt.
The trimmer is gonna be hard to find. It's only dissipating a little over 1/8th watt so a 1/2 watt would be fine. I wouldn't risk a 1/4 watt. I question if it needs to be replaced. With the tubes pulled, sweep it while watching it with an ohm meter. If it's not broke, don't fix it.
-
Thank you again Butterylicious!
Well, here's my parts order. I will get the trimmer locally IF I need it. I'd rather not de-solder it, etc. if it works.
Doug didn't have the 270 Ohm 1/2 Watt resistor, so I'll get that one here as well.
-------------------------------------------------------------------------------
Item ID........ Detail.... ...Qty ....Each ..Extend Description................
-------------------------------------------------------------------------------
1/2 watt metal. 1.5K...... 2 $0.20 $0.40 1/2 watt metal film resistor
1N4007......... .......... 3 $0.25 $0.75 1N4007 diode...............
47/500v........ .......... 4 $6.25 $25.00 47/500v....................
50/50v......... Sprague... 4 $2.15 $8.60 sprague atom cap...........
Grommet 1/2.... .......... 1 $0.15 $0.15 rubber grommet.............
Grommet 3/8.... .......... 1 $0.15 $0.15 rubber grommet.............
cable.......... .......... 3 $0.32 $0.96 mini shielded cable........
fuse large..... .5 amp.... 1 $0.80 $0.80 large Slo-blo fuse.........
strain......... .......... 1 $0.50 $0.50 power cord strain relief...
-------------------------------------------------------------------------------
His prices are reasonable, and NO TAX!
-
>
His prices are reasonable, and NO TAX!
And his service is exceptional. Warm your iron up for Wednesday evening.
-
as far as your unbal out.
weber sells the way cool tiny trannys https://taweber.powweb.com/store/balun.jpg
I've used it on a couple amps now and love it, I've started putting them in my speaker cabs as well, so any amp I plug into the speakers has a balnced out. I usually forego the R1 and put a 100K lin pot in there, it gives you a little more versitilty I feel.
only problem with these is they are so inexpensive, I have to wait till I make another order to get one, so I add 2 or 3 on each order I make from there.
IMHO I'd pass on the panel mount XLR they sell there, it's plastic, it works, but it wouldn't survive gigging, go buy a neutric or switchcraft metal one.
-
Thanks for the suggestion, but I'm going to stick with unbalanced out, and keep the collector's value intact. I'm never going to run more than an 18" cord anyway to the compressor right underneath it. If I send any farther, it would be from the compressor's output, which is balanced.
-
Got a package in the mail today, I should be able to start tomorrow. I'll post pics, etc. when I'm done!
-
What series Orange Drop caps were in there? 715, 716 or PS6? Thanks Should be printed on side of the caps next to value.
-
(http://www.geofex.com/FX_images/alembpre.gif)
FYI...
pointing out the error in V1b just so's everybody knows. Plate and cathode are reversed. RG must have had a brain fart when drawing up the schem. I built one and looking at that error had me scratching my head for a moment...
-
I completed everything so far with no problems, I'll post details and pictures soon.
I just ordered all the 'Orange Drops' from Doug as well, I'm still getting just a bit of hum from one channel, especially if I put my finger near / on one of the Orange Drops. BUT overall, it's much quieter, which is cool, and I think the punch and liveliness of the amp is back. I really should have made level matched recordings of both before and after, But subjectively, it has a lot more punch than before.
-
What series Orange Drops were in the preamp when you started? There are three different kinds. Thanks
-
What series Orange Drops were in the preamp when you started? There are three different kinds. Thanks
These are Sprague 6PS. The two cap sizes are P10's and P47's. Why do you ask?
-
Thank you very much I build tube guitar amps and know that there are distinct differences in sound between the three kinds of ODs. I was just wondering which ones were giving the F-2B its' MOJO! I akso know that in certain amps, The TW Express comes to mind, that a certain cap is almost mandatory. Thanks again for the reply.
-
Distinct differences? Such as? I was under the impression that the material they were made of was more to do with voltage and design. IE all of them rated for 500V and over are polystyrene. I thought also of swapping in some Mallory's as well.
At the end of the day though, I'm becoming less and less convinced that individual components, etc. do much to 'change' the sound...
-
Your opinion and you are entitled to it, but that does not make it fact. I am not going to get into a long drawn out debate. Ny health will not allow it. If you are interested you may do the research that I have done to find the answer. The data exists. Not onl choicr of componrnts but their physical location and wire dress also matters. Thanks again for the info.
-
Not trying to get into a pissin' match at all. I just wonder what tonal differences are said to exist? I didn't know you could use more than one kind of orange drop with that voltage requirement. Is there a site you can recommend I read?
-
Alembic F2-B refit
Sorry for the long delay in posting pictures.
I'm pretty well done, I've decided to forgo the balanced output, but the rest of the suggestions made here I have implemented.
The result is a unit which is much MUCH quieter - almost silent, even at extreme volumes / gain. A bit of thermal tube noise, and maybe just a tiny bit from the remaining CC resistors, but a huge improvement. No hum whatsoever. The active electronics in my bass are now far and away the noisiest part of my signal chain.
It has also regained the punch and clarity I felt it had been lacking.
Here's a rundown, and pictures.
New wiring as needed.
ALL Capacitors replaced.
All new metal film resistors and diodes in the power supply.
Channel CC resistors replaced with MF resistors where needed.
New grounded power cord, re-wired for proper grounding.
Added choke between X4mr and P.S.
Added small caps between tube output and tone stack.
Tinned circuit board, & riveted to chassis.
Cleaned or replaced all jacks, cleaned chassis, tube mounts, new screws as needed.
(http://farm3.static.flickr.com/2679/4200406999_482c40fb60_o.jpg)
(http://farm3.static.flickr.com/2680/4201159972_6336babb8e_o.jpg)
-
Here's another picture:
(http://farm3.static.flickr.com/2541/4201160082_df9d2083ba_o.jpg)
-
I haven't been around much lately but I wanted to say congrats and nice work. It looks really good. I bet you're glad to have it back working all proper and stuff. I love it when people pop in here and fix thangs themselves. DIY baby.
-
Thanks! Indeed it is VERY satisfying. I liked what I learned as well.
Hey, check this out -
I'm thinking of making another pre-amp, really more of a direct box, based on this circuit. More of a simplified reverse engineering job than anything...
What I'd like to do is make the same power supply, and then use a 12AX7 - and nothing else - as the preamp. No tone stack, extra caps, etc.
I'd just run the input jack straight to the first stage, and connect the pins to go directly to the second stage, and then straight to the output jack.
The only extra might be a first and /or second stage gain control. The tone would then be entirely based on the tube, (I think) and the bass plugged into it.
I'm a complete novice, but I think it will work (?) My curiosity as a 'minimalist' has had me thinking about this for some time now.
It seems my most satisfying 'mods' lately have been in streamlining how much circuitry the tone signal has to go through, and I'm really liking the results. I disconnected the onboard E.Q., and bass/treb controls on one of my basses recently, and found I heard more punch and clarity.
What do you think? Any suggestions?
-
Less is more... Bauhaus
No doubt bypassing tone controls on a guitar/bass guitar will add punch. Any tone control (except for an active) will have some dB loss and obviously some frequency response attenuation. Get rid of it, no losses = more punch.
Attached is an analysis of the AB763 preamp circuit which the Alembic is directly copied from. The 2 gain stages are effectively the same. (In actuality the 1st stage has about double the load {RL} on it but that's only <1dB difference so lets just count on thumbs and call them both 35dB, AV 50. The EQ has a huge mid scoop so lets just round this stage to -15dB or a AV of 0.2
35-15+35=55dB or 50*.2*50=500 AV of 500 is 54dB... close enough
A gain of 500 is ample... excessive even. Now take out the negative factor in the equation above and you get AV=2500 & 70dB You gotta have a volume pot and you prolly won't be able to turn it past 1 w/o overdrive.
Often less is still less.
-
> Attached is an analysis of the AB763 preamp
No actual attachment.... you mean your inline text analysis?
> The only extra might be a ....gain control
Agree with Butter's analysis. Gain will be in the 2,000+ range. This is probably TOO much for some situations. Output of 0.1V is common. Times 2,000, is 200V. Which is a little more than a 2V-in power amplifier wants to find on its doorstep.
But of course what really happens is the second stage clips at "only" 100V peak, dirts-up your sound, and is still too hot for the power amp.
You can turn-down between stages, indeed it is handy to have a knob right there near your plug-jack (albeit electrically between stages). But two cascaded 12AX7, and no other loss, you can never use the upper half of the pot. It's just too much.
It would also be nice to have a somewhat lower output impedance, in case you ever use a longer cord from preamp to power amp.
Which all suggests a different tube-type than 12AX7. But we won't mess with success.
Throw-away some gain and excess level at the second plate resistor. The 12AX7 still "feels" the 100K load we know works so well. The output gets 1/10th (or 1/5th) of the 12AX7's full swing. The output impedance is 9K (or 17K), which will drive 30 feet of cable to the top of the audio band.
Pick the first or second tap down in testing. (And if I'm full of snow again, you still got the 102K tap at the plate.)
-
DoH!!!!! OK it's attached now. :angel
Question for oatmeal:
When you use the Alembic, where do you usually keep the level set?
The reason I ask is because I have breadboarded this (Fender) circuit more than once and have even tried as a stand alone preamp where I found it to have waaay too much gain.
Here is a schematic for a bass preamp I designed/plagiarized about 6 months ago.
http://www.sotxampco.com/Schematics/STAC/Chanel-B-8.pdf
-
A gain of 500 is ample... excessive even. Now take out the negative factor in the equation above and you get AV=2500 & 70dB You gotta have a volume pot and you prolly won't be able to turn it past 1 w/o overdrive.
So, maybe in keeping with the original design, maybe only use one gain control for the first stage and don't attenuate the second?Which all suggests a different tube-type than 12AX7. But we won't mess with success.
Brainstorm! - I have thought of substituting a 6SL7 into the Alembic circuit before. I read somewhere that this was the original impetus for a 12AX7. There is some talk about it being a nice sound too, although more likely to be microphonic. Would this knock the gain down enough?
Another option might be to do it with a 12AU-7 or something with less mu ?Throw-away some gain and excess level at the second plate resistor. The 12AX7 still "feels" the 100K load we know works so well. The output gets 1/10th (or 1/5th) of the 12AX7's full swing. The output impedance is 9K (or 17K), which will drive 30 feet of cable to the top of the audio band.
If I read this right, you're saying it becomes 'Low-Z' out, without even having to use a transformer - I.E., the perfect direct box!When you use the Alembic, where do you usually keep the level set?
The reason I ask is because I have breadboarded this (Fender) circuit more than once and have even tried as a stand alone preamp where I found it to have waaay too much gain.
Gain is at 4 to 5 (out of 10) I have used it as low as 2 or 3, which makes it VERY clean. Beyond 5 or 6 and it drives whatever I have downstream (compressor or power amp) too hard.Here is a schematic for a bass preamp I designed/plagiarized about 6 months ago.
plagiarism is the sincerest form of flattery!
-
There's Low-Z and then there's balanced Low-Z. Most direct boxes are balanced.
-
>There's Low-Z and then there's balanced Low-Z. Most direct boxes are balanced.
>If I read this right, you're saying it becomes 'Low-Z' out, without even having to use a transformer - I.E., the perfect direct box!
There's more than a couple of ways to do this. If you want to avoid costly and colorful transformers, one could wire 1/2 of a 12AU7 as a split load phase inverter and get your balanced out that way. The other 1/2 could be a cathode follower for your unbalanced output. Personally I happen to like colorful little transformers.
>Brainstorm! - I have thought of substituting a 6SL7 into the Alembic circuit before.
Go for it. The 6SL7 is a schweet sounding tube.
>Gain is at 4 to 5 (out of 10) I have used it as low as 2 or 3, which makes it VERY clean. Beyond 5 or 6 and it drives whatever I have downstream (compressor or power amp) too hard.
This vallidates what PRR said about 2 - 1/2's of a 12AX7 cascaded together.Let's grab my numbers from before:
V1 * Tone Stack * V2 = VGain 50 * .2 * 50 = 500 or 54dB
Now lets insert a volume pot set at 50%.
v1*tone loss*vol pot loss*V2 = 50*.2*.5*50=250 or 48dB
Let's call this Maximum usable gain. I think it's still on the high side. Your milage may vary.
If we set the volume pot at 10% we get
50*.2*.1*50=50 or 34dB which is really a good number to shoot for.
>What I'd like to do is make the same power supply, and then use a 12AX7 - and nothing else - as the preamp. No tone stack, extra caps, etc.
There is a very good way to do this and limit gain and decrease distortion. Instead of wiring the 12AX7 as a grounded cathode, wire it as a plate follower. This is what I did on my preamp. The major difference between a grounded cathode and a plate follower is negative feedback. Yes you're adding a couple of resistors and a capacitor, but the payoff is much less distortion and much improved gain control. I realize my schematic is hard to read. The software I drew it in is "vintage" and it needed to be drawn complete since it's built onto a PCB. I can redraw it in SCH, broken down into it's fundamental stages if you'd like (later when I have some free time). I'll go ahead and explain it just for drill:
The 1st stage (V1b) is a plate follower. Gain is set by the relationship between R4/R1 = 1M/100K=10 or 20dB
Next is the volume control. It's gain varies between 0 & 1. Let's assume it's at 50% so call it .5 or -6dB
Just in front of the volume control is a low cut switch RC network. Ignore that for now.
The 2nd stage (V1a) is a grounded cathode. Let's just call it's gain 50 or 34dB
Next is a James Tone Stack with a loss of 28dB (Vg=.04)
This is the basic preamp section.
Let's total this up:
Gain= 10*.5*34*.04=10
dB=20-6+34-28=20
20dB is a gain of 10
With the volume dimed, it's a gain of 20 or 26dB
Off the output of the tone stack, the signal splits to the 2 1/2's of the 12AU7.
V2b is a cathode follower. It has a gain of less than 1, like .93 but for now just call it 1. This feeds a 1/4" jack and is my unbalanced output with an impedance of about 450 ohms.
V2a is another plate follower with a gain of about 10. This feeds an output transformer with a gain loss of about .1 V2a's gain was set to compensate for the loss of the output transformer, effectively = 1. The secondary of the OPT hits a polarity switch and a ground lift switch, both quite necessary for a balanced output.
The power supply quite stiff (300uF). It takes a while to charge which 1/2 the reason I put a 555 timer in. The other 1/2 is to allow the filaments to heat up. The timer is simply a 60 second delay keeping the outputs disconnected until the circuit is warmed up. Since I already had 12VDC, it was obviously prudent to use this for filaments.
OK.... so you prolly don't need all the fluff and you said you want to take a minimalist approach using a 6SL7. I'd suggest trying a bypassed plate follower shooting for a gain of about 30. Insert a volume pot and use the other 1/2 as either a cathode follower if you want an unbalanced out or as a split load if you want a balanced output... although you might not get a low enough Z out with a 6SL7.
-
Attached (hopefully) are copies of my circuit w/o the relays or power supply and other misc fluff. I tried to draw it clearly to help you conceptualize a circuit that suits your taste.
-Richard
-
you could use a 6SN7 for the CF stages and 6SL7 for gain w/ GC & PF.
sub 5691 for 6SL7 and/or 5692 for 6SN7...
--ISO
-
> 6SL7 ... I read somewhere that this was the original impetus for a 12AX7
12AX7 is a dual 6F5 (http://www.mif.pg.gda.pl/homepages/frank/sheets/049/6/6F5.pdf).
The 12AU7/6SN7 is a very good clean tube.
6SL7 is juiced-up for more gain and less linearity.
12AX7 sits on the edge of instability to get the maximum gain which can be mass-produced and used for audio.
Tube-swaps are quite valid. But recall that "everything" uses 12AX7. I suspect the ear wants the non-linearity. That if we truly wanted "clean", today, we'd use a good chip. If you gonna fool with tubes, your best bet is to stay with the pack.
> you're saying it becomes 'Low-Z' out
NOT the 200-600 ohms that Pro Audio expects.
OTOH, not the 50+K that you get out of a pickup or the plate of a high-gain tube.
It's a good unbalanced line which can be run across a stage. Not over the lighting racks to the far end of the barn. (For tough rooms, add a high-quality 10K:10K balancing transformer; but real-tough rooms want a proper Line Driver, not a gimmicked 12AX7.)
And it also throws-away roughly the excess gain of two stages of 12AX7, with few side-effects.
> 34dB which is really a good number to shoot for.
20mV to 1V is gain of 50 which is 34dB. This alone would be enough. But we need a volume control, and golly we just gotta have tone controls. So you need two stages. Conceptually you could manage with 1/2 12AX7 and 1/2 12AU7. There is actually a dual-triode like that. But you can't get a replacement in Lodi on saturday night. So you use the universal 12AX7.
Cutting-out the tone control really leaves too much gain. Gain is easy to throw-away, just a question of how best to do it.
The James stack is interesting, because (unlike the Fender) it has a nearly-flat no-effect setting. (BTW: gain of 10K:600 is -12dB not -20dB.)
If you could find a good transformer 60K:600 which would accept 2mA of DC.... but today such a thing costs more than all the other parts in an Alembic. It isn't a standard part even at the boutique winders, AFAIK. You can find similar iron but response will be poor.
-
" ... There is a very good way to do this and limit gain and decrease distortion. Instead of wiring the 12AX7 as a grounded cathode, wire it as a plate follower. This is what I did on my preamp. The major difference between a grounded cathode and a plate follower is negative feedback. Yes you're adding a couple of resistors and a capacitor, but the payoff is much less distortion and much improved gain control. I realize my schematic is hard to read. The software I drew it in is "vintage" and it needed to be drawn complete since it's built onto a PCB. I can redraw it in SCH, broken down into it's fundamental stages if you'd like ... "
" ... Let's total this up: Gain= 10*.5*34*.04=10 dB=20-6+34-28=20 20dB is a gain of 10 With the volume dimed, it's a gain of 20 or 26dB Off the output of the tone stack, the signal splits to the 2 1/2's of the 12AU7. V2b is a cathode follower. It has a gain of less than 1, like .93 but for now just call it 1. This feeds a 1/4" jack and is my unbalanced output with an impedance of about 450 ohms. ... "
Uhhh.... Guys I love all the help I'm getting here, and I am learning bit by bit, but with this you may as well be showing trigonometry to a sea slug.
So, essentially it isn't as easy as making a power supply, connecting it to a tube, along with an input, a couple gain knobs, and an output... ?
I have this 'thing' I acquired on eBay. It's a jumble of wires, with a 12AX7 attached, let me see... It has an outboard (wall wart) transformer at 15 watts. This is connected to a 3300uF 25V cap and then the tube.
There are two gain knobs, connected with some resistors (one is pretty large - a 10w 150ohm 'brick') and a small cap or two to the tube, along with a 1/4" in plug and 1/4" out for my bass. I use it as a low level pre to my headphone amplifier, and it seems to work fine for practice, and even recording.
My ambition is just to make one with a hotter output and zero tone modification other than the tube. And enough to drive a compressor or power amp, because this one doesn't have the balls.
I think what I'm reading here is that isn't really a possibility, as simple as I have envisioned it, Is that right? I have a pile of 6SL7's laying around, and it would be great to use those instead, but again it seems like it's not really feasible?
In other words, plain English, suitable for dumb guys please gentlemen. I'm learning little by little, but so far this is like trying to get a drink from a fire-hose!
-
Now I'm confused.... Do you want a stereo preamp to drive a headphone amp and or power amp, or do you want a bass preamp? You could build a fine (either one) using 6SL7's.
-
:grin: I see how that might be confusing now that I read it...
The 'jumble of wires thing' I use with my headphones is so that I can practice without bothering my neighbors. I was just using it as an example. It doesn't have the cajones to really drive a power amp and get a good thick sound, it's just enough to add a little gain going into my headphone amp for practice.
What I want is a single channel bass preamp. I was just thinking that if my 'jumble of wires thing' works for a little bit of gain with just a couple resistors and a wall wart, why not make a 'real' power supply like the one in the Alembic, and drive the tube with that - except no tone stack or anything. Just a gain/volume knob on each side of the tube, to adjust gain structure from clean to dirty sounding.
You guys were saying there needed to be a tone stack (I think) because the tube needed it as a pad right? Why not just put a resistor there?
-
Here ya go. I whipped you up a little 1 bottle preamp. Filaments are 12.6 cuz that was the only suitable transformer I could find. I did find one with 6.3V but the HT was about 40 times greater than necessary.
-
Buttery -
This is awesome, thanks! :occasion14: Even I can figure how to wire this one!
Questions:
Does Hoffman Hoffman sell a P-T442 transformer? I didn't see one. Looks like they're easy to get elsewhere though.
So, S2 is a boost switch - for gain I'm assuming, or will it boost treble as on the Alembic?
Would I be able to place a 'post gain knob' somewhere between R8 and R9?
Can I substitute a 6SL-7 for the 12AX-7 if I give the PT secondary a 'dummy load' - maybe another 6SL-7, or a resistor in series, etc?
If I could find a transformer (like another stancor with 6.3 V on the secondary) Could that enable a 6SL-7 to be substituted in this circuit?
Can you recommend a 'tube pre-amps' for dummies book, so I might understand some of what you wrote about earlier? Power amplification is never really an issue - I'm a solid state guy there... But small signals, and Pre-amps are what I'm mostly interested in.
Thanks again for doing this!
-
>Does Hoffman Hoffman sell a P-T442 transformer?
No, Antique Electronics has it. (Sorry Doug)
>So, S2 is a boost switch - for gain I'm assuming, or will it boost treble as on the Alembic?
It boosts gain. The cap you select will affect lo freq cut off. I think Fender used a 15uF, Marshall used a .68uF for a lot of lo cut. A 25 should pass full frequency. It's something to play with. Less than 2V present, clip leads are your friend.
>Would I be able to place a 'post gain knob' somewhere between R8 and R9?
Make R9 a pot. Tie the wiper to J2. Not much reason to. I don't think you can over-drive V2 because it's a cathode follower whose input threshold is still 8V even with a 10k load. Waste of a pot IMO.
>Can I substitute a 6SL-7 for the 12AX-7
Circuit-wise yes. Filaments are the only issue. I found an 80mA transformer, you need less than 5mA. If you're gonna series 2 6SL7 filaments, go ahead and make a stereo unit. Use switching input jacks and wire so when just the left jack is plugged in, it drives both preamps. :wink: Make R10 a 33k if you build stereo. You may have to play with this value to get roughly 250V. I'm guestimating 1mA per stage. You wanna drop 100V across R10. That's high voltage, clip leads are NOT your friend.
>Can you recommend a 'tube pre-amps' for dummies book
http://www.angelfire.com/electronic/funwithtubes/
-
It boosts gain. The cap you select will affect lo freq cut off. I think Fender used a 15uF, Marshall used a .68uF for a lot of lo cut. A 25 should pass full frequency. It's something to play with. Less than 2V present, clip leads are your friend.
I see what you are saying about a gain pot on R9. Can I put one on R4 and delete the the gain switch and cap?Filaments are the only issue. I found an 80mA transformer, you need less than 5mA.
Wow, that much difference? I guess it isn't as easy as just using 'half as much.' If you're gonna series 2 6SL7 filaments, go ahead and make a stereo unit. Use switching input jacks and wire so when just the left jack is plugged in, it drives both preamps.
Say, this is a great idea! Use the PT you listed and just drive two filaments... Would I just use the second jack to drive the second channel / tube only? This might be a great way to compare tubes too!
-
R4 sets the bias of V1. You can vary it between something like 820 ohms and 3k3, but 1k5 is optimal. Adding the bypass cap (C1) gives you about 6dB of additional gain outta that stage. 25uF should pass full audio bandwidth and you should start hearing a decrease in lo freq at about 15uf. As you decrease this value, it will get tinnier. Play with the cap value, not the resistor. Most likely for bass, you'll wanna run with a 25uF switched in. If you put a string of stomp boxes in front of it, you'll prolly want it switched out.
You could use a 500mA transformer if you wanted to. you could put sugar in your coffee with a backhoe too. I speced an economical little 15mA transformer. More than enough HT to drive 2 bottles, just enough filament power to drive 2 12AX7's or 2 6SL7's.
-
You could use a 500mA transformer if you wanted to. you could put sugar in your coffee with a backhoe too. I speced an economical little 15mA transformer. More than enough HT to drive 2 bottles, just enough filament power to drive 2 12AX7's or 2 6SL7's.
Ohhh... Your statement 'Filaments are the only issue' threw me.
In other words, I can put whichever, a 12AX-7 or a 6SL-7 in there with no problems...
-
I can see how I was confusing. I was talking about the HT secondary and the filament secondary at the same time. Let's do filaments first.
A 12AX7 can be heated with either 6.3V or 12.6V. For 6V operation, tie 4 and 5 together and use pin 9. For 12V operation use pins 4 & 5 and ignore pin 9. There's 2 internal filaments in each bottle wired to a common point (pin 9). If you wire them in parallel, then your source voltage is 6.3VAC @.3A Wired in series you need 12.6VAC @.15A
A 6SL7 only has 1 filament winding, 6.3V @ .3A If we have 2 - 6SL7's, we can wire their filaments in series and feed from a 12.6V source. Same deal as what's internal to a 12AX7 except your using 2 tubes. For 2 tubes you need either .6A @ 6.3V or .3A @ 12.6V
Where I threw you was talking about the ampacity need for the HT. I counted on thumbs and called it 1mA per stage, so you need 2mA for mono and 4 mA for stereo at 250V. So armed with these values, I started looking for tiny transformers. The smallest transformer I could find that had 6.3V filaments had 80mA of HT. Way more than you need. The transformer I pick has .3A of 12.6V, exactly enough for 2 of these bottles. The HT secondary is 125V @ 15mA. Wired into a voltage doubler rectifier, you get 350ish volts @ 7.5mA This transformer is perfect for little 1 and 2 bottle preamps.
-
A 12AX7 can be heated with either 6.3V or 12.6V. For 6V operation, tie 4 and 5 together and use pin 9. For 12V operation use pins 4 & 5 and ignore pin 9. There's 2 internal filaments in each bottle wired to a common point (pin 9). If you wire them in parallel, then your source voltage is 6.3VAC @.3A Wired in series you need 12.6VAC @.15A A 6SL7 only has 1 filament winding, 6.3V @ .3A If we have 2 - 6SL7's, we can wire their filaments in series and feed from a 12.6V source. Same deal as what's internal to a 12AX7 except your using 2 tubes. For 2 tubes you need either .6A @ 6.3V or .3A @ 12.6V
Ahh So. Now I completely understand, thanks. So yeah, if I'm going to do 6SL7's, it'd almost be dumb not to do it in stereo. I also understand now why you can't just swap one for the other, without considering how the filament voltage will end up.
Now, I went back and re-read your post on the last page about 28 times...."... There is a very good way to do this and limit gain and decrease distortion. Instead of wiring the 12AX7 as a grounded cathode, wire it as a plate follower. This is what I did on my preamp. The major difference between a grounded cathode and a plate follower is negative feedback. Yes you're adding a couple of resistors and a capacitor, but the payoff is much less distortion and much improved gain control. ..."
OK.... so you prolly don't need all the fluff and you said you want to take a minimalist approach using a 6SL7. I'd suggest trying a bypassed plate follower shooting for a gain of about 30. Insert a volume pot and use the other 1/2 as either a cathode follower if you want an unbalanced out or as a split load if you want a balanced output... although you might not get a low enough Z out with a 6SL7.
It seems to me that I would want the benefits of some negative feedback if I can get it so easily. It isn't like it is 'changing' the tone by adding tone circuits, etc. It's just using the same circuit in a better, more controlled way... Is it a simple thing to make some changes in your 'oatmeal' design to do that? I don't understand what a 'cathode follower' or a 'grounded cathode' (although the latter seems self explanatory...) are, but it seems an easy option for the first half of the tube. Then to have a 'split load' on the second half of the tube, to make a relatively decent balanced out. - I won't be vaulting any light rigs over barns...
I will try to do some homework and read about the above mentioned terms as well...Go for it. The 6SL7 is a schweet sounding tube.
Hooowww schweeeett iiisss iiittt??? In your subjective opinion of course as a fellow bass player. - Is it 'better' or just 'different' than the 12AX-7?
-
If you just want a headphone amp why don't you plug the alembic into a solid state headphone amp and put a pot inbetween so you can overdrive the second gain stage in the alembic if needed?
maybe I'm not seeing the forest through the trees? :embarrassed:
Personally I prefer a little dc biased cathode follower sound (tweed bassman, plexi, SVT) for bass...but that's just me.
jamie
-
Assuming you want to run within the limits of this transformer, you are limited to 2 - 6SL7's. That gives you 4 stages to play with. I drew a gain stage followed by a driver stage. You could double that and have a stereo unit. You could change the driver stage from a cathode follower to a split load and have balanced outputs. You mentioned neg feedback, but with a single gain stage I don't think you'll have enough output left.
I don't know what you have against tone stacks. The James stack hijacked from the SVT is awesome for bass. With 4 stages, I'd build (actually built):
Gain stage with a lit of neg feedback > volume control > Gain stage > Tone Stack >split the signal to a cathode follower for a hi Z out and to a split load for a balanced out.
Heck... 2 bottles, copy the Alembic circuit but use 6SL7's.
-
If you just want a headphone amp why don't you plug the alembic into a solid state headphone amp and put a pot inbetween so you can overdrive the second gain stage in the alembic if needed?
My original intent was to build a pre-amp, like the Alembic F2-B, but without the time stack. The reason is because I have noticed over time that the simpler I can make my signal path, the more 'punch' and 'truth' I can get out of my basses.
Then I had the bright idea of simply eliminating anything extraneous to pre-amping at all, and simply let the tube be the sound.
Butterylicious has been more than awesome in helping me understand all this. He drew me a schematic for what I wanted which I will build as soon as I can.
Now, after reading his latest post, I'm thinking of trying the 'James' stack, along with his design as well.
-
You could change the driver stage from a cathode follower to a split load and have balanced outputs.
That sounds perfect, what do I do differently to your 'oatmeal' design to make this so?
I don't know what you have against tone stacks. The James stack hijacked from the SVT is awesome for bass. With 4 stages, I'd build (actually built):
Gain stage with a lit of neg feedback > volume control > Gain stage > Tone Stack >split the signal to a cathode follower for a hi Z out and to a split load for a balanced out.
This is the schem. you sent me before, right? I once had an SVT pre-amp, but remember not liking it's sound as much as the Alembic. But you're right, it does seem like it's worth a shot. Maybe my taste has changed in 20 years.
Would you mind if I built your pre-amp? Can you share with me the complete schem, with the power supply, etc.?
Heck... 2 bottles, copy the Alembic circuit but use 6SL7's.
Based on the last few posts on X4MR's, I think I could 'try' this sound by swapping a 6SL-7 into one of the channels, and leave the other channel tubeless - thus preserving 6 volts. If I liked it, I could then build it.
-
I've attached a stereo 6SL7 version of the Oatmeal Preamp with balanced outputs. I was quite hammered last night so consider this a DWI (designing while intoxicated) drawing. I have not checked it for errors. I'm home sick today so I'll check it out later on. Lemme breadboard it before you go out spending a bunch of money. Looks OK on paper, it might suck in 3D.
>Would you mind if I built your pre-amp? Can you share with me the complete schem, with the power supply, etc.?
Go for it. Here's the schematic:
http://www.sotxampco.com/Schematics/STAC/Chanel-B-8.pdf
In fact, I have 1 blank PC board left. It's rev1 and has a couple of tight fitting holes but you can have it at cost if you want it. It's kinda pricey to finish with chassis -n- all + the peerless transformers are a fleaBay battle.
>Hooowww schweeeett iiisss iiittt??? In your subjective opinion of course as a fellow bass player. - Is it 'better' or just 'different' than the 12AX-7?
The 6SL7 is one of the last octal dual triodes. It's curves look very much like that of the later 12AX7. Opinions are subjective, but it's a good audio tube. Bass player???
-
> a P-T442 transformer?
With 6V tubes, the power supply from a Champ would work. Just do NOT wire a Champ PT the way Sluckey shows for the P-T442! Wire like a Champ, altho two 1N4007 is neater than a 5V3 rect.
BTW: 12SL7 (no dash!) is same-as 6SL7 except 12V heater.
> I found an 80mA transformer, you need less than 5mA.
Same deal. Champ PT can deliver 40mA-100mA, so is "too big for this job". However sometimes you only need a coffee spoon but can only find the dogfood scoop. At least it's not a backhoe.
There's also this for 12V tubes: PA211 Upgrade and Replacement transformer for Dynaco PAS2, PAS3 and PAS3X preamplifiers (http://store.triodestore.com/pa2powtrandy.html) 330-0-330VAC at 15 ma, plus 12V at 800ma ---- $35
> You wanna drop 100V across R10.
Or if playing with different PT: you want to be in the 250V area at point "A". Build the PT rect and first cap(s), measure what you got there. Slucky's PT will give about 350V, the Dyna part will give 450V, the Champ PT about the same (depends who wound it). So the higher-volt PTs may need 100K-150K at "R10". A dozen 15K in series is a handy tool.
> a split load and have balanced outputs
Well..... good enough to waggle two grids on the same chassis. But interconnects can be long and have substantial capacitance. The split-load's balance is a matter of debate, and IMHO not a good way to drive a "balanced input jack". You want a lower impedance and less coupling (load on cathode radically affects gain at plate).
-
>> a P-T442 transformer?
I picked this transformer cuz it's $15.00 cheap, just barely big enough, and a 125V HT is a little safer for a 1st time project. The filaments are simple enough to work around. Nothing wrong with a voltage doubler.
>The split-load's balance is a matter of debate, and IMHO not a good way to drive a "balanced input jack".
I agree. I like transformers but transformers are expensive. I said I was gonna bread board the circuit, but it dawned upon me I don't have a power amp with a balanced input. Attached is what I had originally drawn but with cathode followers instead of split loads. I will breadboard 1 side of this circuit.
-
Bread board experiment is complete. Too much gain. Changed to a plate follower and all was good.
See attachments.
-
http://www.edcorusa.com/Products/ShowProduct.aspx?ID=631
transformer in link above will get you right in the 250VDC ballpark without burning off excess.
costs 46 and change american.
it's rated @ 200ma... so ya, kind of like sending a battleship to take out a dinghy, but what the heck...
--ISO
-
100V @ 4mA is a whopping 0.4 watts. What's the big deal? I say run with the $15.00 transformer. You'd be hard pressed to ying-yang a pair of wall warts for that price.
xfmr - $15
jacks - $10
Resistors & caps - $10.00
Sockets - $5
box - $20.00
misc - $10
It's like a $70.00 project + tubes (which he already has)
A $70.00 stereo tube preamp, how can you go wrong?
Build the Oatmeal 1.1
-
Bread board experiment is complete. Too much gain. Changed to a plate follower and all was good.
I simply cannot WAIT to build this!!
As soon as I get $50 from my next gig...!
It's like a $70.00 project + tubes (which he already has) A $70.00 stereo tube preamp, how can you go wrong? Build the Oatmeal 1.1
I even have the 47u / 450's from the Alembic rebuild I did. I know at least 2 of them tested out just fine. I replaced just to be consistent. I have the resistors and jacks too. I cannot WAIT to build this!
I think I will kind of breadbox it too, until I figure out what kind of chassis I want. It's so easy, I might just go point to point with no board or anything. I'm kinda leaning toward a single space rack-mount. One knob, one jack per channel. Mysterious and oh so sexy! Anyone know a good place to buy a chassis?
-
Thank you, Butterylicious. This thread got me through some stuck spots on the preamp I'm building. I've read 3 or 4 really good preamp threads from you. Huge help. Much appreciatated!
-
>Thank you, Butterylicious.
You're most welcome. I'm glad I can help especially since I'm still learning.
>>Too much gain.
I played around with the circuit some more. "Too much gain" for my crappy ss amp and cheap speakers. Your mileage may vary. If you don't have enough gain, revert to my earlier schematic. Remove Rfb and change R2 back to a 27k. W/o Rfb, the 47k will cut into the hi frequency response. With a 27k, you should clear 50kHz.
My estimation of 1mA per stage is pretty much right on. V1a is slightly less than 1mA and V1b is slightly more. It all works out in the wash.
>Anyone know a good place to buy a chassis?
You mean rack chassis right?
http://www.par-metal.com/
So much for $70.00 :rolleyes:
>I think I will kind of breadbox it too, until I figure out what kind of chassis I want.
Perf board and an old cake pan. Works great but pisses off the Mrs.
"MY MOTHER GAVE ME THAT CAKE PAN!"
-
I have a question on voltage doublers pertaining to this inexpensive transformer I've been talking about. I am playing with an embellishment of the circuit above using a 12AT7 and a 12AU7 since I literally have a gallon of AT's and closer to 2 gallons of AU's. My plan requires approximately 8mA of B+. This transformer is rated at 15mA. If I understand correctly, a voltage doubler can deliver 1/2 the mA available from the PT so that gives me 7.5mA I'm already at the max rated current for filaments. Do you think I'm begging for trouble?
Thanks!
-Richard
-
Judging by the picture on the AES site the core seems to be generously sized, so I guess that you can try to max it.
-
That's what I'm hoping for. It seems like most of the good transformer builders spec their products conservatively. For $15.00, it's not like I'd be out all that much if it can't withstand my torture.
-
Do you think I'm begging for trouble?
yes.
weather it'll work or not is not the issue, of course it'll work... what is the issues are long term reliability, core saturation and long term temperature rise... play for a couple of hours and you probably won't be able to touch it... forget to turn it off for a day or two and you just may have a melt down. remember this: any piece of electronic equipment is a good as it's power supply, no matter how good it sounds...
hammond winds that same series of transformer with larger cores, it's just that AES doesn't stock them - find out who does and take the high road. :smiley:
-
Hello!!!! That is a Hammond.... or at least it was on a Hammond page when I found it.
-
I played around with the circuit some more. "Too much gain" for my crappy ss amp and cheap speakers. Your mileage may vary. If you don't have enough gain, revert to my earlier schematic.
I'm active in all my instrument electronics, so we'll see. Even so, adjusting doesn't seem like a big deal.
>Anyone know a good place to buy a chassis?
You mean rack chassis right?
http://www.par-metal.com/
So much for $70.00 :rolleyes:
Holy smokes, screw that! I'll make one out of wood first...
"MY MOTHER GAVE ME THAT CAKE PAN!"
"Well, if you'd have put it to better use than burning up some old cake mix, you'd still have it! ..." :shocked:
I'm still single BTW, wonder why...
-
I'm getting ready to order and make 'The Oatmeal.' I've been messing around with my current setup as well. Something I've realized I just don't understand is input/output impedance. What it is, what I need, how to measure, etc.
I kinda thought it had something to do with resistance, and how loud or hot the signal would be, but I guess there's much more to it.
What will the output impedance be with 'The Oatmeal' and will it drive today's modern +4 dB equipment?
There's another question - can anyone explain or show me where to read - in dumb guy terms - what all this +4dBu -10dbV stuff is, and how it all relates??
Maybe I should not worry about it, build this thing, and plug it into my compressor to see if it sounds good...
-
>There's another question - can anyone explain or show me where to read - in dumb guy terms - what all this +4dBu -10dbV stuff is, and how it all relates??
Here's a very good link. It has dB conversion calculators and clear definitions in the box below the calc.
http://www.sengpielaudio.com/calculator-db-volt.htm
>I just don't understand is input/output impedance.
Impedance is measured in ohms therefore it is resistive, but the terms implies that it changes. Consider an 8 ohm speaker. It's actually a coil of wire floating in a magnetic field and not a resistor, but sitting there doing nothing, it measures roughly 8 ohms. As soon as you start pushing signal through it, the resistive component is all over the place. I think of impedance as the resistance in quiescent state.
Input impedance is the reflective resistance, or load you'd put on a previous stage. Ideally you want this to be very high so as to not tax the previous stage. Consider the input of this circuit. We have a 1 meg to ground in parallel with the grid of a 6SL7. I don't remember the grid impedance of a 6SL7 but it's much greater than 1 meg, so effectively we've set the input impedance at 1 meg. 1 meg is a very light load.
Output impedance is well... the output resistance. It's comprised of amplifier's impedance in parallel with a fixed load. Unlike input impedance, output impedance isn't as easy to set. It's fairly dominated by the tube and how it is wired. Plugging the values into TubeCad for V1a, I get an output impedance of 29K. V1a is a grounded cathode amplifier. Plugging the values in for V1b, I get an output impedance of 624 ohms. V1b is a cathode follower. It has no gain but a low output impedance. It's the trade off. You want lo Z for traveling long distances and you want your output impedance to be much less than the input impedance of the next stage, 1/10th is an ideal design target. I have a sand state power amp whose input impedance is 10K. If i tapped off V1a, I'd be loading it down and therefore decreasing gain. 29k is not less than 10k and these are in parallel so effectively I change the output impedance to roughly 7.5k. Now tapping off V1b, we're putting 624 in parallel with 10k = 590 which is still very close to 624.
>What will the output impedance be with 'The Oatmeal' and will it drive today's modern +4 dB equipment?
Call it 600 ohms and yes.
-
You know, this time, I think I actually understand you! Speaker impedance is well known to me, so being able to take the next step isn't such a stretch.
That calc. page is awesome, I could have used that many times in the past too! The dbU and dbV conversion thing always puzzles me.
Did you say you had bread-boarded V.2 and V.3? Did you have the parts around or did you order them from Hoff.? If you made up a parts/order list, can I have it? I think I'll go through the schem. and order tomorrow, but I don't want to forget anything.
I think the Plate Follower may work best, the Alembic drives the compressor I have after it pretty hard - have to be very careful RE: distortion.
If I were to want to add the 'James tone stack' later, would I be able to just cut into this circuit and add it, or do I need to make the decision now?
-
I have a garage full of parts. I built from parts on hand.
-
Well, here's the parts list. Looks like I can get it all from Hoff, except the chassis and X4MR. Look it over, did I miss anything? (I have wire, solder, etc.)
1 Power cord & IEC input
1 Fuse (Amps?)
1 Transformer Hammond P-T442
1 Power Switch 2-pole
1 LED & resistor(?) for 12V power on/off indicator
2 Octal tube sockets & mounts
2 knobs
4 Switchcraft (mono, switching)
2 Audio taper pots 500k (for gain control)
1 Breadboard & pegs
1 Chassis
2 diodes 1N4007
2 cap 100u/350
1 cap 47u/350V
2 cap 35u/25V
2 cap 0.1UF/100V cathode bypass
2 cap 2.2UF/100V cathode bypass
1 resistor 25k/2W (* is 22k or 27k okay?)
6 resistor 1M
4 resistor 47k 1/4W
2 resistor 1.5k 1/4W
4 resistor 100k 1/4W
2 resistor 1k 1/4W
-
> I don't remember the grid impedance of a 6SL7
Over 100Meg at low frequency.
About 60pFd too. This is 100Meg at 26Hz, 10Meg at 260Hz, 1Meg at 2.6KHz, 100K at 26KHz, 10K at 260KHz.
Together, ~~1Meg from DC to nearly 2KHz, falling to below 100K above the top of the audio band.
> I get an output impedance of 29K.
This stage will drive the one above to the top of the audio band, but not a whole lot more.
Note that the 47K in the oatmeal-1 has an effect on input impedance.
-
> what all this +4dBu
That's when you KNOW your level. Sound for the Olympic Games, the Nightly News, etc is adjusted to standard level on the network, so that any affiliate can cut-in without any surprise or overload.
For you and a sound person, the actual voltage is unimportant. It will vary a LOT with your music, your mood, from rehearsal to take/performance, and from piece to piece. You want enuff voltage to overwhelm any buzz the wire picks up, but typically not a problem. You would like enough voltage to flick the "signal" light on the mixer input, but most mixers can pull-up quite weak signals. You don't want so much that anything overloads. The design fits these goals.
-
Impedance....
Wall-power examples.
I have a 60 Watt 120V lamp. Obviously it must suck 0.5 Amperes at 120V to eat 60 Watts. The ratio 120V/0.5A is the working impedance: 240 ohms.
(I say "working" because an incandescent lamp will read lower when cold; pay no attention to that.)
I have a heat-blower motor which, at start-up, pulls 60A at 120V: 120V/60A= 2 ohms.
Those are load impedances.
What are source impedances?
There is a hydro-power dam upriver here. If it were wired for 120V it might _safely_ put out 4,000 Amps: a 0.03 ohm load. Say I go there, measure 120V exactly, then plug-in my 60W lamp and measure 119.9997V. A 0.5A load causes a 0.0003V drop. 0.0003V/0.5A= 0.0006 ohms output impedance.
Back at my house, at the end of a long feeder wire, that same 60W lamp and 0.5A load causes 0.1V drop. 0.1V/0.5A= 0.2 ohm output impedance of my 400-foot wire.
It is usually "best" to use low-Z sources and hi-Z loads. For example, with my 0.1 ohm feeder wire, one 60W lamp causes a negligible 0.04% drop and 0.04% power-waste in my feeder. "Good power practice" allows 2% drop, I can have forty-eight 60W lamps all at once.
My feeder wire is not "best practice". When my blower kicks-on, the 60A draw in 0.1 ohm feeder drops my voltage by 60A*0.1ohm= 6 Volts, a 5% drop. This causes lamps to dim. That's why "best practice" calls for 2% drop. However in this case, a 2%-drop feeder would cost me $10K++. I have decided that my lamp-dim is more acceptable than the cost of no-dim feeder. The "cost of sag" is often a real trade-off.
Audio:
In power situations we may assume a single 50 or 60Hz frequency. Audio runs 20Hz-20KHz. Many real sources and loads are not the same voltage/current ratio at all frequencies.
In domestic power, the "best practice" says 2% max sag because the eye is good at detecting lamp-dim (and because many lamps' output falls faster than voltage). The ear is no so sensitive, and hardly notices 10% "dimming" (1dB level drop).
So the common rule for audio is to make your loads >10 times higher than your sources.
Imagine 1.000 Volt coming out of a 1 ohm source. If we load with infinity, we get 1.000V. Load with 100 ohms, get 0.99V. Load with 10 ohms, get 0.9V (actually 0.90909...V). Load with 1 ohm, get 0.5V. The ">10 rule" means you have 90% of the most you could get, and 90% is as good as 100% for most audio purposes.
When impedances vary with frequency, pick the worst-case.
A gitar pickup with cable is 5K resistance at low/mid frequency but rises to 100K at some medium-high frequency (and then falls off). Take the 100K worst-case. You want to load with 10 times higher: 1Meg.
Many audio chips have output impedances below 1K. Most line-level box-loads are designed for >10K input impedance. In current custom, often 22K.
Back when tubes ruled, you could not expect a source (such as preamp) to be less than 10K output impedance. Most line inputs were over 100K input impedance.
If you hang a modern recorder on an old tube hi-fi preamp: 22K load on 8K source, you only get 73% of the un-loaded voltage. This may be acceptable. Maybe it is the only preamp you have. Maybe it has cherished flavor (but be sure the flavor is not spoiled by the heavy load).
Just for completeness:
Looong telephone lines, so long that the wire parasitics matter, have impedance of 100-900 ohms (depending on construction). And a very long line can echo, and this can be minimized with a "matching" resistor. So there is a vast body of gear with 150/600 ohm nominal impedance. Also when tubes were COSTLY, you did not waste them making low-Z outputs. You transformed to nominal impedance, then did the math to figure what really happened. If an output is true 600 ohms, and an input is true 600 ohms, the voltage drops 50%; however tube-cost is minimized.
Since the cost of a tube fell below a week's-pay, most gear has been designed for low output Z and high input Z. The few situations where you "must" match can be loaded with resistors. Most "pro line inputs" are over 10K actual impedance.
The "oatmeal-1" output stage has internal impedance a bit under 1K, low-enuff for 10K loads. It won't deliver much more than 1mA output current, but 1mA in 10K load is 10V peak, and musical instrument interfaces should rarely need more than 1V or so on the line.
-
>1 Transformer Hammond P-T442
I was wrong, it's not a Hammond, but it's on the Hammond page. Go figure.... Run with it.
>2 cap 0.1UF/100V cathode bypass
2 cap 2.2UF/100V cathode bypass
These are coupling caps, go 350V or greater.
>1 resistor 25k/2W (* is 22k or 27k okay?)
Yes, 22k will put you a bit above 250V, 27k a bit below. Again, this is a shoot from the hip design so you might need to play with this value. It won't explode one way or the other. Higher voltage tends to add a little punch up to a point where it starts to sound crappy. When I bread-boarded this circuit, all I really built was the amp circuit. I tapped off an existing power supply.
>6 resistor 1M
4 resistor 47k 1/4W
2 resistor 1.5k 1/4W
4 resistor 100k 1/4W
2 resistor 1k 1/4W
Go with 1/2 watt. 1/4 watt would be fine but for a few pennies more, you get something easier to work with. Oh yeah, 1k is peachy for the LED current regulating resistor.
-
>Looong telephone lines, so long that the wire parasitics matter
Triggers a tangent. The origin of the spring reverb. (I know we discussed this before, in fact I think PRR posted this link)
http://www.accutronicsreverb.com/history.htm
(copy-n-paste)
Laurens Hammond needed to find a way to add reverberation to the living room. He discovered that Bell Labs had devised an electromechanical device to simulate a single delay experienced on long distant calls. The device used two springs to transmit the delaying signal and four additional springs to dampen and “center” the driver saddle. While the dampening were housed in long tubes filled with oil, one of the springs transmitting the delay signal ended in a short tube which, by varying the amount of oil in the tube, varied the decay time. After modifying the reverb to create many echoes, it was perfect for Hammond’s needs.
-
Hey Buttery -
Could I re-introduce the 'pseudo-balance' configuration from your 1.1 split-load design into the 1.3 plate follower design I'm making? (or does that completely change the plate follower design...?)
Then I could use XLR outs, and feed the balanced inputs on my compressor.
Ordered the chassis yesterday, I finally have some time coming up, I'll be building in the next week.
-
>Could I re-introduce the 'pseudo-balance' configuration from your 1.1 split-load design into the 1.3 plate follower design I'm making?
Yes you could but according to PRR
>The split-load's balance is a matter of debate, and IMHO not a good way to drive a "balanced input jack".
I am inclined to agree. I've actually never used a split load as a direct out. I've always used a transformer, unfortunately they get kinda buckish. The CineMag CM-2810 would be ideal. You simply hang it right off the existing output.
-
Hmmm...
A Google search for CineMag "CM-2810" yields nothing, nor does a search of their website. Is that the right number?
But yeah, it seems a transformer is the way to go. If it's too pricey though, I'll suck it up with a 1/4" tip/sleeve. I don't need the output to be balanced, and 600 Ohms is plenty low enough output impedance already to drive the 10K load of he next unit's input. I was just hoping to keep it 'tidy' with all XLR to XLR connections.
One other thing -
Say I decide to put the James Tone Stack in later. Can I just cut in and 'insert' it someplace after the fact, or will it involve some major circuit re-design?
TIA !
-
How about one of these?
http://4sale.sbszoo.com/xform-aud.htm
-
root URL:
http://cinemag.biz/index.html
device of interest:
http://www.cinemag.biz/output/CM-9589.pdf
-
Edcor USA. http://www.edcorusa.com/Products/ShowProduct.aspx?ID=334
-
Here's a revised schematic. If you're gonna add balanced out, you might wanna add a polarity switch (optional) and you for certain want the ground lift switch.
-
Great!
A question -
At J102 - the signal lead would need to be connected to the signal lead from the top channel in order to mix the two, right? As it is drawn, it seems channel B (bottom) isn't connected to an output?
Also, there's no need to have a separate channel out unless I wanted to run stereo for some reason right? I can just connect the two channels for a blend as far as I can tell...
What do you guys think of this faceplate layout? There are two ideas - channel A or B. I was thinking of opening a hole to view the tube. Dumb idea? (tone stack to come later).
I went with the jacks toward center, and the channels working 'outward' as that seems to be where my eye gravitates in my current rack... Dumb idea?
I wouldn't do the 'Hire me, I'm easy red' color, it would be a very muted deep red over black... IF I color it at all.
(http://farm5.static.flickr.com/4060/4429013971_14d786b421_o.jpg)
-
I like the chassis layout. If you're going to add a tone stack, at that time you'll probably need to eliminate Rfb & change R2 to a 22 or 33k. W/O the feedback, gain will increase probably enough to compensate for the losses associated with the tone stack. Also w/o the feedback, R2 will affect the hi end. I would suspect that 47k will cut into the range of hearing but it might be OK.
The way I showed the output, if the 1/4" jack is plugged in, then the XLR output and it's associated transformer are disconnected, otherwise your output transformer primary is in parallel with your hiZ output and it'll sound like there's a sock stuffed in it. It's an either or output situation but..... Since the input jacks are wired such that when using the (left) mono jack, sends signal to both halves of the unit, you can use the hiZ out on 1 channel and the loZ out on the other. You'll have separate level controls (& tone stacks) if you go this way. If you want hiZ and loZ simultaneously on each channel, you'll need to add another stage which blows the whole power supply since we maxed out that tiny cheapo power xfmr.
I finished the revised schematic.
-
... Also w/o the feedback, R2 will affect the hi end. I would suspect that 47k will cut into the range of hearing but it might be OK.
So, using a lower value resistor (like 22k or 33k) will not affect high frequencies, but the 47k possibly might??
The way I showed the output, if the 1/4" jack is plugged in, then the XLR output and it's associated transformer are disconnected, otherwise your output transformer primary is in parallel with your hiZ output and it'll sound like there's a sock stuffed in it. It's an either or output situation but..... Since the input jacks are wired such that when using the (left) mono jack, sends signal to both halves of the unit, you can use the hiZ out on 1 channel and the loZ out on the other. You'll have separate level controls (& tone stacks) if you go this way.
I'd never use both Hi-Z and Lo-Z out's simultaneously. I'll only ever use a single Lo-Z output, and only one input at a time.
Since that is the case, could I simply wire channel 1 direct, so that is NOT feeding channel 2, and then sum the two outputs to a single OPT? This way, no matter what, I get one Lo-Z output, no matter which input I use. (not to mention save some $)
I finished the revised schematic.
Wow! I'm even learning how to read schematics here, this actually (magically) makes total sense to me now.
-
I'd never use both Hi-Z and Lo-Z out's simultaneously. I'll only ever use a single Lo-Z output, and only one input at a time.
Since that is the case, could I simply wire channel 1 direct, so that is NOT feeding channel 2, and then sum the two outputs to a single OPT? This way, no matter what, I get one Lo-Z output, no matter which input I use. (not to mention save some $)
In other words, could I do this? : (see attachment)
-
I don't think you can sum the output as shown. Obviously you could simply delete 1 channel all together and recalculate the power supply, or you could make a 2 in 1 out mixer as shown below. Personally I'd stick with the 1.21 version. With a mono input you get dual outputs with separate level controls, a glorified active direct box if you will, or a straight up stereo in/stereo out.
-
Dang!
6 week wait for the Edcor output transformer(s).
Looking at the Alembic f2-B circuit, (and my unit) it sums the two outputs using a 47k resistor on each. I actually got rid of the resistors for more output, and it's working fine so far... I think it's worth a shot.
If it doesn't work though, I will stick with 1.21 version. I'll have to run around back when I want to switch channels though. - Only one input to the compressor...
Chassis holes drilled/stamped!
-
Schematic question, and picture.
In the oval, there is an arrow pointing to the pot [EDIT: NEVERMIND, FIGURED IT OUT WHEN DRAWING THE WIRING SCHEME] C4/C104 cap.
Also, I'm guessing the dashed line means that the two pot 'cans' are connected via a grounding bus?
Here's a 'so far' pic. Power supply and power wiring done. Heaters and B+ wired.
All that's left to do is the audio path. Oh, and wait 6 weeks for the blasted output transformer(s).
-
The dotted line represents a "ganged" pot but isn't necessary or even preferable if used as a dual mono. Looking good so far.
-
Hey B,
Got channel 'A' all wired. All I get though is a very small and very distorted signal. What I'm wondering is if I have the two 'stages' or units of the tube crossed.
In the schematics I've studied, the first unit corresponds to pins 4,5 and 6. The second unit corresponds to pins 1,2 and 3. On this Oatmeal schem, the input signal goes into the second unit first, 1,2,3) and then to the first unit. Is there a difference in gain etc. between the two stages? Or are they identical and it doesn't matter which is first?
It would make total sense if I had the wiring exactly backwards - giving me a small, distorted sound...
-
The 2 triodes match. It does not matter which 1/2 you use. Double check your wiring. Make sure filaments light up on both tubes. Even if you don't have the 2nd bottle wired for signal, you still need the filaments wired in series since you're using a 12V supply for 2 - 6V tubes.
OK, so check some DC voltages now.
First look at point A. Is it 250VDC (or close +/- 15%)?
Now look at the cathode of the first stage.(pin 3 or 6 depending which 1/2 you used). It should be in the neighborhood of 1.5VDC. The plate (pin 2 or 5) should be.... call it 155VDC.
Next look at the 2nd stage. Obviously the plate is 250V. The cathode should be 100ish VDC.
-
The 2 triodes match. It does not matter which 1/2 you use. Double check your wiring. Make sure filaments light up on both tubes. Even if you don't have the 2nd bottle wired for signal, you still need the filaments wired in series since you're using a 12V supply for 2 - 6V tubes.
All good -
OK, so check some DC voltages now.
First look at point A. Is it 250VDC (or close +/- 15%)?
Not even close. It fluctuates widely from millivolts to about 80 volts, depending on how hot the X4mr is. I double checked the power supply wiring, I think it's right. The transformer gets too hot to keep a finger on for more than a few seconds. The heater voltage is only about 4V and the DC out from the primary leg is only about 40 volts. Looks like I have a dud? I think I should be seeing about 125VDC entering my power supply (?)
-
Sounds like you have a shorted or reversed filter cap, or you miswired the voltage doubler power supply.
Disconnect everything from the PT secondary and measure it's AC voltage. Filament voltage should be right now too if the transformer isn't injured.
Next check your orientation (wiring) of D1 & D2, C5, C6 & C7. Reconnect the PT secondary but disconnect the amp circuit from the power supply at point A. Check voltage now. W/O load it should be something around 265ish VDC.
-
Sounds like you have a shorted or reversed filter cap, or you miswired the voltage doubler power supply.
One diode - D2 - was backwards I think, which I repaired. The stripe side is negative - cathode, and the plain side is positive-anode, right? In the schematic, the arrow points toward the positive end, right?
Disconnect everything from the PT secondary and measure it's AC voltage.
If I place my Multi-meter contacts one on each of the two leads, I get 120VAC (there is nothing DC.) I've looked over and over, it must be that I can't read schematics. The + is the positive side of the cap, and the U is the negative right?
Filament voltage should be right now too if the transformer isn't injured.
it is, about 5.5 - 5.8 VAC.
Next check your orientation (wiring) of D1 & D2, C5, C6 & C7. Reconnect the PT secondary but disconnect the amp circuit from the power supply at point A. Check voltage now. W/O load it should be something around 265ish VDC.
Nothing I can find, I'm stumped. Have a look. I had one of the red leads between the 2 diodes, and the other at the +/- junction of C5 and C6
-
something is sucking too much current - perhaps there is a short under the board? based on your description of where the PT leads went, it seems to be wired correctly. does the PT have a CT? if yes, is it grounded? it shouldn't be. confirm both diodes and the caps are good.
http://www.allaboutcircuits.com/vol_3/chpt_3/2.html
-
>The stripe side is negative
Stripe is +.
-
No transformer CT, and no shorts. The diodes seem to be good, I don't have the means to test capacitance.
One thing I did do, is to cut the bus between c6 and c5. With that done, I do get about 60VDC at B+, but that's it.
Also, the grounding signs in the schematic... - doesn't the AC stiff have to go back to neutral, (like the AC heaters, and AC ground from the power supply before it's converted to DC?) and only the DC stuff gets grounded to chassis/safety, right? I played around a little with this, but nothing conclusive.
I'm pretty sure the transformer is fine. Taking it away from any connections causes it to measure fine. When connected though, it's all the same as I've described.
-
any chance of getting pics of the entire build?
:smiley:
-
ISO -
Yep, here ya go, thanks for any help!
Two alternate views of the PS. The cap in the middle of the pics is C5. C6's NEG was attached to C6's POS side, and one of two leads from the transformer was attached to this. I have snipped the bus between them, and that's how I got the 60VDC on B+.
Also the rest of the build. Some is messed up as I've pulled apart stuff to test...
-
I think you have a bad diode. Your layout looks correct, but just in case, attached is a redraw of the power supply.
How to check a diode:
Set your VOM to ohms. With at lest 1 end of each diode out of circuit, connect your meter across the diode, then reverse your meter leads. 1 way you will get almost a short reading, the other way you'll get an open circuit.
>One thing I did do, is to cut the bus between c6 and c5. With that done, I do get about 60VDC at B+, but that's it.
When you did this, which cap (c6 or c7) did you leave the PT secondary connected to?
I think the PT is fine, but something is obviously shorted.
-
I think you have a bad diode. 1 way you will get almost a short reading, the other way you'll get an open circuit.
D1 = .6 M Ohms one way, Open circuit (no connection) the other. D2 = .7 MOhms one way, open circuit the other. Both = bad, right?
EDIT: JUST TESTED A NEW ONE AND GOT THE SAME READING. ALSO .5V FORWARD VOLTAGE WITH THE DIODE CHECKER FUNCTION ON ALL.
When you did this, which cap (c6 or c7) did you leave the PT secondary connected to?
I left power between the two diodes, and to the + of C6 (C5's Neg was no longer connected).
-
Diodes are good. C5 looks burnt on the end. It's really too close to that turret IMO. Measure with VOM. It should NOT read short.
>I left power between the two diodes, and to the + of C6
So C5 is now out of circuit. Is the transformer still getting hot? What do filaments read?
SAFETY NOTE:
You do know those caps will store a charge w/o a load connected right? Be careful and bleed the caps before touching.
-
Diodes are good. C5 looks burnt on the end. It's really too close to that turret IMO.
Yeah, that's me soldering and unsoldering the power lead for tests. Measure with VOM. It should NOT read short.
C7 and C5 have varying resistance BUT - C6 has none! That would then short, right?
>I left power between the two diodes, and to the + of C6
So C5 is now out of circuit. Is the transformer still getting hot?
yesWhat do filaments read?
About 4.5VAC or so when lit.SAFETY NOTE:
You do know those caps will store a charge w/o a load connected right? Be careful and bleed the caps before touching.
Yes, but thanks for the heads up. To bleed, I'd connect a grounded alligator to the +, right? Either way, I've been zapped before, it makes you very conscious of where your hands go!
-
Checking an electrolytic cap for shorts:
Take the cap out of circuit. Put you VOM on ohms, Black to - Red to +. This will put a tiny charge on the cap. You meter should read greater than 1 mega ohm and slowly climb. This would be a non-shorted cap. If you're reading anything substantially less than 1M, it's trashed.
>About 4.5VAC or so when lit.
Is that across each tube or off the PT filament secondary?
You did wire the filaments in series right?
Pull the tubes. What do the voltages do now?
(Power supply point A is disconnected from the tubes right?)
-
> About 4.5VAC or so when lit.
Is that across each tube or off the PT filament secondary?
The 4.5 -5 is when measuring lead to ground on a single tube with the tubes in the sockets.
With the tubes out I get about 40VAC measuring from one tube's lead to ground.
Pull the tubes. What do the voltages do now?
I get 12VAC measuring both leads.
You did wire the filaments in series right?
Yes in series - One lead goes to one tube's 7 with 8 grounded, the other lead goes to the second tube's pin 8 with pin 7 grounded.
(Power supply point A is disconnected from the tubes right?)
Power supply point A has never been disconnected, no. It runs to each tube's pin 5
If you're reading anything substantially less than 1M, it's trashed.
There is no connection when I test the cap - OL is all I get on the meter. (in other words, infinite ohms)
-
OH Kay. Isolate the power supply from both pin 5's at point A. We're trying to get the PSU up with no load to make sure it's OK.
Put the tubes back in. Disconnect the tubes from point A. Put the PSU back together (D1 &2, C5 6 & 7, and R10). Recheck voltages.
-
Yes in series - One lead goes to one tube's 7 with 8 grounded, the other lead goes to the second tube's pin 8 with pin 7 grounded.
what you describe is not a series heater circuit... wire transformer lead 1 to pin 8 of V1 - pin 7 of V1 wired to pin 7 of V2 - pin 8 of V2 to the second filament lead of transformer.
transformer(1) ---- 8 V1 7 ---- 7 V2 8 ---- transformer(2)
-
Yes in series - One lead goes to one tube's 7 with 8 grounded, the other lead goes to the second tube's pin 8 with pin 7 grounded.
what you describe is not a series heater circuit... wire transformer lead 1 to pin 8 of V1 - pin 7 of V1 wired to pin 7 of V2 - pin 8 of V2 to the second filament lead of transformer.
transformer(1) ---- 8 V1 7 ---- 7 V2 8 ---- transformer(2)
Yes it is. Virtual center tap between the 2 tubes.
-
Yes in series - One lead goes to one tube's 7 with 8 grounded, the other lead goes to the second tube's pin 8 with pin 7 grounded.
what you describe is not a series heater circuit... wire transformer lead 1 to pin 8 of V1 - pin 7 of V1 wired to pin 7 of V2 - pin 8 of V2 to the second filament lead of transformer.
transformer(1) ---- 8 V1 7 ---- 7 V2 8 ---- transformer(2)
Yes it is. Virtual center tap between the 2 tubes.
yes, you are correct - why V tap? how do you balance? i guess it really doesn't matter on w/ class A.
-
You wanna ground the filament circuit somewhere. Seems ideal for both tubes to get grounded. Balance from = filament current.
-
Re-assembled the power supply as before - with C6 cap (?).
Removed B+ from both pin 5's. X4mr still hot, AC measures about 40 VAC from point A to ground. DC voltage is negligible (Less than a volt)
Heater filament voltage is at about 5.5 to 5.8 VAC
-
When I remove the two leads from the transformer and test them, the transformer is still outputting 120 VAC.
I can't see that there would be any shorts, There is nothing on the underside of the PS except the turret backs, and the two leads from the transformer. I've even removed the point A feed entirely...
It's now GOT to be something in the PS itself, it isn't connected to anything else. The diodes I think are okay, and the resistor is ok. It has to be one or more of the caps (?)
Maybe I should just order all new parts and make a new one from scratch...
-
Something is shorted, caps or diodes or just a plain hard wired shorted.
>Maybe I should just order all new parts and make a new one from scratch...
I'm going to say yes. Even if you get this up and running, you're going to have hum issues. I'm certain I see a ground loop and your layout has problems. I'm gonna sketch up a layout real quick. I would suggest you use these
(http://www.hoffmanamps.com/MyStore/catalog/images/IMG_2369_small.jpg)
Cheap and effective. Easier than drilling a turret board. I love em.
-
I've seen those in a lot of high quality, old style point to point builds, which I guess I'm trying to emulate. Good idea, only I just ordered new parts from Hoff (I'm impatient) . And, Dang him, he's so fast, he had them shipped out an hour later so I can't change it. I can probably pick some up at my local store.
Thanks for doing the layout, it helps guys like me who are schema-tarded.
As a side note to the 'original' topic of this thread:
I just did a quick session today of a couple cover tunes, one with a fretless, and one with a fretted bass. The engineer was blown away by how quiet my F2-B was, and also how punchy and phat. I just plugged in to his feed, and the tone was there instantly. You really get to hear what your gear sounds like when you record it.
-
I recognize the style you're emulating. It begs for trouble. Have a look at the attachment. I tried to accommodate your existing chassis config, but you're gonna hafta move the tubes. I didn't draw the PT or line supply. What you got is fine. Separate chassis grounding point for the power cord. Keep the filament wires as far away from everything else as humanly possible. I didn't draw a balanced output. Get a couple of extra tag strips for that. Pay close attention to the grounding scheme. No extra wires. No need for a #12 ground bus.
-
That certainly looks a WHOLE lot easier...
I'll see if I can pick some strips up tomorrow. If not, I'll get some from Hoff.
Other than to mount the tubes outside the chassis, (Thus defeating the single rack space) I don't know where else to put them. I need the clearance above them to remove/replace. I can't really go to one side or the other, there's the PS on one side, and the upcoming OT on the other. I think I could replace the tube mount board and get the sockets all the way on the back side of it and still have just enough clearance. I can also run longer filament leads, maybe along the back wall and then straight in.
I think if I get rid of my 10 GA. ground thingie, I could mount a row of tag strips, and go directly to the pots / input jacks from there?
You know, I'm having so much fun with this... :laugh: ... Do you still have the circuit boards you had made for the bass pre-amp you designed? I think after this torture, I'll be ready for something even more painful, LOL. If so, how much would you want, and do you know an approximate component cost?
-
You could prolly turn the audio section 90 degrees and make it it. The tag strip area is roughly 2.25" x 6".
I have 1 unpopulated board left. I can't remember what the BOM priced out at but it wasn't cheap.
-
> doesn't matter on w/ class A.
Yes it does.
> Balance from = filament current.
I don't see that. What we mostly want is voltage cancellation, a "virtual Neutral" in electrician's terms, AT each socket.
That silly AC is always wobbling. "Freeze" it at a peak. With the conventional guitar-amp heater wiring, we have +9V on one lead and -9V on the other lead. A point in the middle, or a point at a distance, feels the difference: Zero volts.
With your wiring, V1 feels the difference of +9V and zero, or 4-some V; V2 feels the difference of -9V and zero or negative 4-some V. It is in fact equivalent in hum-effect to grounding one side of the heater, as the oldest amps did.
There isn't a clean answer for 12V power and 6V preamp tubes.
If you could stand 12SL7, you would wire normal with two 100r resistors. And 12SL7 are still a buck cheaper than 6SL7. And they may not be making any new 12SL7. But I dunno if the karma is identical.
Does the PT have 12VAC at well over 0.6A? While I say DC heat is a frill, and that good twisted 6V is usually penty fine, a bass front end for studio work might get my blessing on DC heat.
-
I've seen that 12SL7 has identical ratings except for the heater voltage, and it'd be easy to wire a switch or something to use both types... But what does the tube sound like? Is it identical in sound?
There do seem to be quite a few on ebay.
I'll either pick that stuff up tomorrow, or order it. I want to do this at least semi-well. Even if it is my first learning project...
-
Oh yeah, what's involved in changing an AC heater to DC? Does that open a whole new can of worms?
There's AC in my F2-B as well....
-
Success! :grin:
Power supply rebuilt, about 200V to point A. I changed out everything except the original PT. There had to have been a short somewhere, maybe under the board. The PT is nice and cool now, and everything is fine.
To increase to the specified 250V range, I can lessen R-10's value a bit, I assume. Is it crucial that I hit as close to 250V as possible?
So, then I totally re-did the circuit using your layout - except I couldn't fit both terminal blocks, so I used one, and changed a couple things. (I'll take a pic tomorrow) I re-did the routing for the heaters as you suggested, as well as the whole grounding scheme. - Several other things too. I didn't move the tubes - Took the risk on hum, but so far it's dead quiet.
You were right, those tubes are markedly different than the 12AX-7. I was instantly reminded of an old Portaflex. Big fat whompin' tube-ness. It makes me want to slap the black right out of my bass :grin:
Channel 2 isn't right yet. It has about half the gain, and my 3-way switch for R102 (between 0, 33 and 68K) doesn't function either, but I think this is minor. I'll bet the switch is broken, or the feedback cap is wrong, I'll try to trace it down tomorrow....
Channel 1 however is perfect! I might mess with the R2 resistors a bit, but it already goes clean to overdriven just with the gain knob.
I swapped several tubes, and the actual tube used in the circuit affects the sound much more than in my Alembic. The Alembic is subtle at best between tubes - this is not at all.
Tone stack? I dunno yet, I want to play this tomorrow through my rig before I make that decision, but so far so good!
-
Rock on! No, it's not critical you hit 250V, but you have the right solution. You'll probably get more "punch" with a voltage increase up to a point where distortion will take over. Go too far and you cook the tube. I'm working on another project for Kagliostro and right now I wanna rip the tone controls out and throw the parts back to the fishes. If you like it w/o tone controls, don't mess with it.
-
How high can I go with it? No such thing as too much punch, but yeah, don't want distortion, nor do I want to blow up the tubes.
Pix tomorrow.
-
How high can I go with it? No such thing as too much punch, but yeah, don't want distortion, nor do I want to blow up the tubes.
Pix tomorrow.
experiment... find the sweet spot you like.
-
I'm reading that 300 volts is the max, is that right? Will this PS do it, or will I need some bigger / more caps?
And yeah, I think regardless, this pre-amp will be tone stack free. That is what this whole concept was about, and I must say I'm excited about the results I'm just starting to get. Now I want to build another - like on the PCB you have left. Money's tight, but I could just do it little by little. Do you have a parts list?
Oh, and check this out - I went back and re-read some of the more technical posts made earlier in this thread - it makes SO much more sense to me now! I think I'm learning!
-
> 300 volts is the max, is that right?
On the tube. (And 330V is factory-safe if measured; we know these tubes will stand more.)
But the supply voltage SPLITS between tube and 100K resistor. 500V maybe 600V will not leave 300V on the tube, nor will it cause excess dissipation.
One other detail. The "a" section of each tube runs with cathode near ground, the "b" section of each tube runs with cathode at high voltage, nearly half of supply voltage. The heaters inside both cathodes are connected together. The heater-cathode insulation has a voltage rating. G.E. says 6SL7 heater-cathode voltage should not exceed 90V.
If heaters are grounded, as B's plan shows, the "b" section's cathode should not be over 90V. That suggests the supply voltage must be less than about 200V.
Actually the "b" section could be biased for a lower cathode voltage. Change its 100K to 47K. No real change in audio performance but significantly lower cathode DC voltage and lower heater-cathode insulation stress. Now over 300V supply should not be a problem. And I don't see a reason to need over about 250V supply, for this gain and normal instrument pickup output. Which is a very traditional B+ level, and what Butterylicious suggested.
-
Most interesting. I'm kind of understanding these talks a bit more now.
Basically, I could lower R8/108 to 47k, (or maybe 68k?) and drop the R10 resistor a bit allowing more flow, but less voltage. Interesting based on the second part of what you said - normal instrument pickup output. Mine's anything but normal. I use active EMG's and like to thwack the heck out of my bass - I'm not an easy player, I really dig into it. Dynamic range and headroom - while keeping that transient 'punch' has always been what I'm after. So, giving the tube plenty of reserve to keep up, sounds like it's worth experimenting with.
I do think the 6SL is a better deal for me than the 12AX-7 It still breaks up, but it sounds a lot nicer (to me) doing it. The I don't like how the 12A breaks up on bass. It's great on guitar though.
The big point is that by changing these, I'm not going to blow anything up, right? Is there a point at which I'll need to add more / bigger caps, or is there 'plenty'? Maybe I should put in a couple switches.
As ISO said, Play around until you get the sound you like.
-
>I don't see a reason to need over about 250V supply, for this gain and normal instrument pickup output. Which is a very traditional B+ level, and what Butterylicious suggested.
Right on. I picked 250V for several reasons. 1st it is a traditional value which means as you learn, and start reading tube data sheets (http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6SL7GT.pdf) you can find information at this voltage w/o having to recalculate or estimate for a different voltage. 250V is relatively a "safe" voltage. If you get across this B+ it will certainly get your attention but it probably won't send you screaming from the room. It shouldn't explode if you wire something up wrong. It'll probably just make the PT get real hot. :wink: Finally because I'm lazy. Working at typical voltages are quicker easier to design.
>The big point is that by changing these, I'm not going to blow anything up, right?
You've got the unit working correctly, so you're pretty much past the point of doing any real damage. I assume you've gotten the 2nd channel to work right. Please do this before you get into experimenting with the circuit, ie. keep the horse in front of the cart. If 1 channel isn't conducting properly, you probably are drawing the correct current, so you're shooting yourself in the foot to be messing with the power supply voltage before you take care of this.
The power supply cannot possibly make anything greater than 350VDC. I doubt it's actually making 350V since the PT is so small. I didn't have one of these on hand so it was all a guesstiment. Here's what you need to do. Measure the voltage at the point where D1, C5, & R10 come together. This is your actual max B+. I normally call this point A, but since there's no tap here, I didn't want to confuse you (until now) :angel Let's call this point "X" so it has a name without a + symbol. Now measure the voltage at point "A". Subtract the "A" voltage from the "X" voltage. This is the voltage dropped across R10. Now take that voltage drop and divide it by the value of R10. This is your total current. My estimate looks like this: 350/250=100V 100/25000=.004 or 4mA. You mileage may vary. Ok, now take your voltage drop and multiply it by the calculated current. 100*.004 = .4 watts Using a 1/2 watt resistor would be foolish and destine for failure, a 1 watt would be safe, and a 2 watt will never get beyond a little warm.
OK, now you're armed with all the info you need to recalculate R10 for a higher voltage. Pick the voltage you want to run and subtract this from the measured voltage "X". Take this calculated voltage drop and divide it by the current you calculated. This is your new R10 value. 250Vshould be optimal. 300V [/i]should[/i] safe. Given the limits of this power supply, as long as you've wired correctly, it's gonna be hard to toast a tube.
>I could lower R8/108 to 47k, (or maybe 68k?)
Could and probably should. Of course you'll be flowing a little more current which affects everything I just talked about in the power supply. Welcome to the merry go round. It's not that drastic of a change but this is typical of playing with a circuit. Fix the power supply, jack with the circuit, & refix the power supply. (repeat if necessary)
-
Buttery - Thanks again for all this, you've definitely gone way beyond.
I actually understand - everything - you just said, and it's a great working order for today's experimenting!
Here are some pix, there's extra room over on the left for the OT(s) when they get here. In 5 weeks. Still, since the output is lower in impedance than my Alembic, I'm thinking I'll already have better luck with it.
I'll report back when I do some tweeking.
-
>Buttery - Thanks again for all this, you've definitely gone way beyond.
You are most welcome. This has been fun for me. There are 3 basic levels of learning, what you read, what you do, and what you teach. Typically one retains very little of what they read, they retain more of what they do, but almost everything they teach they retain so this is most beneficial for me. I'm glad your happy with the project. Your rebuild look +100% better. If you get hooked on this stuff (which BTW it IS addictive), you'll get better and better.
-Richard
-
You ar 100% correct. Not blowing my horn, but I'm a supercalifragialisticexpialidocious diver. Mixed gas, deep decompression, wrecks, the works. I remember learning to dive involved reading which helped get the ideas, but it was immediately (same day) followed by doing. - And then repetition. That's where the 'learning' really happened for me.
Once I had some salt under my weight belt, I learned and got certified to teach diving, and over time became able to teach all the way up into the technical diving realm. Going back and teaching the very first open water course, and the 'second level' stuff, even though I've now taught it maybe 40 times, is STILL beneficial to me. It's funny how you re-connect those basic blocks with some of the most abstract idiosyncrasies, once you're looking top down.
There you have it. A 'how-to' on connecting tube pre-amps to scuba diving.
-
Hey guys, this is great fun! :grin:
Fixed channel 2, it was a wrong value at R103 (note to self - watch that decimal! 1K is a far cry from 100K.
I'm wondering if, say, a 150K @ R3/R103 might increase clean output a bit more? The amp is clean at a gain setting of about 2, but not enough output. Setting of 5 (50%) is about where I need to be for gain, but it gets a little too 'fuzzy' at 5 as well.
Then I did as instructed RE: power supply / cathode ground and settled with a 6.8K on the PS, and a 47K on the c. ground. The results are 248 V at B+, 293V at point A. Today I think I will shoot for 250V with a resistor of 68, maybe higher, to combat some of the distortion.
So, I played at length with it in an actual band situation last night, as well as solo - loud through my amp... and here's what I took for notes...
After playing 'for real' my notes were:
1) Very Ampeg-y, perhaps too much. It is a little too warm (i.e. distorted/over-driven) for my tastes, and a little too 'clanky' in the mid-range. Some of this may be resolved in changing tubes.
2) Even more punchy, close and 'present' than my Alembic (Quite a feat actually, we're now approaching the punch-iness of Solid State - without Solid State sterility and 'gack')
3) Wouldn't want to mess with overall 'tone' too much, but would like a 'DEEP' and a 'BRIGHT' switch. A sweepable midrange at about 6dB cut would be the bees knees, but now I'm dreaming I think.
EDIT: One other thing, I'm noticing that my heater voltage is about 5.3V each tube. I'm also noticing that when I swap out a tube, the voltage on the non-swapped tube stays higher, like around 6.5V until the swapped tube heats up. During this time, the tube with the higher voltage sounds MARKEDLY better - Much less distortion, more clarity, etc. It's not subtle, the improvement is huge. I can then noticeably hear it fade away though as it's heater voltage re-stabilizes at 5.3V.
This is the same transformer I was using when I had the short for two days, and it got extremely hot. Is it possible I cooked it? What can I do to get that voltage back up in the 6 to 6.3 specified voltage range? (I have a new transformer ready to go, but was going to return it - should I try it instead??)
Comments / suggestions?
-
It's for certain, the biggest issue is low heater voltage. I put a dummy tube in one slot and got the working tube to about 5.8 - 6 Volts. The sound was markedly cleaner, and distortion free. Put a working one in there, and it's back to distorted, 'brown' sound.
I also tried an RCA Red Base 5691 6SL7 which I use for another amp in here, and it wouldn't make a sound. Popped it right back in the other amp, and it works fine. The other amp measures a solid 6 Volts.
So, how do I increase the heater voltage?
-
leave one tube out or get a stiffer transformer.
-
I was hoping you wouldn't say that. :wink:
Was doing some reading - the Hammond in there has an amperage of 0.3Amps. each 6SL-7 has a heater rating of .3Amps, I'm wondering if having run them in series, it's over-taxing the PT...
Can I use any PT that has a primary of 120AC, and a secondary of 6 - 6.9VAC?
EDIT:
OR - Couldn't I just run the two output leads in PARALLEL rather than series, and use two resistors to get 6.3VAC ?? It has a 12.6V / 300mA secondary output.
-
I was hoping you wouldn't say that. :wink:
Was doing some reading - the Hammond in there has an amperage of 0.3Amps. each 6SL-7 has a heater rating of .3Amps, I'm wondering if having run them in series, it's over-taxing the PT...
Can I use any PT that has a primary of 120AC, and a secondary of 6 - 6.9VAC?
EDIT:
OR - Couldn't I just run the two output leads in PARALLEL rather than series, and use two resistors to get 6.3VAC ?? It has a 12.6V / 300mA secondary output.
the problem is the specs on that tranny, you are operating it at the limit, so no amount of reconfiguration is going to get you any more power. you NEED a larger transformer.
:wink:
edit: here's (http://www.radioshack.com/product/index.jsp?productId=2102700) a $10 fix... make sure it'll fit into your chassis first though.
-
Well, I ordered a dedicated filament tranny. - a Hammond P-T166L6 6.3V CT 2 Amps. I'll have to do some jostling of parts in the chassis, but not a huge deal.
EDIT: Which may just barely not fit. Look what I found instead:
http://www.alliedelec.com/search/productdetail.aspx?SKU=2270042
-
Well shoot. I was afraid of that. Hammond usually rates their transformers conservatively but I fear that isn't the case with this particular transformer. It's listed on the Hammond page but it diesn't look like a Hammond to me. Sorry I gave you a bum steer on the tranny. At least there is a $10.00 fix.
-
Oh, no worries at all. Now I have an extra for my next project!
-
New question -
I have the filament transformer, it's just a bit larger than the P.T.
I'd like to put it in position one. Is this too close to the audio outputs, or will I be okay?
Position two is an option, but I will need to move and re-drill stuff. Not an attractive possibility.
Position three can't happen, because that is where the one (and possibly two) impedance matching transformer(s) will go. Plus, if I decide to go with tone controls later, this space will be even more cramped.
Can I have permission to place it in slot one?
-
Position 2 is probably your best choice. You want all that nasty AC stuff as far as possible from ANY signal. The only other option is to change out the existing transformer all together. I'm still kicking myself over that. A PA211 (http://www.triodeelectronics.com/pa2powtrandy.html) would do the trick, but you'd need to redesign your B+ supply whereas it's a 330-0-330 secondary. I didn't select this one in the first place because I didn't know if you should be playing with such high voltage. Given the initial start up problem, it mighta been scary vs. frustrating.
-
Yeah, that'd smart a little. Don't worry about the tranny, What I have now still results at $10 less than the Dynaco one. But yeah, in retrospect, I think 2X330V would probablybe just enough...
I'll try position 2. I can get it in there pretty easily, I think without moving the PS board. I just wanted to see if 1 was a better place, with the higher powered output signal, but I'll take your word for it.
-
Got the F.T. installed at position 2. Wasn't as difficult as I suspected it might be.
Here are the results of my new measurements.
- Voltage -
Heaters = 6.8 Volts
B+ = 282VDC
Point "A" = 318VDC
---------------------------
- Variable Resistor Values -
R8/108 = 68K
R10 = 68K / 3W
R2 / 102 = 0/18K/33K Ohms (3-way switchable)
Most likely, I'll just play through it for now, before making any more adjustments, it sounds great! The output trannies are still a month away, I can adjust then too.
-
Well, I've been playing and listening to it for a couple weeks. The only thing I'm missing is an Edcor Output Transformer (http://www.edcorusa.com/Products/ShowProduct.aspx?ID=334) because I'm still not driving my compressor with much oomph. If I understand what I've learned here about impedance, I'm basically spinning my engine really fast, to drive a really low gear...
It also has some distortion, but I don't know what I can do to improve that, other than turn the gain down. - But then I don't have enough to drive the following component in my rack.
I'm pretty derned happy with it otherwise, I still love the idea of "I made this" (sort-of) but there are a couple things I think I'd like to do to at least one of the channels.
1). a tone stack. I've fiddled a bit with the duncan calc, and I think the James stack would do me - but am wondering about some options. What about substituting the 'Big Muff' as the midrange control?
2.) I'd love to have a 'deep' switch. I feel like the boom in the room is a bit lacking at times.
-
bump
-
If you add a tone stack, you're going to lose gain. All of these passive stacks are cut only in nature. If you add a tone stack, you're gonna domino into other issues.
But.....
If you need better bass response, change C2 from a .1uF to something more along the lines of 2.2uF. Don't forget that needs to be at least a 400V part.
If you need more gain, change R2 from a 47k to a 33k. This will brighten it up a little bit too but it will not effect the bass response.
-
Ahhh, not that easy, eh. Gotta design the whole circuit together...
Guess I'll have to build another. Darn :smiley:
Here's an idea...
I have the EMG BQC System (http://www.emginc.com/products/index/105/146/4) in my basses. What I don't like about it is that I can never be sure it's 'set' I always have to check it if I put the bass down, etc. I'm thinking about mounting it after the input jack, but before anything in the circuit. essentially running the EQ circuit 'outboard' from my bass. Hmmm...
Maybe I should just add to my collection and build your design. How does it sound?
I'm also looking for a schematic for a Demeter VTBP-201 S (http://www.demeteramps.com/products/bassamplification/vtbp201s.html). I think it would be Da Bomb!
-
Hey guys,
My Edcor impedance matching transformers (http://www.edcorusa.com/Products/ShowProduct.aspx?ID=334) finally arrived. I will use one in my Alembic F2-B, and one in the new pre-amp. Primarily, they will be to lower the output impedance but a side benefit will be balanced outputs as well. There's a monkey in my wrench however, The transformer is center tapped on both windings.
I'm finding conflicting info as to what I ought to do with the CT on either / both the primary and secondary windings.
Two Questions:
1.) I know that Hi-Z signal will go to the (+) side of the primary. Do I simply ground the (-) side of the primary, and leave the CT un-connected to anything, or should ground that too?
2.) On the secondary side, I would go to pin 2 (hot) from the (+) side of the secondary, after that I get confused. Should I go to pin 3 (cold) from the CT, or from the (-) of the secondary? If I go from the (-) side of the winding, should I then ground the CT along with pin 1 (ground)?
-
Ignore both of the center taps.
Primary: Ground 1 side & feed the other.
Secondary: Tie 1 side to pin 2 and the other to pin 3. Reverse them and you've reversed polarity. Shoelace wire a DPDT switch and you've got a phase selector.
Pin 1 should go through a switch to chassis ground. 99 times out of 100, you'll run with the ground lifted to eliminate a ground loop buzz. You could get away with leaving pin 1 open.
-
Very cool, thanks!
Does it matter if the polarity is flipped or not? I know it would matter if you ran the instrument 'direct' first, and then into the pre-amp, but I'd never allow that. :smiley:
-
Phasing becomes an issue in recording and live if there are outputs. Consider your bass rig might be out of phase with the PA. As you turn up, everything gets louder but may appear to get quieter when standing in front of your cabinet. 1 miswired XLR cable from a DI can get funky. That's where the polarity switch comes in handy. The ground lift switch is a must. Again, 99% of the time you'll run with the ground lifted.
I expect your gain to drop considerably from the step down transformer. You'll probably need to tweak the feedback in one or both of the stages to compensate. I talked about this a few posts back.
-Richard
-
Yes, I'm ready for the output gain drop. - But I'm expecting to be transferring power more efficiently, which will enable better gain structure. Right now, both pre-amps don't have any 'oomph' into 20K loads.
I've been reading that since the Alembic has about 40k output, and the 'oatmeal' has about 15K, it's like trying to start a car rolling in high gear... My guess is that even though the current (amps) will drop, the volts (pressure) will increase... or something. I think....