Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: dwinstonwood on January 10, 2021, 09:23:47 pm
-
I've been slowly accumulating parts that would allow me to build yet another amp. It's hard to decide on one project, so I've been thinking about combining some stuff that interests me.
To begin, my apologies to sluckey for directly ripping off large parts of his AC15 schematic. I should have asked you first. :grin: I'll make my own rendition if I actually build this.
I don't know if this idea will work, or if it's been done already. What I have is the input and power sections of the AC15, and the tone stack, recovery gain stage, and cathodyne PI from the AA964 Princeton. But, because it's a cathodyne PI driving a cathode biased power section it sort of has an early Tweed, or Gibson element to it, too.
Anyway, thanks in advance to anyone who points out serious errors in my idea. Thanks!
-
I've ordered some parts; I've decided to built this thing. It's laid out to fit into a 6G2 chassis (with octal to 9-pin adapter plates, of course).
-
That TMB tonestack is gonna be a heavy load on that pentode. Another guy just tried it. Read this...
https://el34world.com/Forum/index.php?topic=26984.0
-
Thanks sluckey. I'm not set on the TMB.
I'm guessing that a 4 or 6 position coupling cap switch - like your AC15 - would be a better wat to go?
Or, just one tone control with maybe a bright switch?
-
I'm guessing that a 4 or 6 position coupling cap switch - like your AC15 - would be a better wat to go?
Yes
-
OK, I ordered the shorting (make-before-break) P-H392 version:
https://www.tubesandmore.com/products/switch-rotary-2-poles-6-positions-shaft
if my research is correct, I don't need to use the 5M1 R's with the shorting version.
Now to sort out the best way to mount the caps... one end of the caps soldered to a terminal strip located close to the switch, the other ends connected directly onto the switch terminals... or, mount the caps on the board and run a bundle of wires up to the switch...
-
I like to mount all the caps directly on the switch. Then you only have to run two wires to the rest of the circuit. Study this...
http://sluckeyamps.com/VAC15/Vox_AC15_mods.pdf
-
Perfect! Thanks sluckey!
-
Revision 5... :icon_biggrin:
-
I've been reading up on this confusing world of impedance matching. I understand enough to see why a high output impedance tube like the EF86 is a bad match in front of a low input impedance TMB tonestack - that it results in too much signal attenuation.
So, I'm wondering what you all think of a pentode/triode tube like the 6U8A used to insert a cathode follower (the triode section) after the pentode section in front of a TMB? As in: Input > pentode > triode CF > TMB...
I don't know what amps have used this, or how it sounds. Does it retain any of the EF86 characteristics?
Thanks!
-
Waiting on parts delivery to get started. Ignore the 6U8A tangent.
I ordered a 6G2 chassis on the auction site and some octal to 9-pin adapters. Thanks bmccowan for the chassis seller tip in another thread. Great price.
Since I have to order my amp cabinets anyway, I’m thinking about a custom 2-10 with a WGS Green Beret mated to something else...
-
Revision 5... :icon_biggrin:
ground pin 7 of V1 (ef86).
--pete
-
The B+ filtering for the AC15 has me confused. :w2:
Although it looks like a "capacitor-input" as far as the rectifier is concerned, it looks like a "choke-input" in terms of the current load - the OPT primaries and power tube plates are connected after the choke.
So, my question: doesn't this configuration increase the current requirements of the choke, i.e., wouldn't the choke need to be rated for 250ma-300ma or more (or, whatever the power tubes draw)?
I'm just trying to learn about all this. Thanks!
-
It's not a choke input filter. See the cap between the recto and the choke? That makes it a cap input filter. That cap would not be present in a choke input filter.
Yes, the choke will need to carry all the current for the entire amp. But my full circuit AC-15 only draws 100mA. The choke you have listed is fine. Don't know why you think the power tubes draw so much current? My full circuit power tubes draw 92mA and my Dual Lite draws 67mA.
-
Hi!
I did something similar.
https://el34world.com/Forum/index.php?topic=8349.msg94519#msg94519 (https://el34world.com/Forum/index.php?topic=8349.msg94519#msg94519)
Work like a charm!
-
Thanks sluckey. I'm still trying to teach myself how to add up the tube current from a schematic. :think1:
Do you simple add up all of the anode current numbers from the data sheets? Like, 1.6ma for a 12AX7, for example?
(And, yes, I did see it as a capacitor-input circuit; I was only referring to the additional current load on the choke) :icon_biggrin:
Thanks chocopower! Yes, your Gañan II looks very similar. Do you like the amp more with the bypass caps in-circuit, or switched out?
-
Do you like the amp more with the bypass caps in-circuit, or switched out?
You mean the coupling caps on the ef86 or the bypass switch on the 12ax7 cathode?
-
I'm still trying to teach myself how to add up the tube current from a schematic.
It's easy if the schematic has voltages on it. Simply divide the cathode voltage for a tube by the cathode resistor value. This will give you the current flowing through that tube. Repeat for all the other tubes and sum all the individual currents. That's all.
For example, look at my AC-15 Lite. http://sluckeyamps.com/hammond/hammond_ao39.pdf
V1 cathode voltage is 2.1V and the cathode resistance to ground is 2200Ω, so the cathode current will be .95mA. Repeat for V2 (notice that the cathode resistance to ground is 820 plus 47K). Do the same for V3 and V4 (doesn't matter that the tubes share a common cathode resistor). My calculator says the total is 68.86mA. What does yours say?
-
Yep, 68.85ma. I think back at some point in time I learned that (when I built my SET stereo amp last March), and since forgot it. Thanks! My project has the same number and types of preamp and power tubes as your AC-15 Lite, so it should sum up around the same, give or take.
You mean the coupling caps on the ef86 or the bypass switch on the 12ax7 cathode?
chocopower, both I guess!
On a different note, I talked with Dave Allen of Allen Amplification about his custom Heyboer transformers. I ended up going with his PT and OPT instead of the Hammonds.
TP22D PT: 270V-0-270V @ 120ma; 6.3 CT@3A;and 5V@2A (won't use with EZ81)
TO20 OPT: 20W; 6,600 ohms to 8 ohms
-
The switch in 12ax7 cathode... i remove it.. 1k5/22uf work better for me.
The rotary switch for the coupling caps on EF86, you cant switched out. At less, you need ONE coupling cap.
I make it variable to control bass response.
For me, with 3 positions is enough, and in amps recently build, i use a 3 position toogle switch.
-
If you check the topic about mine, you'll see i had problems with missing highs.
Those values you use with the EF86 (plate, screen and cathode), seem to me, are gonna show the same result. Too dark and too "gainy".
Maybe is whst you are looking for and you like it, but if its not the case, try the ones i use, adapting them to you supply voltage.
Valvewizar web have a nice example about how to get the more clean signal from a EF86
Saludos!!
-
Thanks David.
I see that you used 330K plate, 2.2M & .047uF screen, and 2.2K cathode.
I have 220K plate, 1M & .1uF screen, and 2.2K cathode.
So, you're saying that your values will make the amp brighter?
I'm using a six-position rotary pot per sluckey's documents. I have a .01uF before the pot, so one of my settings is with no cap, i.e., .01uF.
I've also been looking at Hoffman's AC30 schematic. I will build the amp according to sluckey's AC15 (the original V-1-5 schematic) and live with it for awhile. Then, I might try Hoffman's .047uF before the PI and his .1uF's before the power tubes. It will be a few paychecks before I buy a cabinet, so the chassis will be open to tinker with for a while. Maybe I will experiment with your EF86 values, too.
But, for now, the amp will be built (and de-bugged) as is.
-
Thanks David.
I see that you used 330K plate, 2.2M & .047uF screen, and 2.2K cathode.
I have 220K plate, 1M & .1uF screen, and 2.2K cathode.
So, you're saying that your values will make the amp brighter?
I tried those values and didn't like it.
If you check the last schematic, you'll see i finish using 100k for plate, 470k for screen and 470ohm for cathode.
In some intermediate fase, i even installed a switch who change those 3 values... after play with the amp for a while, i remove it. Fixed values, for the most clean output.
-
Thanks for that info David.
The 6G2 Princeton chassis is going to work fine. I'll need to enlarge the PT opening and drill four #6 holes to mount the board, but that's about it.
I'm messing around with control panel ideas, too (again, I'll laser engrave the acrylic at work).
I'm leaning towards a head cabinet with Vox cloth. But, a 1-12 with a WGS Green Beret would be nice, too.
-
I have a .01uF before the pot, so one of my settings is with no cap, i.e., .01uF.
Be aware that each of the caps on your 6-pole rotary will interact with the 0.01uF blocking cap to further decrease the blocking capacitance, e.g.:
220pF||0.01uF = 215pF*
0.0047uF||0.01uF = 0.0032uF*
etc
*By that I mean the caps are in series, but its my shorthand math for the sum of the inverses, inversed https://en.wikipedia.org/wiki/Parallel_(operator)
THis won't affect the small caps much, but you'll get a higher (i.e. even more subtle difference in) rolloff point than you might expect on the bigger caps.
To minimise the reduction, you could change the 0.01 to 0.1uF (or 0.047uF).
-
...To minimise the reduction, you could change the 0.01 to 0.1uF (or 0.047uF).
Thanks tubeswell! That's really good to know. I have some .1uF's and an .047uF, so I can experiment.
It's hard for me to keep up with the series vs. parallel rules for caps and resistors. :icon_biggrin:
-
I like the range of tones I get with my AC-15. Well, most of them. I don't care for the two pF cap positions. My favorite position is no cap, ie, only the .01 coupler. Ed Chambley used a 12 position switch for even more subtle steps.
But consider that the original AC-15 used only one cap with a shorting switch. You go from full Vox tone to ice pick! The 6 position switch is a definite improvement. :thumbsup:
-
Yeah Steve, this is the one amp project I can't wait to hear. I love the Vox sound. Thanks for letting me pilfer your schematic! Now, I think I want to go with a combo cab so I can play it through a British-type speaker. Guitar Cabinets Direct sells a 12" 6G2 cabinet for $260 - a great price. One Celestion alnico blue is actually more at $279!
-
It's hard for me to keep up with the series vs. parallel rules
my experience in "public education", having the info stored in your brain was best practice
my Navy education best practice was know where the data is when you need it. So a test included, notes, books.....just NOT the human next to you
-
I read somewhere that Einstein never memorized phone number, addresses, etc. Apparently, he said something similar to you, like, "I don't clutter my brain remembering stuff that I can just look up." And, then there are all of the online calculators... I like Digi-Key's the best:
https://www.digikey.com/en/resources/online-conversion-calculators
-
I've been reading up on this confusing world of impedance matching. I understand enough to see why a high output impedance tube like the EF86 is a bad match in front of a low input impedance TMB tonestack - that it results in too much signal attenuation. ...
"Matching" is a term fraught with problems.
We "match impedance" when we seek to achieve maximum power transfer. Or when we're connecting a speaker load to a tube, and the transformer winding-ratio is chosen to give the tube a desired loading when a specific speaker-impedance is attached. Regardless, here we're looking for Source Impedance = Load Impedance.
Preamp stages are mostly Voltage Amplifiers. We don't want maximum power transfer, we want maximum voltage transfer from stage to stage. To get this, we want Source Impedance << Load Impedance, and ideally a load 10x (or more) the source impedance.
Pentodes have a high plate resistance; that's why their curves tip-over horizontal. To get good gain, they have high-Ω plate resistors (though still less than their internal plate resistance). Attach a lossy tone circuit that is few-kΩ to ground, and the pentode loses the ability to throw an output voltage.
-
...we want maximum voltage transfer from stage to stage. To get this, we want Source Impedance << Load Impedance, and ideally a load 10x (or more) the source impedance.
Pentodes have a high plate resistance; that's why their curve tip-over horizontal. To get good gain, they have high-Ω plate resistors (though still less than their internal plate resistance). Attach a lossy tone circuit that is few-kΩ to ground, and the pentode loses the ability to throw an output voltage.
Thanks HotBluePlates. As you say, thinking about voltage is the key to understanding this stuff. Thanks for clearing up my misuse of "matching" when thinking about preamps as opposed to speakers and power.
I found a website that simplifies it to the point that I can start to grasp it. Here's a quote that makes sense to me:
"The output voltage from the source is developed across the input impedance of the destination (often called the load impedance, or simply the load), and therefore the signal voltage is passed from source to destination."
https://www.soundonsound.com/techniques/understanding-impedance
-
I'm still waiting on the transformers; all the other parts on on hand.
Since this "experimental" project is using the cathodyne phase inverter from the Princeton, I came up with a name for it... :icon_biggrin:
-
The Allen/Heyboer iron arrived. It turns out the PT is a perfect fit in the chassis, no drilling or filing needed.
I will need to juggle the choke and OPT placement around a bit, and drill some mounting holes, but that shouldn't be too hard.
-
Before I start drilling holes, is there anything terribly wrong with this setup?
My reasons are: this OPT sits lower than the choke and is less likely to interfere with a speaker basket; the OPT leads line up well with the holes intended for them; and, this is where the original 6G2 OPT mounting holes are located.
My concern: is the choke too far from the cap can? I will route the choke leads under the circuit board away from the pots and front of the amp. They won't cross or be near any other wires.
Thanks!
-
you should rotate the OT 90 degrees. with lay down PT's is hard to determine the bell ends - in this case they face East/West
so your OT Bell ends should face North/South. Hope I make sense. Look at pictures of Fender amps to see how the iron is oriented
-
Thanks Glenn! Well, I actually was looking at photos of original Fender Princeton amps. :icon_biggrin: Of course, mine isn't a Princeton, but it's a Princeton chassis, and I'd like to use the original OPT mounting configuration.
- edited to add both AA964 and 6G2 pics -
-
Lol... well the proof is in the pictures. I stand corrected.... good luck with the build
-
Thanks Glenn. I've read up on a technique where folks connect wall AC to the PT primary and connect headphones to the OPT secondary. Then, they slide the iron around on the chassis to find the least amount of hum.
I don't think I'll go that far. But, if there is a lot of hum (moderate hum doesn't bother me) I guess I could move the choke around after the amp is built to see if it helps.
David
-
I'm making progress - at my usual snail's pace. :icon_biggrin:
Some of the outside turrets along the long edges have only had parts "tack" soldered since they will have tube and pot wires going into them. They look like poor solder joints right now, but this is easier for me. I'll go over every connection one final time.
I left enough free space around the .01uF caps (smallest yellow ones) so that I can go up to .1uF in size later, if I want to.
-
Looks to me like the lay down PT end bells are oriented “up/down”, neither east/west or north/south.
-
Looks to me like the lay down PT end bells are oriented “up/down”, neither east/west or north/south.
Yes, that's how I see them, too. Thanks.
Next thing to do is wire up the filament connections. I'm going to put them flat down onto the chassis, and pushed up against the back edge.
-
Ready for first power up with dim bulb, no tubes.
-
With only the EZ81 rectifier in:
Node A: 419.7VDC
Inner EL84 pin 7 plate: 419.5VDC
Inner EL84 pin 9 screen: 419.0VDC
Outer EL84 pin 7 plate: 419.4VDC
Outer EL84 pin 9 screen: 419.1VDC
EF86 pin 6 plate: 403.7VDC
12AX7 pin 1 plate: 409.2VDC
12AX7 pin 6 plate: 414.1VDC
Those voltages seem high, even though I know they'll drop with the other tubes installed.
Should I proceed by putting in the other tubes and hooking up a speaker? I don't want to burn up my EL84s. :icon_biggrin:
Thanks!
-
Should I proceed by putting in the other tubes and hooking up a speaker?
Yes. Use the dim bulb for a few minutes. If no problems then plug straight into the wall. Now you can test for sound and measure voltages.
-
Thanks Steve! I'm down to one 200w incandescent bulb. Will that be ok, or should I run to the store and get a 40w?
Thanks.
-
Why would you use a 200W bulb to check a 15W amp? You been watching Uncle Doug again? Get a 40W appliance bulb. You may even want a 25W appliance bulb while they are still being sold.
-
...You been watching Uncle Doug again?...
:laugh: yep
Cool, I need to go to Kroger anyway.
-
No pyrotechnics with the 40w appliance bulb. But, the tubes barely glowed.
Plugged straight into the wall the EL84 plates settled down to 316.6VDC and 316.8VDC after about five minutes.
I'll measure the little tubes in a while, after I relax. I still get nervous with these first power ups - I'm working in a townhouse with attached units. I don't want to start a fire! :icon_biggrin:
Anyway, those two readings look good to me. We'll find out if I might need to adjust the two B+ dropping resistors for the first two tubes...
Thanks again Steve!
-
No pyrotechnics with the 40w appliance bulb. But, the tubes barely glowed.
And that's what I want to see. I bet the bulb was glowing dimly too. At this point of testing I'm not interested in voltage readings or even if the amp can make a noise. Once I see this, I'm ready to plug into the wall for some excitement!
-
Yes, the bulb was glowing, but not too bright.
Plugged straight into the wall, I measured the voltage across the shared 150 Ohm cathode resistor at 12.4V. The bias calculator spit out this data:
Total Cathode Current = 82.67 mA
Total Plate Current = 78.12 mA (Cathode Current minus approximate Screen Current of 5.5%)
Plate Current per Tube = 39.06 mA
Plate Dissipation per Tube = 12.34 Watts
Plate Dissipation per Tube % = 102.86 %
They're revving pretty good. :icon_biggrin: I'll plug in a guitar.
-
I had to solder up a longer speaker cable. :icon_biggrin:
It sounds really good! There's a big range in tone with the switch. From super bright to pretty warm. I need to look at the switch and write down what capacitance each one is.
It's definitely a very different sound than my 6G3 Deluxe with 6V6s. It's too soon to give a good description, but maybe more raw and bright for sure. It's what I was hoping for. It's why I built it, for the AC15 sound.
But, there are a couple of issues. I'm getting loud pops when I turn the tone switch. Not so with the volume turned down. I'm sure I bought a make-before-break switch. Do I still need to add resistors?
Also, some tone setting are very quiet and others create some hiss. Maybe that's just the nature of the smaller caps letting it through?
I'm calling this project a success! Here are the schematic and layout files.
-
I have a small pop on both of my amps. Not really objectional. Might be in church. Resistors will help. Maybe look closely at your switch while slowly operating it. Can you see the wiper blade making before breaking?
-
I'll take a look. I also realized that I didn't really ground the switch. I wasn't sure exactly how ground it. But, it is making contact with the chassis.
-
No reason to ground the switch.
-
Thanks. I'll give some 5M1 1/4W resistors a try. From what I've read the pop is from the caps discharging (which didn't occur to me), so the bleeder resistors make sense. It looks like Matchless uses them, too.
I thought I'd swap in a 160 Ohm cathode resistor for fun; that should drop my plate dissipation down around 95%-ish. And, since I've never bought a single NOS tube before, this amp might be a good excuse to try them. Some of the NOS Russian 6n14n's are relatively inexpensive.
-
AES has some Amperex Bugle Boys for $150 each. I like the $11 JJs just fine. :icon_biggrin:
-
Just for the record, the 9-pin tube voltages:
EF86 pin 6 plate: 87.1VDC
12AX7 pin 1 plate: 151.5VDC
12AX7 pin 6 plate: 203.1VDC
I'll go ahead and add the voltages to the schematic for my future reference.
-
AES has some Amperex Bugle Boys for $150 each. I like the $11 JJs just fine. :icon_biggrin:
Yikes! That's eccentric rock star territory. $280 on an Alnico Blue or Gold would be money better spent, IMO.
The Reflektor 6n14n's I was looking into are $22 at KCA. But, yeah, I've bought JJ's for my other amps and I think they sound great.
-
AES has some Amperex Bugle Boys for $150 each.
$190. But they're out of stock. They do have ribbed-plate Telefunkens for $300/ea if you really need to buy-it-now. :l2:
Fortunately, I've gotten the old-tube hookup a few times in the recent past, and might be set for life. Sometimes it really helps to be able to buy in bulk when you stumble into a good opportunity.
Just for the record, the 9-pin tube voltages:
EF86 pin 6 plate: 87.1VDC
...
EF86 screen voltage matters. A Lot.
If you get into a mood to tinker, or don't feel you have enough grit & grind from the EF86, try using a 3MΩ pot from B+ to ground with the wiper to the screen (probably with the 0.1µF bypass cap from wiper to ground). One end of the pot will clean up the EF86, the other will make it dirtier.
-
Thank you HotBluePlates!
So, is this what you mean? Or, did I understand it all wrong? BTW, your suggestion sheds some light on the AC4's use of a 5M6 screen resistor, which I never understood until now.
What special precautions would need to be taken wiring up a pot with 226VDC B+ feeding into it?
Thanks.
-
I think this project is finished. I changed the 150 Ohm cathode resistor to a 160, and added 4.7M resistors to the tone selector.
It no longer pops when turning the tone switch, and my EL84's are running at 11.78 watts - 98% dissipation. :icon_biggrin: I'm happy with all of the voltages, and the Allen iron.
Plate Inner EL84 = 315.8VDC
Plate Outer EL84 = 315.7VDC
160 Ohm Cathode Resistor = 12.62VDC
Total Cathode Current = 78.88 mA
Total Plate Current = 74.54 mA (Cathode Current minus approximate Screen Current of 5.5%)
Plate Current per Tube = 37.27 mA
Plate Dissipation per Tube = 11.78 Watts
Plate Dissipation per Tube % = 98.14 %
It sounds really good, with a wide range of tone settings. Thanks sluckey for recommending the selector switch!
I updated the schematic and layout, if anyone's interested. I included an option for a grid stopper for the PI (I didn't add it), since there seems to be a general dislike for Cathodyne phase inverter distortion. I won't really know what I think until I get the amp somewhere where I can crank it into distortion. Maybe this weekend.
Thanks everyone for all the help!
-
So, is this what you mean? ...
Yes, at least as a setup for testing.
What special precautions would need to be taken wiring up a pot with 226VDC B+ feeding into it?
Use an average large-size pot found in amplifier volume controls. If it were going to be a permanent arrangement, we'd think harder about it. But my suggestion was only to enable you to listen & fine-tube the pentode.
-
Thanks HotBluePlates. After I posted my reply, I suspected that it was meant as temporary means of determining the best resistor value. One thing I wasn't clear on was which way it changes things: does a higher value (above 1M) make the amp cleaner or dirtier? I'll try to read up on it. I understand the purpose of the screen, I'm just not clear yet on how it's voltage affects the clean/dirty sound.
Thanks!
-
A knock at the door... the cabinet arrived.
I went with a 10" Celestion Creamback. I'm still waiting on the faceplate material; here's the design I drew up. I'll laser etch it at work.
It's almost done.
-
Well, I just got the faceplate and backplate engraved. So, this project is finally "complete."
The next time I build an amp I'll make the faceplates early on, and attach them at the beginning... I had to de-solder the fuse holder and fiddle with that troublesome Fender pilot light bracket. But, no major deal.
I'm still messing around with the EF86 voltages. Right now I have a 330K plate resistor and a 2.2M screen resistor (DC30-ish). I might end up going back to the Vox 220K and 1M resistors. :dontknow:
This is a very light amp. The Celestion Creamback magnet is only 14ozs., and the GCD cabinet is lightweight pine. The amp is also compact (6G2 build format), but plenty loud for me!
-
I've been really happy with the tone(s) of this amp. But, I've always had a louder than preferable hum with the volume up past halfway. This didn't change whether a cable was plugged in or not.
For some reason I built the amp with one four-section cap can - 20/20/10/10uF - to do all of the filtering, using one ground. I wondered if that was the source of the hum. So, today I split the filtering in two: an F&T 16/16uF can for the power section, and a separate 8/8uF dual-cap for the preamp section. Well, that got rid of the hum. I forgot to take pictures, but here's a layout showing the new filtering with separate grounding points.
I was glad that after thinking about the problem for awhile I was able to finally see what was causing it. I know the single cap can with one ground was a mistake none of you all would have made! But, I learned something the hard way in the process. :icon_biggrin: