Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Platefire on July 20, 2018, 11:30:30 pm
-
For instance, I'm currently re-biased my DIY AB763 DR with JJ 6V6S tubes. The tubes are running about 3 to 4 mA apart depending on the setting.
So my approach has been in going for 70% dispensation Calculated:
Watts/Plate Voltage X .70
14/418=.0334928 or 34mA x .7 dispensation = 23.8 mA bias setting
So I was previously running the lower one at 22mA and the higher 24.8 mA to obtain a balance between the two. So I was running the high side slightly over my 23.8 destination to bring the low side up closer to the destination setting.
I got concerned that it was a little to hot so I reset it to 16.9/20. Now I'm concerned that it my be a little too cool for my taste.
How do you balance your bias in these cases???? I always order a balanced set of power tubes but they are always at least this far off. Platefire
BTW---no red plating was occurring on the higher setting but it was mighty hot to the touch----but sounded awful good!
-
Honestly, I'd call those tubes balanced. But if you must have them exactly balanced at idle, a dual bias pot arrangement is a good solution. Not worth it for a 6V6 amp IMO. Take a look at this simple mod to my big Sunn. Schematic at bottom of the page...
http://sluckeyamps.com/sunn/sunn.htm
Keep in mind that even though you perfectly balance the tubes at idle, they probably will not be perfectly balanced when you start playing.
-
Thanks sluckey!
Looked at your Sun. That's definitely the way to go to get them near perfect as you can get.
I'm not really concerned about a 4mA imbalance. I just have never seen a discussion before on the best way to set the imbalance with a single bias pot. Maybe it's so minute of an issue it don't deserve recognition as a problem but since I was in the process of biasing, thought I would bring it up. Do You:
1-Set the hottest tube at the intended dissipation and let the cooler one fall where it may?
2-Set the hotter one slightly above your intended dissipation to bring the cooler one up closer to the intended dissipation?
I've previously tried to hit the dissipation goal right in the middle of the imbalance but now thinking that the #1 method is probably the best approach. I do like the sound and feel of it set slightly on the hot side of normal without setting it so hot I feel I need to be peering into the back of the amp to make sure no red plating is occurring. Platefire
-
I set the hottest tube to 70% although I agree that setting the cooler tube to 70% has a little better tone and feel.
During the '70 CBS era, Fender switched over to a bias balance circuit rather than the older bias adjust circuit. The one pot bias balance circuit does allow you to set the tubes to the same idle current but you give up the ability to set the overall bias hotter or cooler. The only way to set the tubes to say 70% point ***AND*** also adjust so the tubes are exactly balanced is to use a two pot bias system.
Here's Fender's later bias balance circuit. Most people like to modify it back to the older bias adjust circuit...
https://el34world.com/charts/Schematics/files/Fender/Fender_super_reverb_aa1069.pdf
-
I have a 1974 Pro Reverb that was almost exactly like a aa1069 circuit except mine don't have a middle control.So I haven't figured out exactly what the circuit number it is? I converted it to a aa165 Blackface circuit and did modify the bias balance pot to a regular bias pot. Probably should have tried to use the original bias balance to see how it worked because those old vintage tubes presently in there are way out of balance.
Think I'm going to re-bias the DR to 23.8mA(70%) on the hotter tube and let the cooler one fall where it may.I know it probably don't make a Nat's hair worth of difference but I can feel better about having a decisive method to deal with it now. Thanks, Platefire
-
1-Set the hottest tube at the intended dissipation and let the cooler one fall where it may?
That is how I handle it. After purchasing a "matched" set of Winged C EL34B tubes from Tube Depot that were running about 10ma apart I decided to add a dual bias adjust circuit to my JCM800 2204 clone I was building at the time. Problem solved. Now I can put any tubes in it. I built an SLO 100 clone after that but only put dual bias for each push pull pair.
-
I suspect that the DR AB763 circuit, with its >400V HT and 6k6 p-p load, runs 6V6 rather hard under signal, so lower idle dissipation may be beneficial, in order to compensate.
-
Chances are the 'matched pair' you buy from someone else isn't really matched at all, or is only lazily matched. For my own amps, I buy at least twice as many tubes as I need and mix and match them.
-
+1 to sluckey. Also for guitar amps (as opposed to hi-fi) there's a theory that slightly imbalanced power tubes give good synergy.
-
+1 to sluckey also---he always bothers with my crazy questions. Instead of Jim Dandy to the rescue it's sluckey to the rescue:>) I'm not so concerned about the imbalance of so called matched pairs and like it has been said some prefer the sound of imbalanced tubes. I would think that better balanced tube lends itself to a cleaner headroom/purer clean tone and imbalanced would lend itself more to coloration and quicker power tube breakup.I can see the beauty of a bias pot per tube because of the added flexibility to mix all kinds of brands and different rated tubes. Haven't tried that myself yet.
The main point of this thread for me was just see how you folks with only a single bias pot deal with the imbalance. Every time I bias a amp, I have this question come up in my reasoning should I?:
a-Set the bias in the middle of the imbalance and let the hot one run a little too hot that would at the same time bring the cooler one up closer to 70% Dis
b-Set the hot burner to the desired dissipation and let the cooler fall where it may
c-The real risky one is set the cooler one to 70% and let the hotter one fall where it may--long as it wasn't red plating.
I had never seen this discussed much to any degree, so I thought I would bring it up. Previously I was doing the "a" method. Now, I finally decided for myself I chose "b". So this discussion has helped me decide what procedure I would do so I would have a thought out plan of action next time. I don't like running my amps so hot that I feel the need to keep checking the back of the amp to see if it's red plating. So "b" will give me more of a warm and fuzzy. Even at that I will find myself peering in the back, however now! maybe not as often:>) Platefire
-
How about both types in the same amp, bias adjust and balance, is that possible?
Here one interesting video about bias balance:
I have discovered, through experience, that even a few mA difference in cathode currents will raise the temperature of the tube significantly -
Second the video the balance effects the tubes temperature too.
-
Note that bias balance is harder than it might seem. Component values in the circuit are nominal, and drift over time. Each side of the OT windings are a bit different. Bias varies as tubes age and how warmed up they are. How far are you willing to go?
You can use a bias balance pot. Or KOC uses a signal pot at the grid leak resistors to vary signal to ea tube to compensate for bias differences.
-
Can't help but wonder at the old Fender factory in Fullerton Ca., did they just come up with the correct circuit components to achieve a good bias on a certain circuit they had predetermined in the original design of the prototype and on final set up on the assembly line just set the negative bias voltage to a prescribed setting they had on the schematic? or did they double check for an actual dissipation level that they were shooting for.
My guess would be they just adjusted the prescribed negative voltage in a cookie cutter manner for a certain circuit a left it that. Anybody know? Platefire
-
My guess would be they just adjusted the prescribed negative voltage in a cookie cutter manner for a certain circuit a left it that
That's my guess too. No one was obsessed with bias current in the repair shops. Just set the negative voltage to the schematic voltage. Change tubes if they are red or dead. Some shops would replace the output tubes in pairs if one was faulty. Others would only replace the faulty tube.
-
My guess would be they just adjusted the prescribed negative voltage in a cookie cutter manner for a certain circuit a left it that
That's my guess too. No one was obsessed with bias current in the repair shops. Just set the negative voltage to the schematic voltage. Change tubes if they are red or dead. Some shops would replace the output tubes in pairs if one was faulty. Others would only replace the faulty tube.
It was very commonplace to replace only the bad tube back in the tube days, but I thought it still was. I must be doing something wrong. I use old tubes like RCA Blackplates. Since I can buy 10 Unmatched tubes for the price of a "Strong Matched Set" from an Ebay seller, :l2: Date codes match and I tested them a 5 volts on my TV77. These are NIB, but I don't have the boxes.
There is a semi annual huge Hamfest in Arden NC that is part of the state fair. Now I am 55 years old, but they treat me like the kid since I have known a lot of them since they were 55. Last time I was there I bought 5 RCA 12Ax7 Gray in new boxes for $10 and the guy said hell I am 77, I will never use a lot of these tubes. Then he handed me bag of old tubes. Low and behold, there were 2 big boxes and contained GE 6550A. They looked very new so I plugged them into my Leslie 147. They sound great, no red plate. I have no idea how matched they are, but they are not redplating.
Now that is a deal at twice the price.
-
JJ 6V6S is supposed to be tough, so I'd be comfortable running them hotter than other tubes. I would bring the cold tube to the target and let the hot one exceed it.
This is personal preference, like biasing always is. Running them hotter might reduce tube life or risk failures, but it's your choice to accept that risk (or not).
Can you afford replacement tubes? Do you have spares, or would a failure put the amp out of service until you get new ones? Is downtime acceptable or is the amp used for gigs every week?
If you're not worried about those questions, and the amp sounds better biased hot, bias it hot. Or in this case bias one of the tubes hot.
-
I think back to my first tube amp, a Silvertone 1482 purchased used from a local Keyboard player. I played the dickens out of that amp by a record player with a lot of Ventures albums on. A lot of cases of switching amp on and off real regular. The few years I had it, it never blinked, always worked, never one problem that I recall on the original set of tubes. I never even pulled a tube out of it ever. It amazes me now how well I fared without an inkling of tube amp or electronic knowledge. Next amp a Gretsch Variety 2-12. Same thing, endless use, no problems. Next in line a Silverface Bandmaster, same deal, blew some speakers but nothing on the amp. Never got into any amp maintenance problems until I had my Peavey Years:>) Hybred & SS amps. Came back to full tube amps with one of the first USA made Fender Hot Rod Deluxe in 1997 50th anniversary model witch was a trouble free amp until I sold it. After that I got into tube amp building, now I'm squabbling about the bias. :BangHead: Platefire
-
How about both types in the same amp, bias adjust and balance, is that possible?
It is possible. KOC talked about it in TUT2 I believe, but he has later said that two raw adjust pots are better than one raw adjust and one balance. I've done two raw adjust bias pots in quite a few amps and prefer it over a single adjust. YMMV.
Greg
-
How about both types in the same amp, bias adjust and balance, is that possible?
It is possible. KOC talked about it in TUT2 I believe, but he has later said that two raw adjust pots are better than one raw adjust and one balance. I've done two raw adjust bias pots in quite a few amps and prefer it over a single adjust. YMMV.
Greg
I'm unfamiliar with KOC and TUT2 :w2:
When you mentioned both together so I did some digging and found this interesting layout from Rob Robinette, see the picture!
How about that?
-
KOC and TUT2
KOC is Kevion O'Connor of London Power. (That's London, Ontario). He's a major amp guru. Definitely checkout his website.
He wrote a series of about 6 volumes Called The Ultimate Tone, or TUT for short, called TUT1, TUT2, etc. for short. I have TUT1. I found it to be difficult reading, but worth the effort.
-
While we are talking bias, something I wanted to bring up for discussion. I have been installing Mar Vac 10K lockable pots through the chassis to be adjusted and locked down from the exterior of chassis without removing the chassis. Also using the female Banana Plugs for external check points for each tube.
Problem is to adjust the bias properly, you need to know what your plate voltage is running. Kind of defeating your purpose with all these external test jacks and bias adjustment and no known plate voltage without pulling the chassis.
I have thought about running an external test point to the plate pin for a plate voltage reference just like I do on the cathode check points, but then I think of the jacks shorting out and the major catastrophe following.
My questions is, is there any safe easy way to get your plate voltage under a load externally without having to pull your chassis??
The only thing I have come up with is, I do have a Military Socket Adapter Kit MX-1258/U that has multi socket adapters including an octal that has external metal contacts connected to the pins opposite the pins where the tube plugs in. So I could pull one tube, install the adapter, install my tube in the adapter, fire up the amp and when warmed up touch my MM red to the Pin #3 contact and get my reading. That would be rather awkward but possible.
Any other better ideas??? Platefire
-
.
-
Thanks, PRR
So being reduced by the 1Meg makes the voltage much less lethal and you would measure mV and multiply the reading by 1K and that would be your plate voltage? So if my plate voltage was 417VDC my mV reading would be .417 on my meter? I think I like this!
I also suppose it would be no need to put one on each tube, one tube would be close enough? Does the 1Meg need to be several watts? Platefire
-
Use Ohm's law and the Power Formula. Your dropping V plate volts across 1.1M of R. That gives you the current flow (I) through PRR's voltage divider. Use the Power Formula, W = V x I, to determine the Watts dissipated across the entire voltage divider. Divide that for the % for ea resistor.
-
Those resistors would need to be precision resistors so you would be getting true reading. Platefire
-
PRR's circuit is good enough. If you want to go farther: Voltage divider formula is R1/(R1 + R2); 1,000,000 / 1,001,000 = .9990 Not .1!!!
So for ultimate precision you need a "9" value R in there. The math is giving me a headache; and I have some pressing work to do.
Make sure to at least quadruple the wattage ratings so the the R's don't get warm and secretly change value!
Also I use female test pin jacks (rather than banana jacks) for these purposes. Doug sells them.
-
well seems to me it would require a 10 watt, 1M, 1% resistor probably the size of a hand grenade that don't even exist as far as I know. So it's starting to appear to me not feasible. Platefire
-
500 volts across a 1Meg resistor is 0.25W. Double that for safety. A half watt resistor will be fine.
-
the size of a hand grenade
take Steve's recommendations, but if you need some granade sized low ohmers I have 3-4 dale .1 & .2 ohmer 100w that are grenade sized :icon_biggrin:
-
OK Then! I got two NOS solid color 1% 1M resistors as shown in pix. Trying to determine what watt rating they are? Can't find anything on the net on them. Even though they are different sizes, they both say pretty much the same thing as follows: 1M, 1RC, MDF 1%, RN575B, 1004F, 6005
On my Fluke read 1.002M
Can you tell anything from the codes? Thanks, Platefire
-
You do not need precision resistors until you get precision-destruction tubes that die at 30.01 Watts.
You take a 30% slop factor out front ("the 70% rule"). Who cares if it is 65% or 75%?
Which also makes the 1000/1001 "error" quite meaningless.
Actually you "should" observe the Voltage Rating on the resistor. It may be 350V. This suggests two 470K in series (250V each). This changes the bottom resistor to 940r. 1K||15K is more than close enough for jazz or polka. Power in each 470K is just over 1/8W, 1/2W is ample. The ~~1K heat is negligible. All simple math; just do it.
-
I still think 1Meg (1/2W) and 1K ought to be good enough for most.
-
Who cares if it is 65% or 75%?
That's why tubes are great, in my old world .2parts per million was spec :think1:
-
Thanks Very Much PRR
I did a little more research on my pictured Resistor. Turns out it's Military Spec MIL-R-10509A.The 1% tolerance spec is listed in the spec dated Sept 1958
https://33audio.com/enter/data/M10509A.pdf (https://33audio.com/enter/data/M10509A.pdf)
The large resistor is a RN75B and the smaller one is RN70B and the letter codes are as:
R = resistor
N = metal film
75 = 2W @ 70℃ / 1W @ 125℃ (500V Rating)
70 = 0.75W @ 70℃ / 0.5W @ 125℃ (350V Rating)
Also the RN75B is rated at 500V if I'm reading the spec right. I do have other .5 watt 1M's also I can use. I think my 1K is a 1 watter metal film.
So I got everything I need to install, just taking the time to do it. If it works out like I think it will, I've got several other amps I will be installing this on. Platefire
-
I'm unfamiliar with KOC and TUT2 :w2:
When you mentioned both together so I did some digging and found this interesting layout from Rob Robinette, see the picture!
How about that?
Sorry, I made the assumption that most people know who Kevin O' Conner (KOC) is these days. As was said, he is a major amp guru and has a bunch of books out. The books aren't cheap but there is lots of useful info in them. They are sort of hard reading too....he doesn't seem to write things that are as easy to read as Merlin. He came up with the term and circuit for 'power scaling' at least as it relates to guitar amps and a lot of makers have incorporated that into their products also.
Greg
-
I would memorize plate voltage or write it on the amp. Tubes are changed regularly and need re-biasing to reach target current, but plate voltage and target current don't change with "normal" operation and maintenance. At least not in a way worth worrying about-- wall outlet voltage can affect it +/- a few percent but this changes between venues or between your home and a venue, and most people don't re-bias every time the amp is moved.
Voltage sags with changing current, but you're usually biasing new tubes to draw the same current as the old, so once you hit that target, you can assume voltage is the same as before.
If you plan to swap very different tubes (like 6L6 and 6V6) you want to check B+ because it will change lots with a halving/doubling of current.
I found this 1%, 1M, 3 watt resistor rated for 750v: https://www.mouser.com/ProductDetail/Vishay-BC-Components/PR03000201004FAC00?qs=sGAEpiMZZMtlubZbdhIBIEEdsE%252bzDFm9xwpKNsJJfwI%3d
Mouser sells lots of cheap 3W resistors rated for 600v or higher, this only gets tricky/expensive when you look for 1% tolerance. But it's still only $0.95.
PRR's circuit is good enough. If you want to go farther: Voltage divider formula is R1/(R1 + R2); 1,000,000 / 1,001,000 = .9990 Not .1!!!
So for ultimate precision you need a "9" value R in there. The math is giving me a headache; and I have some pressing work to do.
If you wanted to correct this error the bottom (1k) resistor being larger is easiest. How much larger? If you had 1001ohm instead of 1000, you get .999999. 1002 ohm overshoots to 1.000997. That really shows how small the error is, and impractical to fix. You could also reduce the 1M to 999000ohm, but combining that specific value with the high voltage/power rating makes it even harder to get.
Now, this would matter if you were dividing the voltage by 10, you'd get 1/11 or 91% of the reading you're looking for. But when the divider ratio shrinks, so does that sort of error, in this case below component variation even if we buy very good resistors, and way below what we care about.
-
92Volts
I unusually have a file folder on every amp I built and record updates including new tubes, voltage and the bias setting arrived at. So yeah, no problem on finding recorded plate voltages on last maintenance.
I also realize there is probably not going to be much drift as the tube ages and wall voltages change as the electrical grid get taxed or not. So yeah, most likely you would be good biasing to the last recorded plate voltage which is the method I have been doing.
However not knowing for sure is not knowing for sure. You know what that does to us humans! Plus this is something that's been on my mind every since I've been installing external bias test points and externally adjusted bias pots. To me it just seems logical to go to the trouble of installing external bias points and adjustment, that there should be an external reference to plate voltage. This would give you an inkling if the previous voltages are still holding true to satisfy any wondering.
I only re-bias when I install new tubes or I'm already in the amps to do maintenance unrelated to bias. Then I only do a check to see if it's still holding same as last recording because I'm already in the amp and plate voltage is handy. So no I'm not biasing every playing event or re-checking wall AC to re-adjust accordingly.
I guess this is just a personal thing because it seems nobody else in really interested in this external reference.
Who knows, after I pursue it a bit, I might decide it's totally worthless. I'm the kind of fellow sometimes I have to pursue something in order to determine it's not worth pursuing:>)
So I'm thankful PRR shared his knowledge and put this little circuit out there. Even though I must admit I don't understand the math 317VDC plate voltage through a 1,000,000 Ohm resistor or a thousand thousands, then the reading multiplied by 1000 for a true voltage reading. Then I don't understand the 1000K to ground role in this? So I hate to plead ignorance but it would be great to understand what's happening here. I realize the 1M is reducing the voltage to a much smaller less dangerous safe testing amount. So I throw myself at the mercy of the Court, that so far I'm going by faith that PRR knows what he's doing.
Thanks 92Volts for the reference to the 700V 1 % resistor, that's a lot more hedge than 500V one I got. If I carry this out, I might very well order some of those. Platefire
-
... I don't understand the math 317VDC plate voltage through a 1,000,000 Ohm resistor or a thousand thousands, then the reading multiplied by 1000 for a true voltage reading. Then I don't understand the 1000K to ground role in this? ...
1M + 1k = 1,001,000R (which is the total resistance between B+ and ground).
As you say, the 1M/1k reduces the B+ down by a factor of one thousand:
1,000R/1,001,000R x 371 = 0.37V (which is what you would see at the 1M/1k junction if you had 371V at the input)
0.37V/1000R = 0.00037A or 0.37mA (which is the current through 1k. As the two resistors in the voltage divider are in series, the same 0.37mA will be seen through 1M, hence: 370.63*/1,000,000 = 0.00037A)
*371 - 0.37
0.00037A x 0.37V = 0.0001W (a 0.25W resistor would be more than adequate for 1k.)
0.00037A x 370.63V = 0.14W (a 0.25W resistor would be marginal, but a 0.5W resistor would be adequate for 1M)
Is that what you were looking for?
-
Thanks very much Tubeswell!
I messed up stating my plate voltage was 317 when actually it's 417 but that don't make any difference in that the math principle is still the same. I got my calculator out and been going over it. I'll be trying to wrap my pea picking brain around it. I'm glad you laid it all out in detail so I can refer back to it when needed. Platefire
-
Some of the best Vintage Fenders I ever heard came into my shop with power tubes 10ma apart or more
I would not get too wrapped up in all that
The power tubes I buy have been burned in for 24 hours
I pay for platinum matching to avoid customers emailing me about 4 maa difference :)
-
> I don't understand the 1000K to ground role
1k to ground.
Take 100k and 100k across 200V. The voltage at the junction is half. (If not, re-do hard math including meter-loading, about 1% for this case and a 10Meg DMM.)
Take 1000k and 1k. The voltage at the junction is about one-thousandths of what you put in. (Yes, 1/1001, which is too close to care; and the 1k will not meter-load enuff to see.)
As the final trick: most DMMs read low-Volts as "mV". So your 415V reads as 415mV. You do have to mentally mask-off the "m", but you do not have to shove the decimal point around. (Otherwise, a 100:1 divider would be fine. But turning "4.15V" into 415V in your mind requires more concentration than I sometimes have, when working with hot glass and high voltage.)
The 1Meg means that voltages as high as 1,000V (well above anything with handles has inside) can give at-most 1mA of current. Outside an open-heart hospital, the lowest "safe" current in humans is usually taken as 5mA; 50mA is where we really need to be careful. So the 1Meg is intended to make this quite non-lethal. It is additionally safe because the 1K leaks more than your body can, and limits 1000V to just 1V; most codes get strict above 25V.
It does depend on miraculous modern DMMs with high-high sensitivity and good precision at low voltage. But they sell now for $10, or free with a $50 buy at Harbor Fright.
-
Think I got it! (https://wikimedia.org/api/rest_v1/media/math/render/svg/5d55415f24b63635bc017c3287b406c480a54472)
(https://wikimedia.org/api/rest_v1/media/math/render/svg/5d55415f24b63635bc017c3287b406c480a54472)
R1 and R2 Combined is 1,001,000 divided into R2 1000K=.0000999
Plate Voltage is 417 X .0000999=.0416583 or .0417 (This would be the MM reading in mV)
Since mV is One Thousands of a Volt we multiply by 1000 x .0417= 417VDC
The Voltage Divider Circuit pictured below was the pattern with the correct resistor values added in PRR's schematic. I did a little studying, so I hope I'm on the right track:>) Platefire
-
> divided into R2 1000K
R2 is 1K, not 1000K!!
You are just being lexdysic.
-
Yes, I'm that and a whole lot more! :icon_biggrin: !skanht
-
In Answer to Doug's post, I've fully come to terms that the
imbalance between push/pull power tubes. So I'm not fighting
a slight imbalance. So that's a given, I can be happy with that
What I've tried to explain what this thread is about is in
most cases an imbalance of a certain amount is acceptable
and may be even preferred depending on the individual.
My question is centered around adjusting the bias to best
deal with the known imbalance. So everybody seems to be thinking Iwant to obtain a perfect balance---no---just best way to adjust the imbalance.
Whatever it might be-4mA or 10mA difference.
My final decision after this discussion is to adjust the high
side to approximately 70% and let the low side fall where it may.
That's my story and I'm sticking to it😂
-
Can't leave it alone can you? What do you keep editing?
-
Sorry! Guess sometime I get carried away. Always appreciate your help. platefire
-
Just had to ask. I know you've edited that post twice today and I have no clue what you changed. Is it just grammatical corrections? Changing content? What? I can't tell that anything in the message changed?
If there's still a question, just ask. When you edit and re-edit at least let us know why. I follow your messages because I'm interested in you and what you are doing.
-
Well I was just trying to clairify my language. It seems to me
that most that read this post, their comments are centered around
getting a perfectly balanced set of tubes but my interest was in
how others tilt their bias to work around the given Imbalance using
a single bias pot. I have always been kind undecided how to manage
that. So I felt I needed to keep directing the conversation back the
original question. So I may have failed in my efforts to do that?
It probably didn't help running a rabbit trail with the voltage
divider about getting a plate voltage reference.
So to me the results are very much a success in that I came to a
conclusion on how to adjust around the Imbalance plus I learned
how to build a safe plate voltage test point that I've been wanting
to do for a while, but just didn't know how. So ye Haw! Platefire
-
I think we all missed your point. I was the first to jump down the rabbit hole. :wink:
-
Sorry Platefire, I missed the point too. I was wondering since you aren't a rookie, why you were concerend. Now I understand. Truthfully, on a guitar amp I am more likely to adjust signal level from the phase inverter to achieve the tube tone I am looking for. We cannot know it that LTPI with that "balanced triode" will do it. Also, transformer winding not balanced generally.
I have left the balance pot on a silver fender and simply bias with a resistor. Completely fixed. I did copy a Hifi Macintosh bias balance circit, but there again, I have not use it on a Guitar amp. Hifi generally doesn't run tubes as hard.
I just received a 4 point auto bias board same as used in some newer Hifi gear. It works well, but before I installed I checked signal level and current from the PI, 6sn7 and it had drifted, so I had to reduce resistance on one side a smideg.
I did use a Dual Bias pot on a build and I am sort of split in this. At IDLE you can match well for as long as the tubes will hold. The main problem is we generally do not play at idle. I like analogue needle Simpson 260 meters, because they really are cool, really! Anyway, everytime I have dummyloaded and increased the signal gain I can measure i get imbalance, sometimes one tube will pass more current at half signal and the other will at full signal. Swap places.
The best new production tubes I have bought are from Upscale Audio. The guy that owns the place is Kevin Deal and he is so straightforward you will think he is not very nice, but he just wants to make sure you do not have any problems. So I guess what I am saying is I wouldn't be concerned with balanced tubes unless I want the cleanest output power and even then I need a balanced signal.
I have played guitar many times using a Hifi amp, but usually I am playing something hollow or semi hollow.
I hope this reply is more in-line with your topic.
-
Hay Sluckey and Ed
No problem. At no time was I mad or offended even though maybe my text language might of led you to think that. Yeah! besides slightly off balance OT's, PI's and supposedly matched tubes that are not, I've come to realize there is a lot off-sets in Push/Pull to try to pull together by hook or crook. No I'm not a Rookie but for sure I'm technically inept compared to you guys. I've never even thought about balancing a PI or have I tried the dual bias pot thing. My taste in amp tone keeps changing and seems like I'm leaning more and more to cleaner amp tone. I just let my pedals do the dirt and want my amps clean. So like you say, better balancing I reckon is the path to more pure clean tone. So who knows if I keep pursuing the path of purer clean I may find myself trying to pull the imbalance closer together.
I spent the first 15 years of my tube amp experience just building new tube amps and converting/re-purposing other old tube amps. Seems now I reached the end of my building, re purposing and now I'm just playing them and finally getting around to fine tuning them. Evey time I pick up a guitar I find myself fine tuning the setup and same thing with my amp builds going back over everything. So every time I rotate an amp out
you might see a thread pop up on it:>) Thanks, Platefire
-
Required reading: http://www.aikenamps.com/the-last-word-on-biasing (http://www.aikenamps.com/the-last-word-on-biasing)
This covers the various points mentioned in this thread.
I submit that it is ultimately impossible to know if your power tubes are in balance to a high level of precision.
* If you measure ea side for cross-over distortion, the measurements are inherently imprecise.
* Real tubes are not bogeys. Actual plate impedance, etc., will vary from tube to tube. If you balance bias by using the cathode current method, the operating points of each tube may differ, yielding different gain & harmonic characteristics.
* If you equalize output signal voltage of ea side, then, do you have equal signal voltage into ea tube, or will you adjust that per the KOC method? To get equal signal voltage out of ea power tube will require that they be biased differently.
* Will you leave your amp on for 20 min's or more before you play it?
There is no demonstrable reason to believe that highly precise balancing, even if possible, improves clean tone.
-
There is no demonstrable reason to believe that highly precise balancing, even if possible, improves clean tone.
Obivously I misunderstood something somewhere. I can certainly demonstrate causing distortion by imbalance of a cathodyne very easily. I can reduce it again as well.
I do not disagree with any of your comments and I am quite sure tubes are all somewhat different. It must be in your definition of what highly balanced means, but I agree it seems impossible to perfectly balance.
I was not speaking to a guitar amp when writing and should not have mentioned the balance. The reason I did is I have a Stereo KT88 and have improved the clean output wattage and balanced so preceived volume is the same with both pots positioned the same.
Now, I have no way of know if you could even hear the actual distortion or the decrease, but my pots are "balanced." This is a different type of balance, I get that. It is basically a dual monoblock. That is why I shouldn't have mentioned it. The topic is Bias.
-
Sorry Platefire, I missed the point too. I was wondering since you aren't a rookie, why you were concerend. Now I understand. Truthfully, on a guitar amp I am more likely to adjust signal level from the phase inverter to achieve the tube tone I am looking for. We cannot know it that LTPI with that "balanced triode" will do it. Also, transformer winding not balanced generally.
I have left the balance pot on a silver fender and simply bias with a resistor. Completely fixed. I did copy a Hifi Macintosh bias balance circit, but there again, I have not use it on a Guitar amp. Hifi generally doesn't run tubes as hard.
I just received a 4 point auto bias board same as used in some newer Hifi gear. It works well, but before I installed I checked signal level and current from the PI, 6sn7 and it had drifted, so I had to reduce resistance on one side a smideg.
I did use a Dual Bias pot on a build and I am sort of split in this. At IDLE you can match well for as long as the tubes will hold. The main problem is we generally do not play at idle. I like analogue needle Simpson 260 meters, because they really are cool, really! Anyway, everytime I have dummyloaded and increased the signal gain I can measure i get imbalance, sometimes one tube will pass more current at half signal and the other will at full signal. Swap places.
The best new production tubes I have bought are from Upscale Audio. The guy that owns the place is Kevin Deal and he is so straightforward you will think he is not very nice, but he just wants to make sure you do not have any problems. So I guess what I am saying is I wouldn't be concerned with balanced tubes unless I want the cleanest output power and even then I need a balanced signal.
I have played guitar many times using a Hifi amp, but usually I am playing something hollow or semi hollow.
I hope this reply is more in-line with your topic.
Not " completely fixed " you should have put a Hoffman 50k bias pot on the bias board... for the next set of output tubes installed... what you did , is what we all hate.. no adjustable bias .
-
Tube amp Fixed Bias, the most highly charged and controversial subject on Tube amp Forums.
sluckey I'm appointing you to be the Chief bias referee and Arbitrator!
-
enjoy! :l2:
-
:l4:
-
Not " completely fixed " you should have put a Hoffman 50k bias pot on the bias board... for the next set of output tubes installed... what you did , is what we all hate.. no adjustable bias .
It warms my heart knowing you are bothered by this.
-
Tube amp Fixed Bias
that's why I became good at xSE and using trannies for PI in PP, simple, clean, well received, life is good :icon_biggrin:
-
Shooter
I recall being really surprised first time seeing an old 40's amp schematic with a transformer for a PI. So I learned that to be pretty common back in the day. So are you saying your doing builds doing that again??? And what is xSE? Platefire
-
Yeah, I'm interested in knowing more about using transformers as a PI. Only stuff I've found so far pertains to radios. Haven't found much yet on guitar amps.
-
Checkout: Gibson GA15RVT
In the '40's preamp tubes were low mu. Output tubes were triodes with low mu, and needed huge input signal voltage, like around 115 VAC. A transformer could boost signal voltage better than lo mu tubes to drive a power triode. Also, signal bandwidth of OT's was then limited to about 100 Hz - 8000 Hz. So interstage tranny's didn't need to do better.
With the advent of tetrodes, then pentodes, power tubes became less signal hungry. And hi mu preamp tubes arose. A preamp tube is cheaper and smaller than a hunk of iron. By the early '60's OT's could pass the full audio bandwidth of human hearing: 20 - 20,000 Hz. You need to wind a high grade, expensive interstage tranny to match that -- which a cheap tube can do. Interstage tranny's all but disappeared.
Lately, hi quality interstage tranny's are making comeback in a small niche of tube hi-fi amps, for purists who want to avoid caps in the signal path.
-
Thanks for the info! I'll check that out.
-
Yeah, the Gibson GA15RVT and Gibson GA20RVT from the '60s both use interstage phase inverter transformers. Those are the latest amps that I know of that used them. Some of the much older Gibson BR series used them. I'm sure there are others too, but I can't recall any.
You can find all the Gibson models mentioned in Hoffman's schematic library.
-
Cool, thanks! I found an old post on a forum that one of the members mentioned some Fender 400W amp that used them also. Edit: Fender400 PS?
-
:hijack1:
your doing builds doing that again
yup, jj got close, the idea came from one of the 23 variations of the GA-5_
I built a fenderish pre, USA-1 tranny pi and 6V6s, never bonded
went to the plexi-ish and a KT88 SE (no PI :icon_biggrin:)
was liked, but just didn't get quite there
married the plexi-ish and the tranny PI with a matchless clubman PA, it became a hit, sold with in 2 days of demo day
what I found poking around, the iron adds a unique distortion, on a scope you get a leading edge spike, ("overshoot" in my old life). It's slower than a "digital" transient. I think this smacks the crap outta the PA tubes, that are already biased at MaxPdiss thenthe main signal gets there and the tubes are "pre-bent" :icon_biggrin: :dontknow:
from the guitar folk; "You nailed it" The owner toured and opened for Foghat, even filled in when someone from foghat got sick. so i'll take his word it works, for everyone, no.
xSE is single-ended, no pi needed :icon_biggrin: x is "how many tubes"
I soulda said self-biased instead of SE.
I like simple, but good, I've found you can get great sound from simple amps, will they trem-a-verb at 100W, no, that's why PA's and Pedals were designed :icon_biggrin:
-
Not hijacked -- an interstage tranny PI might in theory be better balanced than a tube PI, and won't drift. But (I think) except for special toroidal windings, there will be some difference in the 2 windings, because one is wound over the other.
-
Good discussion. Not concerned about hijacking. I did see Foghat many
times in the 70's. All lead playing was slide on an SG through a Marshal.
I just wonder if this is the same guy using Shooters amp? Thanks
for sharing the schematic. Amazing how old technology sometimes
comes back around. Platefire
-
is the same guy using Shooters amp
He was in the opening band, forgot name, but in the 2nd go round from the '80's
might in theory be better
I like it because it's simple, low part count, hard to screw up, cost's less than tube PI, if you're into marketing, could be a selling point.
the downside, it smokes easy, 50mA is about max, hence the cap coupled, which "adds" its own tonal kinks
can't find the link, but there is some well written smart guys that explain the deep weeds "logic" for why it's still a viable thing
-
Well I ordered 5 of the 1 meg, 1%, 3 watts, 750 Volts working Resisters that 92Volts referenced on post# 35. I will install the Plate Voltage test point in one of my amps to test the work ability of it. If it worksas good as hoped, will install it on my other amps that already has external bias test points also. Platefire