Who has added separate bias pots and what did you find?
Moderators: pompeiisneaks, Colossal
-
iknowjohnny
- Posts: 1070
- Joined: Thu Apr 24, 2008 2:10 am
- Location: los angeles
Who has added separate bias pots and what did you find?
I'd like to hear opinions for those who have built amps with a bias pot for each tube and what you found when you experimented with balancing them verses setting them unbalanced. How far off did you try and did the amp sound richer and smoother one way then the other? You always read debates about whether guitar amps are better off matched or not, and i would think those who have tried this would have the answer, subjectivity aside.
Re: Who has added separate bias pots and what did you find?
I built an amp once with separate bias pots. Not for the reasons you mention but to have an amp that I could use some of my unmatched orphan tubes in. I don't think there was anything special about it as far as gaining tone other than being able to use a pair of tubes that normally would bias 10 to 15 ma apart.
If you wanted to experiment with this, the easiest way would be to purposefully use an unmatched pair.
Matching (or mismatching) the bias is only one piece of this issue; another is the phase inverter. The phase inverter circuit is not perfectly matched and then there is the phase inverter tube, which may or may not have matched sections.
If you really wanted to experiment with matching, then I would suggest setting up something like a PPIMV with trimpots so you could match the signals driving the power tubes.
Quite honestly, I don't worry much about this. I build with the 82K/100K resistors and usually the tube that gets stuck in that spot is one that I don't like for a gain stage. A good, reliable tube is what goes in there. It seems like no matter how many tubes I buy, I am always scrounging for that last one.
If you wanted to experiment with this, the easiest way would be to purposefully use an unmatched pair.
Matching (or mismatching) the bias is only one piece of this issue; another is the phase inverter. The phase inverter circuit is not perfectly matched and then there is the phase inverter tube, which may or may not have matched sections.
If you really wanted to experiment with matching, then I would suggest setting up something like a PPIMV with trimpots so you could match the signals driving the power tubes.
Quite honestly, I don't worry much about this. I build with the 82K/100K resistors and usually the tube that gets stuck in that spot is one that I don't like for a gain stage. A good, reliable tube is what goes in there. It seems like no matter how many tubes I buy, I am always scrounging for that last one.
-
Andy Le Blanc
- Posts: 2582
- Joined: Sat Dec 22, 2007 1:16 am
- Location: central Maine
Re: Who has added separate bias pots and what did you find?
separate pots for bias is good up to a point...... for two tubes its practical
but for four or more it become a problem of rationing current..... the adjustment
of one effects all the others until a equalibrium is reached..... and unless the
the bias supply was designed for the load ... you might not get what you hoped for...
the textbook demonstration for "push-pull" requires two signals of equal magnitude
that are 180 dgrees out of phase...... inverters that need two gain stages require
identical gain stages..... this means DC and AC balance where it can be found
AC balance is a priority to get the least distortion and the most power.....
but a guitar amp IS ment to color the tone so its up to your ears...
but for four or more it become a problem of rationing current..... the adjustment
of one effects all the others until a equalibrium is reached..... and unless the
the bias supply was designed for the load ... you might not get what you hoped for...
the textbook demonstration for "push-pull" requires two signals of equal magnitude
that are 180 dgrees out of phase...... inverters that need two gain stages require
identical gain stages..... this means DC and AC balance where it can be found
AC balance is a priority to get the least distortion and the most power.....
but a guitar amp IS ment to color the tone so its up to your ears...
lazymaryamps
Re: Who has added separate bias pots and what did you find?
I did this to an amp where the OT DC resistance was different on the 2 primary windings. I was able to balance the output stages using a scope, but the typical method of measuring current did nothing for the amp.
I prefer a single bias control and a PI balance like Dumble uses to balance the PI and output.
I prefer a single bias control and a PI balance like Dumble uses to balance the PI and output.
Re: Who has added separate bias pots and what did you find?
I've done it too, it works, but when you factor in time, parts, chassis/layout space, etc., buying a matched set becomes a much more attractive and elegant solution, unless you absolutely need NOS KT66's or something....
Life is a tale told by an idiot -- full of sound and fury, signifying nothing.
...in other words: rock and roll!
...in other words: rock and roll!
Re: Who has added separate bias pots and what did you find?
Bob, sorry for the newb question but you've got me wondering...what does this actually do? I checked out a Dumble layout and I see the 5k trimmer there between the PI plate resistors. So if you balance the output to the tubes, how does this affect the tube bias? The bias voltage going to the tubes is going to be the same (220/220k). But suppose your tubes were supposed to be "matched" but weren't, wouldn't you still want (theoretically) both tubes to be running at the same current? I thought the bias adjustment was supposed to compensate for variances in tube manufacture...does balancing the PI output do this? Aren't the differences in plate resistors intentionally used to unbalance the PI? Still trying to figure all this stuff out...Bob-I wrote:I prefer a single bias control and a PI balance like Dumble uses to balance the PI and output.
Thanks
-
iknowjohnny
- Posts: 1070
- Joined: Thu Apr 24, 2008 2:10 am
- Location: los angeles
Re: Who has added separate bias pots and what did you find?
I had a conversation at 18watt about this a while back. Balancing involves the DCV at the plates and AC signal going to the power tubes. If you balance the DC the AC signal becomes UNbalanced, which is a desirable thing in guitar amps. (more on that in a sec) It was done like that (82k/100k plate resistors) back in the day to help balance the signal to the power tubes because at the time there weren't any thioughts about creating guitar distortion. it was all about clean and loud. later on it was determined that using 100k/100k is the better way for guitar amps because an unbalanced AC signal doesn't cancel out even order harmonics. thats how it was explained to me anyways. and i went and looked at a lot of late model boutique amps and sure enough they are using balanced resistors in many of them.
as for the effect, i can tell you that today after changing my NFB circuit so the signals are unbalanced, it does make a big difference in the tone if the balance is quite off. the tone becomes much softer and loses the harsh edge on the high end. Much like a cut knob does. i don't so much notice more harmonics tho. if anything maybe less, but i would say less odd order harmonics because the tone becomes so much smoother. I checked the AC levels at the PT grids with a keyboard organ signal sustaining thru the amp and one side was 2-1/2 times the signal as the other ! but i gotta say, i believe i'm going to keep it like that because it sounds better. But that was changing the balance on the cathode side of the tube with the plate R's remaining 100k/100k. whether theres a difference i dunno. But i couldn't get much difference in signal level at all using 100k/82k.
as for the effect, i can tell you that today after changing my NFB circuit so the signals are unbalanced, it does make a big difference in the tone if the balance is quite off. the tone becomes much softer and loses the harsh edge on the high end. Much like a cut knob does. i don't so much notice more harmonics tho. if anything maybe less, but i would say less odd order harmonics because the tone becomes so much smoother. I checked the AC levels at the PT grids with a keyboard organ signal sustaining thru the amp and one side was 2-1/2 times the signal as the other ! but i gotta say, i believe i'm going to keep it like that because it sounds better. But that was changing the balance on the cathode side of the tube with the plate R's remaining 100k/100k. whether theres a difference i dunno. But i couldn't get much difference in signal level at all using 100k/82k.
-
Andy Le Blanc
- Posts: 2582
- Joined: Sat Dec 22, 2007 1:16 am
- Location: central Maine
Re: Who has added separate bias pots and what did you find?
all inverters reproduce a range of frequency where both signals are 180 degrees
out of phase and put back together thru push-pull nearly perfectly.....
or a linear response...... signal outside this range produces various levels
of even and odd order distortion products...... balancing the inverters plate V
and the AC signal output and the current of the output tube bias increases the
range or linear responce of the amp..... you still get the characteristic spectrum
of distortion products of what ever inverter type and other factors associated
with how your amp is built, tube type, output circuit, power supply, tube type etc.....
if your useing so much feedback the you have to severly imbalance your
design to get good tone color then the issue is how the feedback is employed...
out of phase and put back together thru push-pull nearly perfectly.....
or a linear response...... signal outside this range produces various levels
of even and odd order distortion products...... balancing the inverters plate V
and the AC signal output and the current of the output tube bias increases the
range or linear responce of the amp..... you still get the characteristic spectrum
of distortion products of what ever inverter type and other factors associated
with how your amp is built, tube type, output circuit, power supply, tube type etc.....
if your useing so much feedback the you have to severly imbalance your
design to get good tone color then the issue is how the feedback is employed...
lazymaryamps
Re: Who has added separate bias pots and what did you find?
IKJ, thanks for the thorough reply. That's interesting about changing the plate resistors to 100k/100k. I looked at the Dumble 124C layout and it has 110k/120k with the 5k trimmer in the middle. I wonder why the difference in signal on your amp on one side of the OT was 2.5x the other? Do you have dual bias on the amp when you are making these observations (you said you were changing the balance on the cathode side)?iknowjohnny wrote:I had a conversation at 18watt about this a while back. Balancing involves the DCV at the plates and AC signal going to the power tubes. If you balance the DC the AC signal becomes UNbalanced, which is a desirable thing in guitar amps. (more on that in a sec) It was done like that (82k/100k plate resistors) back in the day to help balance the signal to the power tubes because at the time there weren't any thioughts about creating guitar distortion. it was all about clean and loud. later on it was determined that using 100k/100k is the better way for guitar amps because an unbalanced AC signal doesn't cancel out even order harmonics. thats how it was explained to me anyways. and i went and looked at a lot of late model boutique amps and sure enough they are using balanced resistors in many of them.
as for the effect, i can tell you that today after changing my NFB circuit so the signals are unbalanced, it does make a big difference in the tone if the balance is quite off. the tone becomes much softer and loses the harsh edge on the high end. Much like a cut knob does. i don't so much notice more harmonics tho. if anything maybe less, but i would say less odd order harmonics because the tone becomes so much smoother. I checked the AC levels at the PT grids with a keyboard organ signal sustaining thru the amp and one side was 2-1/2 times the signal as the other ! but i gotta say, i believe i'm going to keep it like that because it sounds better. But that was changing the balance on the cathode side of the tube with the plate R's remaining 100k/100k. whether theres a difference i dunno. But i couldn't get much difference in signal level at all using 100k/82k.
I guess I'm still trying to understand the relationship between balancing the PI and running dual bias pots. I think I get the advantage to being able to fine adjust the balance in the PI, but wouldn't being able to independently bias both tubes still be a good thing?
Based on your comments, I'm guessing there is a line where what may be technically ideal and what sounds good?
Re: Who has added separate bias pots and what did you find?
Not exactly. The 2 sides of the PI don't amplify exactly the same, therefore most amps use different plate load resistors, typically 82k/100k, Dumble uses 91K and 110K (typically) with a 5 or 10K trimmer between them. Adjusting the trimmer balances the output of the 2 sides of the PI.Colossal wrote:Bob, sorry for the newb question but you've got me wondering...what does this actually do? I checked out a Dumble layout and I see the 5k trimmer there between the PI plate resistors. So if you balance the output to the tubes, how does this affect the tube bias? The bias voltage going to the tubes is going to be the same (220/220k). But suppose your tubes were supposed to be "matched" but weren't, wouldn't you still want (theoretically) both tubes to be running at the same current? I thought the bias adjustment was supposed to compensate for variances in tube manufacture...does balancing the PI output do this? Aren't the differences in plate resistors intentionally used to unbalance the PI? Still trying to figure all this stuff out...Bob-I wrote:I prefer a single bias control and a PI balance like Dumble uses to balance the PI and output.
Thanks
This has nothing at all to do with the bias of the power tubes but it can be used to balance the output so that the positive side matches the negative side. The difference in tone is subtle, but then everything Dumble did was subtle. It's the sum of all these little things that make the amp what it is.
That said, the smoothness of the Dumble isn't for everyone. Some folks like things more raw so maybe a balanced PI isn't the best for everyone.
- David Root
- Posts: 3540
- Joined: Fri Aug 04, 2006 3:00 pm
- Location: Chilliwack BC
Locking Bias Pots
The ability to choose where you want to be matchingwise is a big plus to me. Just use locking pots so that they won't change after you get them where you want them.