Page 1 of 1

Expanding Biasing Range in Stereo 35, Part II

PostPosted: Thu May 14, 2015 7:27 pm
by petercapo
I’m revisiting the individual bias adjust kit from Dynakitparts (that appears to be the same as the biasing setup in the DIYTube ST35). I played around with some different resistor values (on paper) for the two parallel branches. I think I came up with a change to the resistor combination that would expand the overall biasing range.

Disregarding the 10Ω resistors, the original values in the two parallel branches are 470Ω || (1KΩ + 10KΩ pot). If I did the math right, this makes the original range of adjustment run from ≈ 320Ω to ≈ 451Ω.

Again disregarding the 10Ω resistors, if I change this to 620Ω || (390Ω + 10KΩ pot), I get a range of ≈ 240Ω to ≈ 585Ω.

I suppose this is all well and good, but I am wondering about the corresponding change in power dissipated in the pot, which is rated at 0.5W. I made an attempt at calculating the quiescent power dissipated in the pot assuming a total resistance for the two branches adjusted to 400Ω with 35mA total cathode current. With the two original fixed resistor values, the pot dissipates around 0.04W. With the changes to the values of the two fixed resistors, power dissipation in the pot increases to 0.10W. Again, this assumes I got the calculations right.

0.10W dissipated in a 0.5W pot under quiescent conditions is not a concern. But, I don’t know how to calculate the power dissipated in the pot with the amp running full tilt. Could the pot’s 0.5W rating be exceeded with the amp running at full power?

Thanks.
Peter

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Fri May 15, 2015 7:11 am
by petercapo
Perhaps I could rephrase. If the Stereo 35 idles at 35mA through each 6BQ5, how much would this current rise if the amp were running at full power? Anyone know?

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Fri May 15, 2015 5:36 pm
by dcriner
For a push-pull amp, power is the same at idle as it is with a varying signal applied.

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Wed Jun 17, 2015 6:00 am
by Shannon Parks
Peter, I think your mods will be OK. Those T93 pots don't start their power de-rating until around 70C, so keeping at 25% or below should be fine. This allows for peak current draw (100% beyond idle at 70mA) and still keeps dissipation at 50% or below pot rating. I think I was trying to be pretty conservative on the original circuit.

Shannon

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Mon Jun 22, 2015 11:18 am
by petercapo
Thanks, Shannon!

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Tue Dec 29, 2015 7:00 pm
by Blooze
Curious. I was wondering about the EFB mod in lowering the bias current to 27mA to extend tube life, but figured it would be a pita with my old revB board. This mod above by petercapo would give me a little more range. I guess I could also swap the 10R for a 100R to bring the whole thing down as well?

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Tue Dec 29, 2015 9:12 pm
by petercapo
From what I understand about EFB, it saves on power tube life without losing power output. If you want to extend power tube life by lowering the bias (with no EFB installed), then the overall clean power output would decrease some, I would think. Increasing the 10 ohm resistor to 100 ohms would be quite a change – I don’t think I would try that.

I cooked up this bias circuit mod to expand the range both up and down to accommodate differences in power tube brands and also variations/adjustments to the B+. If you want to use it exclusively to reduce the bias current, then further reworking of the resistor values (other than the 10 ohm) may result in a greater potential reduction in the bias current. It is my understanding, however, that the issue with power tube longevity is specifically related to the plate power dissipation. Reducing the B+ is another way to reduce the plate power dissipation, while keeping the current at 35mA or so.

Without EFB, I don’t know if it might be better to reduce the stress on the power tubes by reducing only the bias current vs. reducing only the B+. I’ve been thinking about some of the potential implications of one approach vs. the other. Perhaps someone could comment on this. Actually, I hope someone will.

Of course, another way to have the EL84s last longer would be to use the ones that are rated for a higher plate dissipation. A number of EL84s have their plates rated at 12 watts. But, the Sovtek EL84M and, IIRC, the Electro Harmonix EL84 are rated for 14 watts. Some of the original Russian military EL84 selling on ebay for instance should have better longevity. The new Tung Sol EL84 is said to be rugged, though I have not seen a datasheet for it yet.

BTW, I have not yet implemented my bias mod, as shown in the first post of this thread. If you change your fixed resistor values, watch out for their power rating, which might possibly have to be bumped up.

Re: Expanding Biasing Range in Stereo 35, Part II

PostPosted: Sat Jan 02, 2016 2:57 pm
by burnedfingers
petercapo wrote:I’m revisiting the individual bias adjust kit from Dynakitparts (that appears to be the same as the biasing setup in the DIYTube ST35). I played around with some different resistor values (on paper) for the two parallel branches. I think I came up with a change to the resistor combination that would expand the overall biasing range.

Disregarding the 10Ω resistors, the original values in the two parallel branches are 470Ω || (1KΩ + 10KΩ pot). If I did the math right, this makes the original range of adjustment run from ≈ 320Ω to ≈ 451Ω.

Again disregarding the 10Ω resistors, if I change this to 620Ω || (390Ω + 10KΩ pot), I get a range of ≈ 240Ω to ≈ 585Ω.

I suppose this is all well and good, but I am wondering about the corresponding change in power dissipated in the pot, which is rated at 0.5W. I made an attempt at calculating the quiescent power dissipated in the pot assuming a total resistance for the two branches adjusted to 400Ω with 35mA total cathode current. With the two original fixed resistor values, the pot dissipates around 0.04W. With the changes to the values of the two fixed resistors, power dissipation in the pot increases to 0.10W. Again, this assumes I got the calculations right.

0.10W dissipated in a 0.5W pot under quiescent conditions is not a concern. But, I don’t know how to calculate the power dissipated in the pot with the amp running full tilt. Could the pot’s 0.5W rating be exceeded with the amp running at full power?

Thanks.
Peter


Maybe my math is bad but I get 239ohm to 593 ohm for the 620 390 and 10K pot

470 resistor 1K and 10K pot I get 319 to 450 ohms