Bias circuit in Koch Twintone I

[Oliver]

Hi Guys. Can anyone help me figure this out?
I want to modify the bias circuit on my Koch Twintone I. It goes from - 55 vdc to - 35 vdc, which in my opinion is good only for hot EL34 tubes. I would like something in the range of -35 to -25. I gathered I need to change the bias feed resistor but I don't know which one it is in relation to the trimpot. Is it the only component I need to change? I have a schematic, but it is from an earlier batch of Twintones. I can also provide a picture of the circuit board if it helps...
Thanks!!

[Enzo]

Without a schematic, it is darned hard to advise you what parts to change to what.

HAve you actually had tubes that -35v was too cold for?

[Oliver]

Yes, TungSol EL34B run at 20 mv max and my EH at 30 mv (I would like them around 36 mv...)
Here is a recap of the specs I measured

Range of negative voltage on pin 5 with bias pot: -55 V to -35 V
> Plate voltage on pin 3 without tubes: 415 V
> Plate voltage on pin 3 with tubes, bias at -55 V: 413 V
> Plate voltage on pin 3 with tubes, bias at -35 V: about 398 V
> Screen voltage on pin 4, bias at -55 V: 406/408 V
> Screen voltage on pin 4, bias at -35 V: 386/388 V

For the time being, I have ordered tubes a the Tubestore that are selected as running hotter than average, so we'll see. I still want to change the circuit to have more flexibility. Here is part of the schematic. Serial numbers don't match (mine is H848) but the circuits look similar. Koch does not respond to my emails about schematics...
Thanks

[Oliver]

[Enzo]

Ah, well a partial schematic is better than none, but the bias supply is not on this page. It should not be hard to trace out the bias supply by hand, if need be.

A google search finds your schematic in the first post of this thread on our forum.

http://music-electronics-forum.com/t23074/

SO looking at that...

I see 40VAC to start with, and that makes about -56vDC. R97, R98, P13 make a voltage divider, with the bias voltage at the wiper of P13. I don't like reliance on the wiper alone for bias voltage, but that is what you have. I'll get over it.

My calculations come up with a sweep range of 55v down to about 27v. SO since you get only down to 35v, I will assume some part values are off some, probably the trimmer. No matter.

In series across that -56v is the 1k R97, the 22k P13, and the 22k R98. 45k total, and you can calculate the voltage division at the extremes of P13, and get my numbers.

Easy thing to do would be change R98 from 22k to something lower. Unsolder one end of it and lift it out of circuit. Now you can measure the actual resistance of the trimmer. Pots are notorious for not being accurate, so though rated 22k, it might be only 20k or something. Measuring it would give you a more accurate figure for calculating. But even just guessing, I get a range of 54-17v with R98 at 10k, and if 15k then a range down to about 22v.

Changing R98 leaves the upper end near 55v, so you still could tame the hottest tubes. However, instead of decreasing R98, you could alternatively increase R97. If you raised it to 10k instead of 1k, then the top voltage would only be about 45v instead of 55v, but the low end would go down to about 23v.

I'd just stick a 10k in place of the 22k at R98. The stock voltage divider draws only about 1.2ma from the bias supply, and the lower resistance at R98 would only increase that to 1.6ma, so a negligible draw. SO no concerns there.


Before you do any of that, look at the actual parts at R97 and R98 as well as the trimmer at P13. The values may already not match the schematic, so we would redo the arithmetic, but I doubt it would be far off anyway.

[Oliver]

I'm sorry I had provided the wrong part of the schematic, I got it from the post you mention. Looking at my circuit board, I see that the trimpot is 25 K, R97 is 1K and R98 is 33K. 2 differences then... Do you think that I could solder an additional resistance on top of the one in place to decrease the total to some given value, or would that be a half-ass job? It seems a hassle to unplug everything to remove the circuit from the chassis. I'm not too confident on how to drain the voltage caps either ...
thanks for your help

[Enzo]

OK, that explains the difference between your range, and my calculations.

SO 25k pot is close enough to 22k to not matter. 33k for the bottom resistor raises the lowest voltage. SO reduce the 33k. I still think 10k would be a reasonable target.

Can you tack a resistor in parallel? Oh I suppose. I just try to do things "right" here. But I am a pro, charging money for my work. A 27k across the 33k makes close to 15k. a 15k parallel the 33k makes about 10k.

[Oliver]

You're right, I should probably try to do things the right way, so it's neat and tidy. Thanks for all the great info anyway. Can you just tell me what wattage resistor I need ? They look like regular 1/4 watt but I want to be sure
Thanks

[Enzo]

Ohm's Law. WHat voltage is across what resistance? That tells you current, and current times voltage is power. Whatever power you need, then double or triple that for the resistor rating.


1k, 25k, 10k, all in series across 56v. You are not changing the other parts, so let's look at the 10k. And yes, whatever size is already there is probably a good size to use.

Total resistance is 36k. It is across 56v. 10k will then see 10k/36k x 56v = 16v Now 16v across 10k ohms? I = 16/10,000. That is 0.0016A, or 1.6 milliamp. 1.6ma times 16v = 0.026 watts. That is 26 milliwatts. SO a quarter watt resistor has 10 times the dissipation you need. Even a tiny 1/8 watt would still have 5 times your need. SO ANY 10k resistor you are likely to have will work just fine.

[Oliver]

Problem solved!!
Everything works great with your suggestions. Now a bias range to cover all my needs.. many thanks!