adjusting cathode bias

[2manynotes]

i have a Fender Vibrochamp. about a year ago i was at the Torres Engineering site and thought id try the Bias kit for the champs. he claims you CAN adjust the bias on a cathode biased amp, simply by taking a few measurements, and swapping out this resistor. i followed the directions and replaced the 470uf 1w resistor (going to pin 8 of the power tube) with a 840uf 5w (or thereabouts... there were many that came in the kit, and this seemed like the one to use)
the problem is, i have since lost the instructions and all the rest of the resistors, so i dont know if its "right" any longer. (i have replaced the power tube with a new JJ. i also have a new Tung-Sol 6V6) its pretty clean up to about 4, then it gets dirty pretty quick. by 6 or 7 its "full on" overdrive!

SO MY QUESTION IS: what should this bias be? how do i measure? volts? resistance? measure from where to where?? im lost.

anyone?
thanx!
- jim

[bob p]

yes, a cathode biased amp has its bias point changed by changing the value of the cathode resistor.

the normal cathode resistor on a champ/vibrochamp is 470 Ohms. the normal cathode voltage across the resistor is about 20-21 VDC. check the voltage referenced to ground on Pin 8 and see what you've got. that measurement is a quick and dirty way to see where you stand, without having to perform current calculations.

[bob p]

one other thing to think about if you've got a Torres modded amp -- its possible that more has been changed in your amp than just the cathode resistor on the power tube.

i'd take a look at the original vibrochamp schematic to determine what else might have been changed on your amp.

[MWJB]

A typical tolex Champ/Vibrochamp dissipates 20-22W with the stock 470ohm resistor . It can take it, but going to 660-900ohms will get the plate dissipation closer to the 14W rating for a 6V6.

Which should you use? Use what sounds best! Just don't go below 470ohms, that's tempting fate.

As Bob P says a "rule of thumb" method would be to measure cathode voltage at pin8 of the 6v6 and divide the voltage by the cathode resistor value.

Then multiply the answer by plate voltage less cathode voltage - this will give you your plate & screen dissipation, now knock off a W to exclude the screen dissipation and you'll be near enough...

Eg, you have 30v at your cathode (more likely than 21v given the probable plate voltage & obversize cathode resistor)...

30v/840ohms = .0357 (35.7mA)
35.7 x 400v = 14.28W
subtract 1W for screen supply = 13.28W plate dissipation.

Not the most accurate method but plenty good enough for you/your VC's purposes.

If you're a W over or under 14W I wouldn't panic, if the amp sounds good that's what matters, nothing is going to fry.

[2manynotes]

you guys are fantastic! im SO GLAD i found this forum! ill be testing this out tonight! i feel like im really getting an education!! between this and my other post, this has been the most productive day of screwing around at work ive ever had! thanx!!

[Konny]

Hey guys,

can anyone tell me how to calculate the exact bias?
I do like rule of thumbs but Im thinking about switching the cathoderesistor in my champclone to go for 6L6 and a bit more power (OT and PT can handle that, no problem )

Thanks,
Swen

[Enzo]

Rule of thumb? It is Ohm's Law. Nothing more fundamental in electronics than that.

VOltage dropped across a resistor is determined by the current flowing through it. V = I x R.

Power dissipated in a tube is determined by the voltage across it - cathode to plate - times the current through it.

Current through the tube will be the same as the current through the cathode resistor. Currents are the same in a series circuit.

There will be a small amount of screen current, but really, you can pretty much ignore that. or subtract a watt as someone suggested.

[Old Tele man]

...which? "exact CATHODE bias" or "exact GRID bias"?

...there's a theoretically-based equation for estimating control-grid bias voltage (-Vgq), which can be viewed as just a "negative" version of Vk:

-Vgq = (0.9)*(3/2)*(Ikq / gm) - (Vsq/µ1 + Vpq/µ2)

...where ("q" suffix denotes quiescent or 'idle' condition):

Vgq = Control grid quiescent bias voltage
Ikq = Cathode quiescent current (single tube)
Vsq = Screen quiescent voltage
Vpq = Plate quiescent voltage
gm = Tube TETRODE transconductance, amps-per-volt (6V6 = 0.0041A/V)
µ1 = Tube TRIODE amplification factor (6V6 = 9.8)
µ2 = Tube TETRODE amplification factor (6V6 = 205)

...for example a 6V6 operating with Vpq = 420V, Vps = 400V and Ikq = 22mA

-Vgq = (0.9)*(3/2)*(0.022/0.0041) - (400/9.8 + 420/205) ~ -35.6Vdc

[MWJB]

There is no negative grid voltage on a champ/vibrochamp (or any other cathode biased amp) at idle. So I'm not sure how this helps.

If we assume that Konny wants to know his plate current, Enzo's reference to Ohm's law has all he needs to know. Or, even easier, if it's running typical Champ type voltages (say <430v from plate to ground), try 470ohms to 1K & anything you have in between...what sounds best IS what's best.

For a 6L6 use a 10W cathode resistor and a 50v to 100v bypass cap.

[Old Tele man]

There is no negative grid voltage on a champ/vibrochamp (or any other cathode biased amp) at idle. So I'm not sure how this helps.

...notice the statement: "...a "negative" version of Vk..." which reiterates the inverse relationship between cathode bias (Vk, a positive voltage) and conrol-grid bias (-Vg, a negative voltage). While the equation is specifically for control-gird voltage, we can still use it by "knowing" that only difference between the -Vg and Vk is the "sign change," which allows the equation to ALSO be used to estimate Vk...ie: a "backdoor" design approach, if you will.

[Enzo]

I would add that switching from fixed bias to cathode bias also effectively lowers the plate voltage. This in the sense that cathode voltage must be subtracted from plate voltage to get voltage across the tube. Of course a grounded cathode is zero to subtract. It would be something to consider, though it would not upset the calculations a lot.

[MWJB]

...plus in most cases, in cathode bias mode the tubes are drawing more idle current than in fixed bias, so in addition to a drop in plate voltage due to subtraction of the cathode voltage, B+ may drop still further...especially in a champ type amp where a relatively small swing in plate current causes a noticable change in B+.

My champ has a fixed/cathode bias switch, depending on how hot or cold I want to bias a 6L6 tube I might see a swing in plate voltage of 40v.

Assuming cathode bias operation - even when you've done the calculations you still need to ear test the result, so from a practical point of view it's still just as quick & easy to sub 2 or 3 different value cathode resistors.