Cathode resistor value for SE 6L6/EL34/6550

[kleuck]

Don't forget that you need -to calculate the plate dissipation- to take in account the actual PLATE voltage, ie between plate and cathode (not plate to ground or B+ to ground or B+ to cathode), and subtract the screen current.
I use 6P3S in a VHT SP6 with a 235 ohms cathode resistor, and it leads to a 19-20 watts of plate dissipation under a B+ of 343 volts (6P3S are given for 20,5 watts)



Edit : well Tubeswell just explained

[Steve A.]

Don't forget that you need -to calculate the plate dissipation- to take in account the actual PLATE voltage, ie between plate and cathode (not plate to ground or B+ to ground or B+ to cathode), and subtract the screen current.

Okay- I haven't been subtracting the screen current. We should be able to figure that out with the voltages and resistances I've been posting (see if my computations below are correct.)

Okay #2- I screwed up the pin numbering and got pin #2 mixed up with pin #5 (Don't ask!) so I was getting 2.87vdc on pin #2 (the filament). I have not yet measured for DC voltage on pin #5 yet. The board is out and I need to replace some parts before putting the amp back together. And I might as well swap out the new 0.022uF cap I put in right before the 4k7 grid stopper since I have the board out.

3k3 screen resistor
470R cathode resistor:
Plate 278v
Screen 248v
Cathode 54v

screen current I = E/R
= 30/3300
= 0.009A or 9mA (right?)
cathode current I= E/R
=54/470
= 0.115A or 115mA
cathode current - screen current = 106mA

106m x 224v = 23.74 watts

BTW Fender has rated some of their 2x 6L6 amps as 60 watts. I always figured that 2x 6L6 was 40 watts (or 50 if you really pushed it). So how do they come up with 60 watts?

Steve Ahola

[jmaf]

BTW Fender has rated some of their 2x 6L6 amps as 60 watts. I always figured that 2x 6L6 was 40 watts (or 50 if you really pushed it). So how do they come up with 60 watts?

Two 6L6GC's in push-pull can deliver 60 watts when they're close to class B. Each tube can handle more than 30 watts for (almost) half the signal, both working together can transfer that power out to the transformer secondary with little loss.

I've made two 6L6GC's pump out 80 watts RMS while fooling around in the lab, and I'm sure others here, from what I've read about what they do to 6V6's, have pushed them even harder. You just can't push an amp like that out the door as a commercial product, so a 60 watt amp using two 6L6's is Fender going as far as their warranty lawyers will allow.

[tubeswell]

3k3 screen resistor
470R cathode resistor:
Plate 278v
Screen 248v
Cathode 54v

screen current I = E/R
= 30/3300
= 0.009A or 9mA (right?)

Without seeing your schematic, I'm guessing the 3k3 may be the plate-to-screen node B+ supply resistor (that is also supplying the pre-amp)? If it is, then the 30V voltage drop you are measuring across it would (most-likely) include the tube current for the pre-amp tubes. (9mA is a bit on the high side for a 6L6 screen current at idle; you'd expect something more like 4-6mA). If you are wanting to measure screen current accurately, you need a dedicated screen resistance (i.e.: in the form of a screen grid resistor going to Pin 4, and you measure the voltage drop across that)

[Steve A.]

Without seeing your schematic, I'm guessing the 3k3 may be the plate-to-screen node B+ supply resistor (that is also supplying the pre-amp)? If it is, then the 30V voltage drop you are measuring across it would (most-likely) include the tube current for the pre-amp tubes. (9mA is a bit on the high side for a 6L6 screen current at idle; you'd expect something more like 4-6mA). If you are wanting to measure screen current accurately, you need a dedicated screen resistance (i.e.: in the form of a screen grid resistor going to Pin 4, and you measure the voltage drop across that)

That is how I have been measuring it- right across the screen grid resistor. The B+2 supply resistor in 1k/5W. I have attached the preliminary (and mistake laden) factory schematic for the PicoValve. I will draw up one with corrections and edit the Dark Star MkII schematic with the changes I made. Still need to put the PicoRocket back together to measure the voltages...

Thanks!

Steve

[Steve A.]

I use 6P3S in a VHT SP6 with a 235 ohms cathode resistor, and it leads to a 19-20 watts of plate dissipation under a B+ of 343 volts (6P3S are given for 20,5 watts)...
[ATTACH=CONFIG]16063[/ATTACH]

So in the chart you uploaded:
rK would be the suggested value of the cathode resistor
iA would be the suggested bias current
UrK would be the maximum voltage drop you'd like to see across the cathode resistor
B+1 would be the suggested maximum plate voltage
B+2 would be the suggested maximum screen voltage

Is any or all of that correct? By "suggested" I mean that some of these can be exceeded without problems in guitar amps...

Can you post the information for 6V6's? I've never heard of the 6P1 or 6P3S tubes- is there an alternate number for them?

Thanks!

Steve Ahola

[kleuck]

The chart is actually a spreadsheet with my measurements, so voltages are not at all suggestions but the actual voltage with each tube and in this amp, B1 is the power amp voltage, B2 the preamp one, the "vintage" mod is with a 50 volts zener dropper.
And as i use different tubes, they do not always have the perfect bias etc, but for the 6P3S it's good.
That's why there is 2 rK, i use a switch to parallel two resistors, to have a correct idle current for 6V6/6P1P (single 470 ohms resistor), and 6P3S & EL84 (470 // 560)
6P1P are actually tiny 6V6 in a small tube, think to the 6AQ5 (different pinout) but curves are the same ; 6P3S are like 6L6, old ones, 20 watts of plate dissipation.
Here's the schematic :http://farm6.static.flickr.com/5186/...afc586dc_o.png
To choose bias point and calculate dissipation etc, look at Merlin's site : The Valve Wizard -Single Ended

[tubeswell]

So R22 in that schematic is actually 3k3. And if there is a 30V drop across it then yes, that's 30/3300 = 9mA screen current (which is more than I thought I would be with a 3k3 g2 resistor). Hmmm

[Steve A.]

So R22 in that schematic is actually 3k3. And if there is a 30V drop across it then yes, that's 30/3300 = 9mA screen current (which is more than I thought I would be with a 3k3 g2 resistor). Hmmm

R22 is 5K6 on the stock PicoValve circuit. I replaced it with a 1k/5W thinking that I could improve the sound of the triode mode- it helped a little but not enough. When I reconfigured the preamp to resemble the Trainwreck Rocket I found that the combination of 1k screen and 250R cathode was frying the cathode resistor.

I will edit the Dark Star schematic to reflect the actual wiring in my PicoRocket (as I like to call it). And will post a corrected schematic to the PicoValve when I get around to doing that.

Steve

[kleuck]

So R22 in that schematic is actually 3k3. And if there is a 30V drop across it then yes, that's 30/3300 = 9mA screen current (which is more than I thought I would be with a 3k3 g2 resistor). Hmmm

Indeed, i got 4,5 mA, strange.

[Steve A.]

OK- I put it back together with a 0.047uF cap ahead of the 4k7 grid stopper (to eliminate the possibility of a leaky cap). BTW the PT is 0-260vac and goes into a fullwave bridge rectifier. With no output tube the B+ is around 368vdc. I tried another 6L6, a 6V6 and an EL84- only the EL84 showed no DC voltage on g2. But the cathode voltage kept climbing and climbing- when it got to 70v I turned off the amp.

EL84 277S 211G1 67.5K 0g2

6L6 286.5S 251.4g1 57.5K 2.1vdc(g2)

6V6 303S 264G1 49K 5.3vdc(g2)

The amp sounds great w/ none of the problems I mentioned at the beginning of this thread. But I need to figure out the weird readings before I button it back up. BTW the output tubes were getting really hot really fast...

Here's a thought- maybe I ought to try increasing the value of the power supply resistor between B+1 and B+2?

FWIW the cathode resistor bypass cap is 330uF 63v- is there any way that could be increasing the voltage on the cathode?

Question: how does lowering the value of the cathode resistor affect the voltage at the cathode? And how does raising the value affect it?

I'm going to sleep!

Steve

[jmaf]

FWIW the cathode resistor bypass cap is 330uF 63v- is there any way that could be increasing the voltage on the cathode?

Question: how does lowering the value of the cathode resistor affect the voltage at the cathode? And how does raising the value affect it?

I don't think the cathode cap would increase the voltage at the cathode, in principle it'll just hold it where the resistor sets it. What a cap could do is the opposite, in fact, start leaking heavily and then drop the voltage at the cathode, you'd see a bulge on top of it if it's electrolytic.

There are two things that I know that would drive the screen grid to excess current: the plate voltage going too far under the screen(which should only happen during signal swings, not idle) and using a too small or no screen resistor.

[kleuck]

Raising the value of the cap makes the amp "stiffer" -almost like a fixed-bias one, though in A class that's not really that audible- and can improve the lows when the amp has a low value (22 or 47µf) but here it's already a huge cap.

EDIT : there's something badly wrong in you amp Steve, first, the voltage drop from your B+ is way too high, and the cathode current is way too high too : 120 mA with the 6L6 and 100 mA with the 6V6, it's crazy, actually, with a 470 Ohms cathode resistor and such low voltages, you should never obtain these currents !

[Steve A.]

Raising the value of the cap makes the amp "stiffer" -almost like a fixed-bias one, though in A class that's not really that audible- and can improve the lows when the amp has a low value (22 or 47µf) but here it's already a huge cap.

EDIT : there's something badly wrong in you amp Steve, first, the voltage drop from your B+ is way too high, and the cathode current is way too high too : 120 mA with the 6L6 and 100 mA with the 6V6, it's crazy, actually, with a 470 Ohms cathode resistor and such low voltages, you should never obtain these currents !

Do you think a problem with the transformers and/or choke could cause behavior like that? Or perhaps something miswired in the power supply? (I did find some mistakes in how they translated the schematic onto the pcb.) It's not amps can be cursed (like some people seem to be ) they are just parts and wire.

It would have been good if I had measured voltages when the amp was still stock.

As the amp is now it sounds great but I have a hunch that I would be changing output tubes and cathode resistors very often. Well, I do like mysteries and intellectual puzzles...

Thanks for all of your help!

Steve

[kleuck]

I do not have enough experience to figure out what can be the issue, sorry, but probably not from the transformer nor the choke.I think it's more a mistake in the power tube wiring, or a dead cap (in the tonestack), you should look for a DC voltage on the grid for example.
Replace the cathode cap with another (new one) to be sure too (even of lower value) or simply remove it to test.