Biasing 5879 pentode preamp tube - calculations

[rockinrob86]

I'm working on a new build amp - don't want to say I designed it, but the first channel isn't copied from anywhere exactly. I wanted to use 5879 pentode for the channel, but I am encountering some unexpected behaviors and want to understand what is happening. I am not sure how to do the math here and would appreciate the formula.

I currently have the 5879 set up in pentode, with b+ of 218 VDC coming to a 470K plate resistor. I have it biased with a 3.3K cathode resistor bypassed by a 25uf cap. 1M screen resistor and .1 screen cap to ground

What I don't understand, is the 470K resistor is dropping 202 VDC! I am measuring 16V on the plate!

Can someone explain what is going on here?

I found some of these circuits shown here http://el34world.com/Forum/index.php?action=dlattach;topic=21831.0;attach=64660;image

mine is basically the divided by 13 circuit, but I hadn't seen this when I did it.

This is the datasheet http://www.r-type.org/pdfs/5879.pdf

I also don't understand why it actually sounds pretty good and "works" - it sounds dark, but is otherwise a guitar sound. I used my tele, cranked it up and it was a usable tone - although I went to tweak it because of the dark tone, and then I realized how the voltages were off.

I know if I drop the cathode resistor the plate voltage will go up - but I think it's time to understand what I'm doing mathematically rather than just playing around. It's time to man up and learn!

 

[PRR]

> 470K plate resistor.

Why so high? Almost any audio load will suck that down. ("Dark tone")

> 470K plate resistor. 1M screen resistor ...! I am measuring 16V on the plate!

Did you look at the R-C amp table on page 2 of the sheet you linked?

Personally, guitar amp, _I_ would take the 0.1Meg or 0.2Meg column, and try both sets of Rg2+Rk to see what worked.

> I know if I drop the cathode resistor the plate voltage will go up - but I think it's time to understand what I'm doing mathematically

You don't say what Rk you used.

Resistor-biased G2 on small signal pentodes is not amenable to "math". Small audio stage operating point is usually way below the Show-Off operating point in data and curves. Resistor decade-box is MUCH faster than brain-burning. 

 

[moamps]

Can someone explain what is going on here?

Did you connect any resistor between g1 and gnd?

 

[rockinrob86]

Thanks guys.

Rk = 3.3K

I selected the components from the datasheet, based roughly around the highest voltage output for 180, with 470k plate, 1M g2, 1M g1,  3.3k rk

Does it matter that the plate voltage is now so low?  Is what I'm hearing the 470K resistor?

 

[rockinrob86]

Did you connect any resistor between g1 and gnd?

1M off of 68K input resistors

 

[PRR]

I think the plate resistor is too darn large. The 470K suggestions will give AM radio response, less than you want for guitar.

As-is, you could just cut the 470K in half, 220K. The 202V drop tends toward 101V drop, putting you about halfway up your 218V supply, a decent point.

 

[ruffrecords]

This is probably a measurement artifact. What meter did you use to measure it.?The problem is with a 470K plate resistor, if you probe the anode, the meter resistance will pull down the anode and you do not get a proper reading. As PRR says, 470K is rather a high value for a pentode plate resistor.

Cheers

Ian

 

[rockinrob86]

I have a fluke 27 meter.

Played around with it tonight, changed the plate resistor to 220k, cathode to 2.2k. Now I'm measuring 214 in, 85 at the plate, 1.7 at the cathode and 46 on g2.

It sounds way better and has a great level of grit. I changed the coupling cap to .005 from .01 and the bass is much better. I need to decide values for my tone switch, but I'm thinking maybe starting at something at like .0001 and then moving up in 50% increments.

It still seems like the plate voltage is low. I may play around with it tomorrow to see what happens if I go to 100k/3.3k or see what a 4.7k cathode resistor does.

 

[rockinrob86]

I finally had the time to figure this one out, now it sounds killer!

Instead of trying to do the math, I realized I just needed to mess around with it, so the first thing I did was assume they did the load lines with g2 at 100v for a reason. I lowered the screen resistor until it was around there and then I took a look at the cathode resistor, halved it and saw that made the screen voltage go in the wrong direction, so then I went up to 3k I believe. Next I ended up lowering the plate resistor, which gave me over 100v on plate and screen - and now it sounded really great. All the high end came back. It is extremely touch sensitive and alive sounding, very cool amp.

I forget exactly where I ended up, but I think it is 100K plate, 470K screen and 3.3K cathode

I am a fan of really cranked up/fuzzed out little amps and this pentode channel on this amp gets super fuzzed when cranked but a lighter touch or lower guitar volume cleans it up great, without lowering the apparent volume much at all!

So my tips for future pentode builders: do what you have to do with the screen resistor and cathose resistor to make the screen around 100, and then manipulate the plate resistor to taste (and over 100vdc). Alot of these other designs in my amp gave me weird screen and/or plate voltages of 30-60vdc. The tube still made sound, sounded like a tube amp and distorted when I turned it up, but the frequency response was a little weird and overall it was a bit "meh".

Thanks for the help!

I made a quick little demo of it https://www.youtube.com/watch?v=3c-rNUtqCDw