Cathode biased EL34 PP question please

[steelerfan88]

 I am basically building a PP/cathode biased Lite IIB circuit using EL34 power. The PT Set I have on hand is the Classic Tone 50w Marshall set with choke.
#40-18089 PT specs:
312-0-312 @ 150ma
6.3v @ 5A CT
5V @ 3A

#40-18025 OT
3.6K Primary w/ 4/8/16R secondary

#40-18058 choke
3hy 250ma

 As I had mentioned this is a LiteIIB layout using the above trannys. I am wondering if I would be in the ballpark of things going with a 250R/10w bias resistor OR would a 180R/10W (with a 50uf/50v bypass cap) be a better starting point. I am shooting for 19w-23W output range with a good classic  rock crunch.....some clean on tap would be nice as well. I have already built the PI to 470R/22K flanked by two 470K's (this worked well in my 6v6 version I built) I realize a 6.6K load would have been nice but I already have this set of iron on hand and wish to use them. Any and all suggestions are GREATLY appreciated 

[sluckey]

Try and see. It'll work with either. Maybe look at some established PP EL34 cathode biased amps such as Matchless for some ideas. Matchless is fond of using separate 270Ω for each tube in their 30 watt amps. I like the idea of separate cathode resistors because that allows you to see how each individual tube is performing.

[shooter]

I do mostly cathode biased and Sluckeys right, with individual cathode R's you can mix-n-match both unmatched and other type tubes.  my pa section is a clone of the matchless clubman, my amp tester say' it's a great rock-n-rollers' amp.  I ran kt88's, 6L6 and el34 during proto-type phase n the 34's fit best sound wise

[PRR]

> 312-0-312 @ 150ma
> 3.6K Primary

The 3.6K load and self-bias leads to 250mA and 100W idle (50W/tube) at like 40W-50W out.

I did similar with 6.6K load and was quite happy.

> #40-1825 OT
3.6K Primary w/ 4/8/16R secondary

?? Maybe 40-18025 ?

Put 8 ohm load on the "4" tap and it will be more right. Downside is this is twice the iron (cost, weight) you need for a 25W.

[steelerfan88]

I think I read where someone said 250R worked well under a pair of EL34's with about 400vdc on the plates. I figure I will be around that same number using my 312-0-312 PT with diode rectifier (400v-420v). I'll follow what another had done with the 250R and see how it sits like that .....easy enough to adjust from there👍

[HotBluePlates]

I think I read where someone said 250R worked well under a pair of EL34's with about 400vdc on the plates. ... I'll follow what another had done with the 250R and see how it sits like that .....easy enough to adjust from there

You may want to install a higher value resistor (like 300Ω) just in case the EL34's you use tend to idle hot.

Then you can parallel another resistor to lower the effective cathode resistance.  For example, 1kΩ in parallel with the 300Ω resistor gives you ~231Ω total.  That might give you an easy way to trim the bias, without having a bunch of values between 200-300Ω on hand.

[shooter]

250R worked well

I wound up with 270R, my B+ is about 370, can't find my volts sheet though, HBP is correct, start bigger n work your way down, beware that as you change R's, re-check all your volts, plate screen, cathode

[steelerfan88]

I think at 250R....I am going to be plenty safe....if anything it will be moderately biased. The JjEl34 valves are pretty tough and I am not going to have a very high B+ to start with. I have built this same circuit using Vibrolux iron with 5881 valves and ended up with a 300R at about 95% disp. at idle which is just about perfect. That PT was a 340-0-340 Spec.....I am using quite a bit less on this build at 322-0-312....we shall see though.
 Another question if I may : generally speaking, where do you typically like your screen voltage to be in relation to the plate voltage ? A few under ? 10-20v under ? Or is it really  ok as long as the screens don't become a secondary plate (that's bad !).AND.....when referencing screen voltage in a cathode biased amp, are comparing the screen readings to the actual plate to ground reading OR the plate to CATHODE reading....there is a difference and what some consider the plate voltage may be something different than what another might interpret it to be.

[PRR]

The screen should be the value to support peak plate current.

There is NO fixed relation, except that "most" audio power pentodes were designed to work OK with specific loads and Vg2 slightly less (to allow for filtering) than Vp.

If you are going beyond typical specs, you may find that Vg2<Vp is enough, or that you need Vg2>Vp (awkward!).

The screen will not SUCK until plate voltage is below 20% of Vg2. This is NOT normally any concern.

[steelerfan88]

Any harm in running the amp at 3.6k load OR.....replacing those EL34 tubes with 6l6/5881 tubes to see how they sound ?

[steelerfan88]

My voltages in the power section after start up :
B+ 426dc
Plate to cathode 397dc
Screens 382dc
Cathode R value 250R/25w
Drop across the cathode "R" 27dc

These were taken with a 6.6k (approx.)load but my OT IS A 3.6k...I simply plugged an 8R speaker into the 4R out jack to get the higher impedance of roughly 6.6K. I sampled the sound of running the amp at the 6.6k load vs. the 3.6k and like what I was hearing just running the 8R speaker into the 8R output (3.6k load) and like that tone much better.....sounds less compressed and more open/airy sounding by far

[shooter]

looks like you're there, my ballpark math says about 21W per tube plate diss.

how's it sound?  are you using NFB?  I typically don't on my PSE or pp cathode biased.

[steelerfan88]

No NFB......the amp sounds pretty good however the low E and A strings are producing a slight buzzy overtone when overdriven. I need to solve this issue then I think Ill be good to go.  My PI is at 470R with a 10k tail resistor flanked with two 1M resistors....same as a Plexi.
220k grid leaks......V1 plate is set with a 100k. V1 is a parallel triode. Cathode resistor on V1 is 820R with a 4.7uf bypass. Any ideas on how to better clean up the overdrive and remove that annoying raspy/buzzy overtone ? It's not real bad but enough to bother me though

[shooter]

You'll have to seek higher counsel, I'm not a player so E, A, I guessing the fat ones
My version uses a single input, like the plexi 6V6, so each 1/2V1 has it's own R n bypass cap, my PI is an inter-stage tranny, no tube.

[steelerfan88]

Hopefully Sluckey or PRR can shed some light on the subject...

[shooter]

if you say out of overdriven, how is the bass response?

if it's open, you might try adding something like the 100k, r136, in the attached image.  I believe it's used to "tame" drive into the PI

[steelerfan88]

Well...after a weekend of tweaks and try's I have a great sounding amp 👍
I ended up settling on biasing those EL34's with a 195R/100uf combination. This seemed to tighten up the tone and add better clarity.....especially when driven harder. The cooler bias network I had originally was 250R/50uf and was just not as focused. I settled on a 1.5k/2uf for V1 parallel triode with 82K on the plates. Here again, this seemed to provide a little better clearity. The PI is set 470R/20K flanked with two 1M resistors. Grid leaks are at 330K each now . Power supply is 44uf/44uf/22uf with a 3hy/250ma choke between the plate and screen anodes. One of the changes I made that really tightened up the sound and helped get rid of the buzzy overtones was to simply switch cabinets. Originally I was doing all of my testing with a semi open back 2x12. After finally exhausting all my other tweaking , I thought that maybe the cabinet choice is not ideal. Once I switched to my closed back 2x12, the annoying buzzy overtones all but disappeared....go figure 👍
With the bias network changed to the 195R/100uf, I am dissipating 23.5 watts on those EL34's now at about 95%. The clean to crunch ratio should be just about perfect for gigging. Being cathode biased, she lifts up her dress earlier with less effort and gives up the candy 😊