more headroom possible?

[daz]

FWIW I get really good results using el84's at 365Vp and an efficient speaker (V30 in my own combo). I'm using a sort of cathode bias with a diode clamp on cathode voltage rise right after the onset of clipping. I'm also using a 160R cathode resistor and dissipating 11+W per tube at idle. So a 130R resistor at this Vp is probably drawing excessive current. AND...

I've also had good (but different tonal) results using el84's in fixed bias with the same PT. Because the tubes are biased cooler in fixed bias the plate voltage goes up to the 380 range but I've never had any trouble with tubes because of it. Some Peavey's, Mesa's, Traynor's, etc. run el84's at 400-ish Vp. I've owned a couple at one time or another and never had a problem with tubes. I mean, they don't last for a year or more like more conservatively operated tubes can, more like four to six months. But IMHE they don't fail abruptly or blow up. Run at higher voltages with fixed bias el84's take on a different character from the smooth/chimey VOX thing they're famous for. They really seem to grow a pair and deliver some impressive power. My usual experience with el84's in combo amps is that they become microphonic even before you can wear them out anyway. They're especially bad about this. Since I love this tube I've always taken the motto "Smoke 'em if you got 'em." Fortunately they're still the most affordable tube type.

I've been looking for a higher resistor in my parts boxes. Been reading a lot today and i did get the impression 130R is too small. So i thought i had one thats around 170R but i couldn't find it so i will have to figure out some sort of combination in parallel or series depending on what i have. I have a lot of those metal encased chassis mount ones but many are the same. Some cement too. So i'll see what i can come up with. I do have a ton of 226R ones if that value isn't too high.

[daz]

some prefer this Eminence version anyway and the price is right (I bought one) https://www.ebay.com/itm/12490339636...4AAOSwc1hgwQgC

I was thinking of trying one of those several years back but never did. Much better deal than the 12L.

[daz]

I've been looking for a higher resistor in my parts boxes. Been reading a lot today and i did get the impression 130R is too small. So i thought i had one thats around 170R but i couldn't find it so i will have to figure out some sort of combination in parallel or series depending on what i have. I have a lot of those metal encased chassis mount ones but many are the same. Some cement too. So i'll see what i can come up with. I do have a ton of 226R ones if that value isn't too high.

Hmmm... tried one of those 226R's and it sounds fine and doesn't seem to lose volume. Now i just have to see what cap works best. tries 100, 50, now 22uf. I think the 22 sounds better except seems to have less lows. Some 'sperimentin' to do....

[Chuck H]

Hmmm... tried one of those 226R's and it sounds fine and doesn't seem to lose volume. Now i just have to see what cap works best. tries 100, 50, now 22uf. I think the 22 sounds better except seems to have less lows. Some 'sperimentin' to do....

I've noticed that getting over the 160R value with my Vp of 365 seems to introduce more crossover distortion on cleaner tones. And I really don't prefer that personally. Otherwise I would be trying to idle cooler just on principal. Some "ears" actually want some crossover distortion in their tone or it doesn't sound "right" to them. Your ears are your own. Cathode biased el84 amps typically idle at 100% dissipation for AB1 circuits to compensate for voltage rise at higher conduction levels actually cooling the bias. And class A circuits are often (usually) over 100% at idle for this tube. So I guess I'm not alone in my preferences tonally. You'd think that with cathode bias (sometimes called 'self adjusting') there would be more leeway in the absolute circuit values, but that's not my experience with ear testing.

[daz]

could well be but it's halloween and kids coming to the door and all so i haven't had but a few minutes to play it like this. So i will have to wait till tomorrow to see if i notice that, tho i DID notice a thin-ness in the plain strings compared to the 130. I'll try and combine a few to get close to 160R but i don't think my plates will be 365 like yours. Mine was there with the 130R but the 226R it's in the 390's. so i'm guessing 380v or thereabout with a 160R.

[Mick Bailey]

The MESA Studio 22 runs a pair of EL84s in fixed bias and my customers who had these haven't reported excessively short tube life. The ones over here in the UK run at close to 400v on the plates of our mains. For even more clean headroom I used to fit EL84M tubes until they got ridiculous in price. They will withstand higher plate voltage than a regular EL84 and don't break up so early.

Why not make the bias switchable between fixed and cathode bias? That way you get to properly A-B the results and give more tonal flexibility.

[pdf64]

The MESA Studio 22 .

Have you investigated the OT primary impedance?

The 5k6 HT dropper to the screen grid node may be significant in reducing stress on the output valves https://el34world.com/charts/Schemat...ogie_22cal.pdf

[daz]

The MESA Studio 22 runs a pair of EL84s in fixed bias and my customers who had these haven't reported excessively short tube life. The ones over here in the UK run at close to 400v on the plates of our mains. For even more clean headroom I used to fit EL84M tubes until they got ridiculous in price. They will withstand higher plate voltage than a regular EL84 and don't break up so early.

Why not make the bias switchable between fixed and cathode bias? That way you get to properly A-B the results and give more tonal flexibility.

I wanted fixed bias and tried to get the fixed bias to work with the 84s but for some reason i could not get negative bias to the tubes. I kept lowering the values of the resistors ut it would never go into negative territory for some reason. Had i been able to then i'd have chosen the better sounding bias scheme but i'll just have to stick with CB. Worked fine with 6v6 but i can't figure out why i can't get it in range with 84s.

EDIT: It showed neg V in the bias circuit but it never gets to the tubes. Now that i think about it, i wonder if i forgot to move the wire from the juntion of the 220ks at the grids. I know i did at least once or most of the 3 or 4 times i tried it but maybe the times i changed the R values in the bias circuit enough to get there i forgot to move it from ground to the bias supply. I'll have to try again today.

[Mick Bailey]

The 5k6 HT dropper to the screen grid node may be significant in reducing stress on the output valves

Agreed - limiting screen current in EL84s prolongs their life, especially with new production tubes running a high plate voltage.

[pdf64]

Perhaps you’re using a full wave bridge rectifier, the HT winding doesn’t have a CT to circuit common?
If so the feed to the bias supply needs to be capacitively coupled.
Scroll down past halfway if that’s gobbledygook http://www.valvewizard.co.uk/bias.html

[daz]

Perhaps you’re using a full wave bridge rectifier, the HT winding doesn’t have a CT to circuit common?
If so the feed to the bias supply needs to be capacitively coupled.
Scroll down past halfway if that’s gobbledygook http://www.valvewizard.co.uk/bias.html

Naa, it has a center tap plus it worked fine with the 6V6's. I didn't just add the bias supply circuit, it was always there and working when it was 6V6. It's strange. No matter what i do it well not read negative voltages even tho the neg v is there are the bias point where it connects to the 220k junction of the grid resistors but it won't read anything across the 1R cathode resistors except positive v. And if i then adjust the resistors to go further in that direction instead of going negative it just stops at zero and never goes into negative MV no matter how far i adjust the resistance in that direction. But it does go negative at the grid resistor junction

[Helmholtz]

The 5k6 HT dropper to the screen grid node may be significant in reducing stress on the output valves

It certainly helps to start with a lower screen supply voltage.
But considering that the reaction time of the screen node is around 1s, I don't expect much limiting of screen current peaks.

[Mick Bailey]

My own preference is to install 2k2 screen resistors to give more protection against current peaks. The Studio 22 circuit would be better in my view with higher value screen resistors - the present arrangement appears to be more in line with protecting against average screen current, but again, the tubes last a good while in these amps just as they are.

[daz]

Man, this is really one of those STRANGE issues that has me scratching my head till i bleed. So it was 6V6 fixed bias till the other day when i decided to try el84 using octal to 9 pin sockets so i can fit el34s. When i did i could not get a nefgative bias reading so i tried them cathode biased. But i kept re soldering it back to fixed to try different values, even copying other fixed bias el84 schematics. I finally gave up but then i had the idea that if something went wrong i could put it back to how it was with the 6v6's and try it. Sure enough, now the 6v6s back in and the entire bias circuit back to how it was, no negative bias. So now i know it's something that went wrong. But what? I checked the caps in the bias circuit, the bias pot, - voltage is there in the bias points and st the 220k grid junction. But measuring across the 1R cathode resistors as i always do will not show negative voltage. If the bias pot is turned all the way till is shorted to ground it read zero. If i turn it the other way it goes up into positive voltages. The 1 R cathode resistors are al always tied together and a wire from that junction to ground and i have check all that for continuity. WTF ?!!! I can't see any possible way this can be happening. Anyone have any ideas?

[daz]

heres a schematic tho the 10k into the bias pot is actually 22k. I think i changed it from 10k when i converted to 6v6 but never changed it on the schematic. https://music-electronics-forum.com/...etch?id=970056