touch sensitivity

[daz]

Ok, so i went back to diode rectification and it's now at 441 on the plates once i adjusted to the same bias current.Definatly better as far as touch, tho that sag isn't there like it was. i think i may prefer this tho. Can't use a sag R because that will just take the voltage down like the tube did. I'm leaving it like this for now at least. Oh, and also pin one and 8 are not together again at long last.

[daz]

I was looking at the 1st tube on the pre amp, the cathode resistor is normally 1 meg not 820 ohms, that value normally shows up on the 3 rd stage as shown

I think thats a typo. No one uses a 1 meg cathode R ! And 820 is quite normal even tho marshall generally uses higher. certainly not 1 meg tho. Not even sure if the amp would work like that.

I don't know about a 1MOhm cathode resistor being normal.

They don't do anything to the amp's tone. They are just there for reading current. I'm surprised you never heard of that, as it's a trick mentioned in about every biasing article online. If you read the voltage across the 1 ohm R's you get a small millivolt reading that represents the plate current. tho you have to check scree R V drop and subtract that. So say you read 40 v across the 1 ohm R and a voltage drop of 5v across the screen R. 40 minus 5=35v which works out to the same in current, IE: 35ma. If i have anything wrong here, someone please tell me. But i'v read this and asked to be sure and was told thats indeed the way to do it.

I checked a few of the earlier Marshall schematics, and none of them seem to show grid stoppers on the power tubes, whereas yours has them at 5k6. Might be worth reducing the value / removing them altogether to see if that helps.

The 2204, one of marshalls most well known and revered amps has tham at 5.6k. I think i may have tried bypassing them once, but i have tried so many things at this point i can't recall. I give that another tyry tho since it's only a matter of clipping a couple clip leads on. EDIT:tried bypassing them, sounds better with the 5.6k's

Have you tried different tubes?

Yes, and these are the best ones i have and the ones i always loved best in my marshalls....the old EL34LS, the REAL LS tubes from the tesla factory before it closed down then was re-opened as JJ.

[Russ]

Pin 1

Referring to Dazs' schem, can someone explain why having pin 1 (suppressor grid) of the el34s connected to the bias supply instead of pin 8 wouldn't cause problems? Seems to me there'd be a big difference between hooking it up to -40some volts and ground.

This is the only schem where I've ever seen it wired like this and it's got me wondering what I'm missing.

[pdf64]

Hi daz, a tube rectifier has a voltage drop and an internal resistance, I think from the weber site that a gz34 is about 17v and 50ohms. Therefore a 50ohm sag resistor with silicon rectifiers would supposedly give the amp the same sag ie dynamic lowering of B+, without the 17V constant volt drop. What value sag resistor did you try? A 47ohm resistor will probably only lose 5V or so at no signal, rising to maybe 20 when it's rockin'.
Regards the 1MOhm cathode resistor, I was refering to R ski's post (1M = 1 million, 1k = 1000). 1 ohm cathode resistors on the power tubes are fine, I'm not aware that any guru has implicated them for bad tone! Peter

[daz]

Referring to Dazs' schem, can someone explain why having pin 1 (suppressor grid) of the el34s connected to the bias supply instead of pin 8 wouldn't cause problems? Seems to me there'd be a big difference between hooking it up to -40some volts and ground.

This is the only schem where I've ever seen it wired like this and it's got me wondering what I'm missing.

I can't tell you the technical reason, but i did it because i showed my schematic at the time to several forums and had at least one heated debate where one of the local gurus and several others said to do it that way.I don't recall their reasoning and i probably didn't understand it anyways. But if you look at my above post you'll see i changed it back this morning.

Hi daz, a tube rectifier has a voltage drop and an internal resistance, I think from the weber site that a gz34 is about 17v and 50ohms. Therefore a 50ohm sag resistor with silicon rectifiers would supposedly give the amp the same sag ie dynamic lowering of B+, without the 17V constant volt drop. What value sag resistor did you try? A 47ohm resistor will probably only lose 5V or so at no signal, rising to maybe 20 when it's rockin'.

I tried a few and settled on a 180 ohm because of the values i had handy that seemed to be the one that gave me noticable sag. Maybe i just didn't take enough time to notice the smaller values i tried, but i'll give it another try. I have a 40 ohm 25 watter i can try.

[pdf64]

Russ, the tone lizard has an excellent explaination / justification of this, see
http://www.tone-lizard.com/Marshall_Myths.htm
Scroll about 3/4 of the way down the page.
Interestingly, the tone lizard does criticise 1 ohm cathode resistors on the power tubes, as they are generally installed in line with the screen grid ground also, thereby contributing to secondary emission runaway. Therefore, keep the screen grid grounded when putting the 1 ohm resistorsbetween the cathode and ground. Unless you try his preference of tieing the screen grid to the bias supply. Peter.

[daz]

Ok, as i said i went back to diode rectification, tied pins 8 and 1, and now tried a 40 ohm sag R. Bingo ! thanks all....it sounds even better now and the touch is back ! Oh, and thanks for that Peter....i haven't read it yet but i'll install a switch to bypass the 1 ohms. I'll put it INSIDE the chassis, as i will never need to use it except when inside.

Amp is sounding and FEELING fabulous ! Thanks !

[Bruce / Mission Amps]

Connecting the suppressor grid to a negative voltage supply (bias voltage) was an old K. O'Conner trick promoted in one of his earlier books to help with tube runaway and to be honest, I think it makes every EL34 amp I've tried it on sound a little sterile or hard... this is very subjective but I wouldn't bother with it myself.
Series 5k6 ohm resistors on the grids is OK for me but the 1 ohm bypass switch is totally silly.

[daz]

Not silly if you value peace of mind. Has nothing to do with whether or not it's necessary. But if 2 people who are both quite knowledgeable like yourself can argue over things like this, who am i to decide there is one definite answer? Consider this..... someone like myself who can build an amp but doesn't understand enough to easily get it sounding fantastic cannot just get a answer to each of the endless questions I have. Throw into the mix the fact that many of them seem debatable between gurus like yourself, so how am i to decide who to listen to on 100 different design details?! I'd be at this for the rest of my like and never get a chance to play the amp !

So while i didn't add a switch, i AM adding a wire from both tubes to ground to bypass them so that i can simply unsolder when i need to use the resistors to check bias, something i will rarely do once the amp goes into it's new home (cab) that will be finished with this weekend. But you haven't gotten rid of me yet.....i'm building another one after this, so lookout !

[daz]

Hmmm.OTOH, thinking about it that would change the bias, wouldn't it? Guess i'll leave that bypass wire off.

[Shea]

Referring to Dazs' schem, can someone explain why having pin 1 (suppressor grid) of the el34s connected to the bias supply instead of pin 8 wouldn't cause problems? Seems to me there'd be a big difference between hooking it up to -40some volts and ground.

This is the only schem where I've ever seen it wired like this and it's got me wondering what I'm missing.

It doesn't cause problems. It works. I've done it before. Some people say it increases plate current a little, all other conditions being the same. I guess the idea is that the suppessor grid will do more of what it's intended to do -- i.e., repel stray electrons back toward the plate -- if it's connected to a negative voltage instead of ground.

IIRC, Traynor wired the sockets in their EL34 amps like this.

Shea

[Shea]

The way it's shown in Daz's schematic, there isn't any filtering between the rectifier and the plate supply (which would deliver a really bumpy signal to the Plates).

Yes, there is. It's the first filter cap. Just because the capacitor looks like it's positioned between the center tap of the output transformer and the choke doesn't mean the cap comes "after" the center tap. It is connected to the center tap, so it's filtering the voltage supply for the center tap. He could have drawn the schematic so that the capacitor' location on the piece of paper looks more like it's in between the rectifier tube and the center tap -- perhaps if he moved it down closer to the rectifier -- but electrically, there wouldn't be any difference.

But the choke nearly always goes in between the plate supply and the screen supply, as others have explained.

Shea

[Steve Conner]

I remember me and Ken Gilbert getting into an argument over the suppressor grid thing years ago. I actually tested it, and it seemed to make practically no difference to the operation of the tube. If anything, the EL34 I tried worked a little better with the suppressor connected to the cathode than with -50V on it.

I don't see any point in bypassing those 1 ohm bias measuring resistors either. Tubes are high enough impedance that 1 ohm of resistance in series with any electrode might as well not exist. Stray inductance in the cathode circuit is another story, though. It can cause nasty oscillations, though grid stopper resistors usually stop them.

[pdf64]

daz, no need for a switch etc, just ground pin 1 screen grid directly and take cathode pin 8 to ground via the 1 ohm resistor. I doubt it makes much difference, I was just reporting the tone lizard's take on this issue. I recommend that site, just read that page if nothing else. Peter.

[tubeswell]

Yes, there is. It's the first filter cap. Just because the capacitor looks like it's positioned between the center tap of the output transformer and the choke doesn't mean the cap comes "after" the center tap. It is connected to the center tap, so it's filtering the voltage supply for the center tap. He could have drawn the schematic so that the capacitor' location on the piece of paper looks more like it's in between the rectifier tube and the center tap -- perhaps if he moved it down closer to the rectifier -- but electrically, there wouldn't be any difference.

But the choke nearly always goes in between the plate supply and the screen supply, as others have explained.

Shea

Yep thanks Shea we got that sorted out yesterday. It was the way the schematic had it drawn that fooled me. (not enough sleep lately I guess). I have already built an amp (5G9) with a choke between plate and screen supply so I know about that setup. Nevertheless I consider myself vindicated about my argument about Pii filters, which it turns out was something they used to do in the old days, which is why it shows up in Jack Darr's book, and that was what I mainly was raving about. ;-) Of course as Enzo had pointed out (which had slipped my mind at the time), the OT primary winding acts a bit like a choke in modulating the current to the plate. I wonder if you could replicate the smoothness of a Pii filter in this way by adding another beefy decoupling cap to the plate pin? (I have noticed that I can never get an accurate reading for the plate voltage on the amps that I have built - it always twitches back and forth about 1V at idle on my meter, but I was wondering whether that was just my meter - anyway its probably just academic).