EL34’s in a 6L6GC amp

[imgumby001]

The manufacturer states they have equipped their kit with enough bias range to accommodate both tube types, but doesn’t state weather or not the pin outs(ie pcb board) is ready to use both without modifications. As I understand the EL34’s will require pins 1 & 8 jumped together and tied to ground. Does any of this sound correct?

[vampwizzard]

which amp is it?

[imgumby001]

It’s a KLD jcm800 circuit, or that’s what they compare it too. I’ve been looking over the schematic for hours now and it showed pins one and eight tied together on the pcb. I guess I answered my own question, or at least thought I did, and took the plunge and fired it up. Sounded awesome!! At least until the tubes started red plating. And that was with the lowest possible bias current!! The amp has 6.3v heaters so I figured I was ok on that. Just have to figure out my on going red plate issue and the el may be the sound I was searching for!!!

[vampwizzard]

Well what are you biased at? 6l6's are 30 watts and El34s are 25.

[Latole]

  As I understand the EL34’s will require pins 1 & 8 jumped together and tied to ground. Does any of this sound correct?

Yes.
Just read 0 ohms with ohmeter between pin 1 and 8 and you'll know you can use EL34
Pin 8 is alway at ground on these class AB amp.

Safety ; amp must be disconnect from the wall outlet and Stanby Switch to "play " position. This will discharge filter caps. Wait 5 minutes.
There is no voltage at pin 1 and 8, just in case you make a mistake with pin 3 ( 450 volts)

[Latole]

It’s a KLD jcm800 circuit, or that’s what they compare it too. I’ve been looking over the schematic for hours now and it showed pins one and eight tied together on the pcb. I guess I answered my own question, or at least thought I did, and took the plunge and fired it up. Sounded awesome!! At least until the tubes started red plating. And that was with the lowest possible bias current!! The amp has 6.3v heaters so I figured I was ok on that. Just have to figure out my on going red plate issue and the el may be the sound I was searching for!!!

With red plating, amps always sound awesome until tubes or fuse burn; Be sure amp have the right fuse, if not you may burn PT
6.3 volts heater have nothing to do with bias.

1- Remove all Output tubes.

Amp power ON:
2 - Read DC negative voltage at pin 5 on each tube ( voltage when tubes redplating)
     You may have too low negative bias voltage

3- Adjust bias pot or bias circuit resistors to have more than  -35 dc Volts . JCM800 50 watts schematic show -57 volts

Put tubes in;
4- Read tubes plate current ma or mv and adjust bias for the right tubes power dissipation. 60% of 25 watts ( EL34)

[jammied]

He has this amp

[jammied]

And this is the voltage chart from the point to point kld jcm25. Using same components and transformers.


Posting this voltage chart because it is a 25 watt amp and the factory recommends -30 on pin 5.

[Latole]

And this is the voltage chart from the point to point kld jcm25. Using same components and transformers.


Posting this voltage chart because it is a 25 watt amp and the factory recommends -30 on pin 5.

Same grid negative voltage for 6L6GC and EL34 !!!!
6L6s are under biased or EL34s are over biased

[jammied]

Yes .




Need to get pin 3 voltage and then take ma readings across 1ohm resistor. And set the plate dissipation.

With the amp completely warmed up.



With el34's that transformer and circuit will be putting out about 375v on pin 3 when across the 1ohm resistors will be 45ma to 46ma. Around 70%

[Latole]

Yes .




Need to get pin 3 voltage and then take ma readings across 1ohm resistor. And set the plate dissipation.

With the amp completely warmed up.



With el34's that transformer and circuit will be putting out about 375v on pin 3 when across the 1ohm resistors will be 45ma to 46ma. Around 70%

+1 for the 1 ohms
I don't like 70% , at high volume, tubes can red plating .
Il go not more at 60% and less with ear test

[jammied]

Well I will throw in a fascinating fact.


If you look at the early from 1995 hot rod deluxe service manual it shows to set the bias pot for -50v at c- test point.


Now if you look at the recent service manuals for the hot rod deluxe it shows to set bias at the 1ohm resistor to +60mvdc. And no mention of the c- voltage


And all years of the hot rod deluxe have the 1ohm resistor as far as I have seen.

[imgumby001]

After breakfast I’m going out to get this straight! I think 60% is what I should shoot for at this point. I tend to run amps on 11, so it seems a better safe than sorry strategy is what I should employ...

[imgumby001]

So, I get 368.3vdc plate to cathode on v4. When I plug that into robrob’s calc I get 40.7mA for 70%. I can tell you I’m almost certain that if I run that I’ll red plate these, or any tubes, that I run in this amp, and I just can’t explain why.

[jammied]

Drop it to 35ma


The signal is driving the bias hotter when played.


The hot rod deluxe is running 6l6gc's at 30 about 40-50%


368v and 40.7ma is 60% with el34's

[jammied]

Ok just put some jj el34's in mine. Same power transformer and bias system as yours and I get 380v on pin 3 and I went to 44ma across the 1ohm resistor. So that's about 67% and no problems.


You seem to be getting alot of voltage sag? Maybe a bad diode in the power supply or bias circuit?


One of my diodes went bad in the power supply and I swapped to a self contained full wave bridge rectifier.  That happened to have fred diodes..


Not sure if that is the difference?


Or maybe my JJ's are worse out

[imgumby001]

How do I know which diode? Just be arbitrary and pull em both? I’m at 395.6vdc plate to cathode and approx 17.5 mA bias, which reads -37.3 pin5 to ground, and its just barely getting cherried in the corner of the plates at full vol, full gain. That’s almost no bias adjustment from all the way cold on the pot.

[jammied]

Dont know.


Before the 6p3s were working and not red plating.  Now you say they are red plating . Along with the other tubes you try?


Maybe retention the pin sockets?


Better to not change anything yet. 


Just interesting to me 


Better let the pros take over

[imgumby001]

Just ran the diode test as indicated on flukes site and all, and I tested all the diodes in the amp not just the bias circuit pair, tested good according to that test. So I don’t know.

[imgumby001]

One of the first things I did was check the tension on the sockets. They’re hard to adjust, as they’re soldered into the pcb. Friggin pcb...

[jammied]

But as I believe it was 2deaf that stated that the 3.4k primary running at 25watts with 6l6's or el34's could be the problem?


Mine is running a 8k primary output transformer.  Maybe why I dont get red plating?

[shooter]

I get 380v on pin 3 and I went to 44ma

works out to ~~ 8k load so good.

395.6vdc plate to cathode and approx 17.5 mA bias

works out to ~~ 22k load, so bad

[sluckey]

Better let the pros take over

The pros may be a bit weary after that other 200 post ramble about the same thing.

[jammied]

Better let the pros take over

The pros may be a bit weary after that other 200 post ramble about the same thing.

I apologize sluckey.


Just trying to figure out why he is having red plating and I never have. And I even tried the stock 3.5k transformer.

[jammied]

I get 380v on pin 3 and I went to 44ma

works out to ~~ 8k load so good.

395.6vdc plate to cathode and approx 17.5 mA bias

works out to ~~ 22k load, so bad

Dont know if it means anything or will help? But I measured the ohms on the primary of the kld 3.5k transformer. 


Red to yellow is 112.3ohms
Red to green is 110.6 ohms
And green to yellow is 221 ohms

[Latole]

Well I will throw in a fascinating fact.


If you look at the early from 1995 hot rod deluxe service manual it shows to set the bias pot for -50v at c- test point.


Now if you look at the recent service manuals for the hot rod deluxe it shows to set bias at the 1ohm resistor to +60mvdc. And no mention of the c- voltage


And all years of the hot rod deluxe have the 1ohm resistor as far as I have seen.

To do a bias, it is not put a minus voltage to pin 5 of a 6L6GC.
It is feeding the tubes with the right cathode / plate current by using the bias circuit minus voltage to adjust for the right cathode / plate current at 60 or 70% tubes dissipation.

One 6L6GC brand may need -50 volts for, lets say 36ma, other 6L6GC brand may need -60 volts for same 36 ma.

[imgumby001]

Jammied don’t worry about it bud. I really do appreciate all your help. Slucky’s right. Nobody is going to help me at this point as I clearly don’t have enough to skills or knowledge to help myself, and holding my hand on this deal is exhausting. I’m going to call it on this amp and move on to something simpler. After I get a few simple circuits right perhaps I’ll have enough experience and knowledge to dive back into this on and figure out what I did wrong. For now it’s going out to the shed. Thanks again for all your help.

[sluckey]

I apologize sluckey.

Don't apologize to me. I'm not a pro. I'm just a hobbyist.

[jammied]

Imgumby001 no matter what you decide with this project. My personal thoughts are a cheapo digital meter that reads ohms, volts and mili volts would be money well spent.


Shooter's calculations show you have 22k resistance. So maybe start checking the output transformer primaries and secondaries. Maybe speakers and so on.


But I'm a caveman my real name is Ugg from the Ah clan   so I caveman things

[HotBluePlates]

I get 380v on pin 3 and I went to 44ma

works out to ~~ 8k load so good.

395.6vdc plate to cathode and approx 17.5 mA bias

works out to ~~ 22k load, so bad

It doesn't work like that.

The math done "380v/0.044A = ~8600Ω" is meaningless because the tube is not a resistor.  If we measured 380v across some unknown resistor and somehow knew 44mA was passing through it, we could discover it is a ~8.6kΩ resistor.

Same deal with "395.6v/0.0175A = ~22.6kΩ."


What's happening is screen & bias voltages applied to the tube result in some amount of plate current (and the exact current is unique that the individual tube-sample).  When a different screen/bias voltage is applied to the tube, some other plate current results.

The difference in plate voltages for the two plate currents is indicative of the "impedance of the power supply" + "the d.c. resistance of the output transformer" but is otherwise probably not too informative to us in this situation.

     -  That is, (395.6v - 380v)/(0.044A - 0.0175A) = 588.7Ω = Supply Impedance + OT DCR.

[shooter]

the tube is not a resistor

If there's an AC signal, I agree, but just idling i'm thinking V/I works in a "working" system
I believe G2 IS playing a part here, hence V/I not working in a "not working" system

[HotBluePlates]

the tube is not a resistor

... just idling i'm thinking V/I works in a "working" system ...

Let's assume you're right.  What do you do with that information?  How is it useful to you?

My suspicion is it will be difficult to find a way to use it, since it doesn't correlate with the tube's internal resistance nor any static/dynamic behavior of the tube.

Back to the thread's topic, it seems one of the tubes is red plating.  If the bias supply is already outputting the most-negative voltage it has (and that voltage seems reasonable for the tube-type), it's probably just a bum tube.



I have a pair of apparently American-made 6L6WGB's that IMO are defective.  When put in a test circuit, they pass double the plate current of the other 30-40 6L6s I have.  I would never modify an amp to be able to use these tubes, because I'd have to modify it back to get the next pair of 6L6s to work.

But as I believe it was 2deaf that stated that the 3.4k primary running at 25watts with 6l6's or el34's could be the problem?

Mine is running a 8k primary output transformer.  Maybe why I dont get red plating?

... I measured the ohms on the primary of the kld 3.5k transformer. 

Red to yellow is 112.3ohms
Red to green is 110.6 ohms
And green to yellow is 221 ohms

The dynamic impedance (3.5kΩ) is not relevant to the idle/static bias.  Each output tube simply sees ~110Ω between its plate & the power supply.  The 3.5kΩ or 8kΩ only comes into play when there's an a.c. signal at the plate due to an a.c. signal at the control grid.

[PRR]

I get 380v on pin 3 and I went to 44ma

works out to ~~ 8k load so good.

That may be a good start in Single-Ended.

In Push-Pull (isn't that where we are?) there's at least a "2" for two tubes, but also we don't have to cover the whole wave in each tube so there is 1.1 to as much as 3 correction from idle to full roar.

[jammied]

Ok I'm trying to be quiet. 


What would the recommended troubleshooting method be with this amp?


From what I understood (and could be completely wrong) . The problem stayed with the tube originally. Then a closer matching pair was installed. Along with biasing the 6p3s tubes within range and the red plating went away.


Now after the original poster installed el34's the red plating returned. And now will red plate even with the 6p3s tubes.


So I'm confused?!


Where should the troubleshooting begin?

[shooter]

begin at reply #31, below the ____________   


And I did slip on the SE peel 
staying on that peel, paraphrasing; what good is that "information"
It's a real quick "check" to see if the system is go/no-go (assume setup of NO signal)
probably holds water in PP also (assuming 1 ohmers and a good meter N no sig)

[HotBluePlates]

... What would the recommended troubleshooting method be with this amp? ...

IMO: tube tester.  Or get a different pair of output tubes (or even a single to match the remaining good-single).

... The problem stayed with the tube originally. Then a closer matching pair was installed. Along with biasing the 6p3s tubes within range and the red plating went away. ...

I can't follow much of anything going on either.  But "if some tubes redplate, but others don't" there's not necessarily an amp-problem to fix.

Depending on the individual tubes, it might be possible to modify the bias supply to get them to work without red plating.  OR the individual tubes could have some internal malfunction leading to red plating.  OR they could be so far from "An Average EL34/6L6" that it's just not reasonable to modify the amp to get them to play nice.

Having tested a lot of tubes with a tester that delivers just about "all the info" on a tube, I can tell you they all vary pretty widely around some "average" and those average values get printed on the data sheet.  This is why even the companies that tout their matching group together pairs/quads/etc, but then have 3-10 categories/ratings to assign to each pair/quad.