Amp repair?!

[Dimi Pana]

I have an Epiphone Valve Junior Combo amplifier that needs repair. It used to play fine until I decided to swap in/out a few different tubes both at pre and power section. I suspect the repeated removal/insertion of tubes and retention spring must have done something to the socket. I simply cannot turn the amp on and start playing and in less that a few minutes there is a weird crackling noise, then some hiss/static noise and then silence. I then check the tube with a tester it's dead, both the heaters (pins 4 & 5) show no continuity. I have ..."lost" two brand new tubes so far, so I am not risking putting another in there unless I identify the problem. This is a very simple design the only thing that could be the culprit is a resistor and cap and of course the voltage that is provided to the power tube. BUT, there is absolutely no problem at all with the preamp tube (which I also swapped in/out many times) so I am kinda stuck. I can follow instructions, I am good at soldering, I have a decent DMM and tube tester (albeit borrowed from a friend), I understand electronic schematics although definitely not an expert. I believe I can do the troubleshooting with someone's guidance. It is simply not worth paying a tech to look at it, it will probably cost me more than the amp's value. I am also posting the schematic, however at this early point I am primarily concerned about asking questions in a Solid State forum about a tube amp. Let me know, I hope I did not break any rules. Thank you!

EDIT (1/23): I uploaded another schematic I found online, the tube section is identical but there are some "interesting"(?!) differences in the power supply. Indeed the transformer outputs 6.3 VDC for the heaters obviously after the bridge it should be a bit higher, so indeed the first schematic is wrong (we already know that). 

[phatt]

First you have to determine which schematic is your amp as I think there are a few variations.
The schematic you have uploaded makes little sense as you have heater voltage at transformer reading 6.3VAC then a full wave bridge and output reading 6.3VDC.
As that can't happen then that means one of those readings is wrong.

Here is a way to check;
If your circuit has that 8Amp Bridge driving into C12 (4,700uF/16V) then you have a DC heater circuit.
If so then set your meter to read DC.
Now with valves removed and amp powered on, read the DC voltage between pins 4&5. Then post your findings. Does not have to be exact,, +/- 20% is ok but 9 VDC will fry the heating filament in the valve.

If you do not have the DC heater circuit then it's obviously going to be AC so set meter to read AC volts and post those numbers.
Phil.

[Dimi Pana]

Phil -

Thanks for your quick and informative answer. I have version 3 of the Epiphone valve Junior which luckily has the DC heater circuit. Thanks for confirming that there's something strange with the schematic as you can see it's lifted from a venerable source but you never know right? I'll try to find a better schematic I'm almost certain there's one floating online for all different versions of the EVJ. I am at work now I will do the measurements you asked me when I get back home. Just to clarify you are asking me to do that with both tubes, not only the power tube that's giving me the problem. Also you want me to measure between 4 and ground and then 5 and ground or between 4 and 5 only? Finally I might be able to obtain an identical working amp from a friend-of-a-friend etc LOL which could be useful for comparison purposes. Much obliged for the help!

[phatt]

Yes pins 4&5 ONLY. no ground needed,, you just want to see if heater sockets are delivering the correct voltage,,Yes check 4&5 on both valve sockets to be certain.

It may be fine but as the schematic is wrong then it's a potential issue.

A couple of other more likely things to check while the power is on and no valves inserted is the Screen grid voltage at pin 9 of the power tube socket and the bias resistor.

 Screen grid check;
Set meter to highest DC voltage,,
With Black lead on chassis (or other convenient ground point)
Probe pin 9 of power tube socket with the red lead.
As long as you have around 300VDC then screen is safe if no voltage then I think that would cause a failure.
While you are there check pin7 it should be a little higher in voltage than the screen voltage.
That covers the 2 main voltages of the power tube. B1 and B2 on schematic.

** Safety Note, when probing high voltage connect the black lead to ground point *FIRST* **

Another test;
With power OFF and meter set to *Ohms*
measure the value of R14 is should be close to 220 Ohms.
(This is a Cathode biased amp and R14 sets the idle for the power valve)
If it reads dead short then unsolder C5 at one end then read the resistor again. If it then reads normal C5 is shorted out and that would burn out a power tube real fast. In which replace C5.

If you are new to this?  Things to be aware of around live circuits.
Remove; Rings, watches, necklaces. Work with your left hand away from chassis. it is tempting to hold chassis with left hand while probing but if you do come into contact with high voltage that voltage will travel through your heart,, hence right hand working is safer.
Don't work in bare feet. And don't work on it after a long drinking session.

Hope it helps,,Phil.

[Dimi Pana]

OK, sorry for delaying to do what you asked me, I am digressing, but we expect snow and subzero temps in Chicago as of later today so I decided to stay home and prepare the "fort" (salt-check, snow blower-check, electric generator-check, etc, you get my point).

But I also managed to find the time and took the measurements, so here it is.

I first checked R14 (did that BEFORE even powering up) and it read 217 Ω, which I think is good (so I did not mess with removing C5) .

Then, with 125 VAC coming from the grid here's what I get:


Preamp tube

Probing DC pin# to ground:

Pin 1 ---> 345 V

Pin 2, 3, 7, 8, 9 ---> 0 V

Pin 4, 5 ---> 9.30 V (!?)

Pin 6 ---> 345

  *Probing between pin 4 and pin 5 ---> 9.45 V (?!? but read comment below about the heater pins)


Power tube

Probing DC pin# to ground:

Pin 1, 2, 3 , 4, 6 and 8 all show 0 V

Pin 5 ---> 9.43 V

Pin 7 ---> 385 V

Pin 9 ---> 384 V

 *Probing between pin 4 and pin 5 ---> 9.61 V (hmmm ?!?)

BUT wait !!!

With time (as circuit warmed up more?) the heater pins voltage also kept going up in both tubes!?! I watched it go all the up past 9.85 V and eventually powered-off at 10.05 V fearing something is (likely) wrong and it was still rising!? Think we found the culprit?

Thanks again for all the advice, waiting for your diagnosis and possible remedies. I hope this is an easy (and not so expensive) repair! 

- Dimi.

[phatt]

Hi Dimi,
OK the EL84 Plate (pin7) and Screen (pin9) are ok.

Cathode resistor is fine.

But Oh dear,, Ouch!! yes my fear was correct.
The heater is way over voltage. 

The heater filament is designed to run on 6.3 Volts (AC or DC)
The problem is that there is only a +/-10% tolerance for filament voltages. Outside of that you are shortening the life span of the Valves.

So 6.3v + 10% = 6.93v max,,, so 7 Volts is the outer limit for safe operation. At 9Volts you are striping the cathode rapidly until the filament gets too hot and burns out. (your comment that you had no continuity between pins 4&5 on the power valve confirms the heater has burnt out)
Bear in mind that those voltages you post are *No load* readings and will be lower with valves inserted,, but likely heater is still too high.

If you pull the 2 filament wires from the pcb and read the AC voltage at the transformer output it should read somewhere around 7Volts AC with no load.
That will tell you that the filament voltage is correct for AC heaters but NO go for DC.
Adding a Full wave bridge rectifier adds 1.4 times to the AC Volts.

So 6.3VAC x 1.4 = 8.82VDC,, So yes some clueless designer screwed up badly.

You could add 5 watt resistors to drop the voltage or Remove the rectifier and run the heaters on ACvolts.
But I'd guess that would mean cutting some tracks on the PCB.
Others here who know this amp well might have a better option for you.

Re the rising voltage,, with no load the voltage will float a bit but no big deal,, with no valves not much can go wrong.
Phil.

[Dimi Pana]

Phil, hi and thank you again for the insight.

The weird thing is this amp was working OK for about a month after I purchased it, early last Nov 2018. Well, I did feel that the front face-plate was (probably) running a little bit on the hot side, but had no other to compare, the amp was working, so I ignored that. Then in mid-December while I was playing it started the crackling and hissing noises, and the amp stopped making sound. I took it to a friend who has the same amp, he removed the tubes, and found the power tube had burned out.

I felt that for a used amp off of eBay that was expected so I embarked on replacing both tubes with new so I can have a fresh start and a baseline. After some research and because I felt the amp distorted too soon, I decided to go with a pair of JJ tubes only instead of a 12ax7, I put a 5751 in the preamp which has 30% less gain and it did help a bit with what I was looking to accomplish.

It is at this point that I tried about 4-5 different brands just to see if my ears could tell the difference. I decided I could not tell any (significant) difference so I settled with the JJs. However, at this point the problem started getting serious. During my tube swapping I did not blow any tubes probably because I never played the amp too much or too hard.

So now that I settled on two JJ preamp + power tubes I cannot play more than 2-3 minutes and the amp sound crackles, hisses and fizzles and the power tube (ONLY the power tube is gone!?)

So what could cause the higher voltage to the heaters? Btw, this version 3 of the EVJ was specifically made with DC heater voltage to eliminate noise which was a problem with earlier versions. So there is no purpose going back to AC.

But what's causing the excess voltage going to the heaters? Could it be the transformer? Or the next logical think would be the rectifier. I look at the schematic and do not see any resistors after the bridge rectifier so I am assuming the path is:

120 VAC ---> transformer ---> 6.3 VAC ---> bridge rect. ---> ~8 VDC ---> tube heaters.

Am I correct?

Also, if for some reason the DC to the filaments is so high how come the preamp tube does not burn out. Plus when I did the first measurement I found over 9 volts, so am I right thinking that the bridge rectifier may have gone bad?

Now that I am getting more comfortable with all this, I will go back and do some more careful and invasive measurements. In the mean time, how can I "advertise" in this forum what my problem is in the hopes that someone particularly expert with this amp can help me out?

Thanks again!     

[galaxiex]

Is it possible that a damaged heater bridge rectifier could cause the high heater voltage?
Eg; one of the diodes shorted or open?

Or... maybe the transformer heater winding has a problem?

[g1]

Reading the heater voltage without the tubes installed is in this case not much help.
You need to check with the tubes installed.
Post your results.

[Dimi Pana]

OK, I can certainly do that (test heater voltages with tubes installed) BUT I am afraid I will burn the power tube again. If I just turn the EVJ ON but with no guitar plugged in and volume all the way down, am I still stressing the tubes, especially the power (e.g. if there's nothing to output) or I am going to be OK for at least a minute or so I need to take the measurement?

On a different note, to test the bridge rectifier I need to remove it from the PCB correct?

--

So just to clarify because the schematic does not help, there are two components that can be failing between 120 VAC input and the tubes heater filaments: The power transformer and the bridge rectifier. I see a capacitor but that is not in series it is going directly to ground. Shouldn't there be some sort of resistor in there too? I am no expert but my limited knowledge tells me there should be at least one controlling the voltage to the heaters. YES/NO?!

Thank you all for the input!

[Dimi Pana]

I was about to test the heater voltages with the tubes on when I had the idea to test the transformer output BEFORE it goes into the bridge rectifier. Well, it is 7.8 VAC guys, so I think we found the problem. For some reason the transformer voltage is too high. Not sure what cause it, but that's what I think is going on. Am I right/wrong? Chime in please! So after it exits the bridge and gets rectified, it will be even more hence the 9+ voltages I was measuring yesterday. Still have not put in the tubes, but does it make any difference at this point? Is it normal that all of a sudden the PT will output more of its nominal voltage. If you think I should still measure the heater voltages with tubes installed let me know - Thank you!

EDIT: While I was thinking what could have caused the PT to do that, I remembered I have a mains step-down transformer at work. I am a PC technician and sometimes we need to bring down the mains AC if it is too high so it won't fry PC equipment. Anyway, I plugged it all in and guess what. With the AC stepped down the heater voltage is a healthy 6.5 Volts. At this point I plugged the tubes in and played my guitar through the amp, no problem at all. I am posting some pictures, so what do you think? I am so lost about WHAT could have caused the PT to go ...crazy!?

[phatt]

Ed;
Hi Dimi,
Wow now you are running hot mate,, good work.
You will be the expert soon.

You posted just as I finished this but I'll post it anyway as it's relevant and may help.
-----
Number 1;
I think it would be wise to establish what AC voltage is coming from the transformer that powers the heater filaments,, FIRST!
Then move on to other possible problems.

Remove Valves
Set meter to read AC volts;
REMOVE the 2 filament wires that go to Bridge rectifier.
(in the pics I see on the net they might be Orange but you better check)
Insert probes in those 2 wires and read the AC voltage,, then post results.

For AC filaments the no load voltage should be around 7VAC, give or take a few points.
If the transformer was wound for full wave rectification then it should read less,, maybe 5VAC.

Yes you can power up with a power tube in place and read the heater pins on the back of socket, it will only take a minute
If it worries you then you can likely remove the high voltage wires from the pcb but leave the filament wires in place to power the heater,, (looks like all the transformer wires are spade cons so should be easy to remove B+) should only take a minute to read the loaded heater voltage, then turn off.

Regards AC hum,, Again misguided design Xpurts.
The PCB tracks are in the wrong place hence heater hum.

my guess;
To lazy to reroute the pcb with better layout so they just slapped a bridge on the heater supply.
which did help hum but now the filaments are running too hot.
Do they care?  I doubt it.
They sell trashy crap with a brand name on it and people buy it cause it has a big name brand on the front. I fixed a couple of Fender pro Juniors (some models had a really bad hum)
The most sensitive signal path in the whole circuit ran a trace right next to the B+ track. 
I simply Replaced that track with a shielded cable directly to valve socket and hum was gone.

And that is just FENDER,, other big names make similar F*** ups.
And you pay big dollars for the privilege of their name on the front.

From the RCA Valve manuals; "Triodes should last for 10,000 Hours at least,, power valves 5,000 hours."
If they don't it's usually because of a crap design in the circuit.
I read a report that a uni had a valve computational circuit which run 24/7 for 30 years.
Yet The best names in guitar amp world still don't get the joke,,, but the shops sell a lot of new valves every 2 years. go figure
Marshall are notorious for burning out EL34's because they run the screens way over voltage,, and still doing it.

Regards Amp distorts too soon;
Easy to understand once you study a bit of history.
Back in the 40/50's these type of amps were popular because a jazz player could lift has volume.
Those old guitars had low output PU's and there was enough power to lift the sound to balance against the Brass section.
Now even the cheapest guitar with crappy PU's has many times the output of those old semi acoustic jazz guitars,,, Result = your Amp cannot play clean with a high gain PU's,, Unless you turn right down.
If you want clean then you need a bigger amp to get clean headroom.
Rant over.
Phil.

[phatt]

Still have not put in the tubes, but does it make any difference at this point? Is it normal that all of a sudden the PT will output more of its nominal voltage. If you think I should still measure the heater voltages with tubes installed let me know - Thank you!

YES!

EDIT: While I was thinking what could have caused the PT to do that, I remembered I have a mains step-down transformer at work. I am a PC technician and sometimes we need to bring down the mains AC if it is too high so it won't fry PC equipment. Anyway, I plugged it all in and guess what. With the AC stepped down the heater voltage is a healthy 6.5 Volts. At this point I plugged the tubes in and played my guitar through the amp, no problem at all. I am posting some pictures, so what do you think? I am so lost about WHAT could have caused the PT to go ...crazy!?

A stupid crazy clueless Valve amp designer
Phil.

[galaxiex]

Is it possible for the transformer heater winding to partially short a few winds and cause the higher AC heater voltage?

[phatt]

Is it possible for the transformer heater winding to partially short a few winds and cause the higher AC heater voltage?

If so surely the voltage would be lower not higher??
Phil.