JCM800 Burning Tubes - HELP!!!

[mikeboone]

A guy brought me a JCM800 that burned a few power tubes in one socket but not the other. It burned the socket pretty badly, so I changed it.

I suspect maybe the OT. But my initial measurements show that it may be OK. With no tubes or load, the SEC has 1 ohm of resistance, the PRI has 89 ohms. It is show 5-4M ohms from PRI to SEC. I compared it to a Fender OT and got similar readings.

With a tube in one socket it powers up and acts normal, even passes audio at a strong level. With a tube in the other socket the amp hums loudly, smokes the tube beteen pins 2 & 3 and blows the HT fuse.

The voltages are the same between the 2 sockets (no tubes)

1 - 4.6VAC
2 - 614VDC
3 - 618VDC
4 - -51.4VDC
5 - 622VDC
6 - 4.5VAC
7&8 - GND

I have a schematic but there is no voltage reference info on it.

Any help would be great.

Thanks,
Mike

[Chuck H]

Either you miread the socket pins or you have it wired wrong. It looks like "your" pin #s are behind by 1 (1 is 2, 2 is 3, 3 is 4, etc.) At least thats how the voltages read. Unless you made an error on the pin #s, you may want to check the wiring on the new socket. Looking into the amp chassis, pin 1 is to the left (clockwise) of the base pin alignment.

Chuck

[mikeboone]

OK Then...

I fixed my little pin assignment faux pas (its been a long day), sorry for the confusion and thanks for the catch.

So when the guy brought me the amp, we put a set of tubes in it, plugged it in and played it for a moment and the tube smoked. I tried the remaining tube in each socket and it seemed OK.


Is there any better (more conclusive) way to test the OT. I suspect that the head may have been turned on with no speaker load. Just a hunch.

[Chuck H]

You could put an AC signal into the secondary (measure this voltage while it's hooked up and running) and measure at the primary. Measuring both ends of the primary will give you the turns ratio (ie: 1 volt into the secondary and 20 volts on the primary would be a 20:1 turns ratio). Measuring each end of the primary to CT should show if anything is wrong.

Chuck

[mikeboone]

I am noticing another thing that makes me concerned is the bias voltage. I have read several threads about EL34 biasing and most state that the bias voltage should be around -42.

This amp has been fitted with a 1 ohm resistor hanging off the cathode to measure bias. Both cathodes are common with the resistor to ground on one.

I am getting a bias voltage range of -45 to -60. At -45 I am hitting 120mA. So I dropped it as low as I could get it (-60)nd am getting around 70mA.

Does this seem OK??? I have mostly worked on Fenders and am accustomed to running 6L6's at around 25-30mA each.

[vintagekiki]

mikemoone@ ...smokes the tube between pins 2 & 3 and blows the HT fuse...

How much voltages on all octal tube sockets without output tubes?

Are you sure that is correct put EL34 in their socket? Tube base nose (between pins 1 & 8) must be locate between pins 1 & 8 tube socket. Tube pins count from bottom side tube base or tube socket.
Blows the HT fuse with or without output tubes in sockets?
If fuse blows with output tubes in their sockets, one or more EL34 is burned (defective).
If fuse blows without output tubes in their sockets, one or more socket is burned (defective).
Recommendations: when change defective tube socket(s) - change output tube(s) Safely is safely.

With analog ohm meter range on 1000 ohm check base EL34 tube or socket(s) on short circuit (burned) between pin2 (filament) and pin3 (anode).
Replace defective tube(s) or socket(s)
Defective tube(s) or tube socket(s) is not for use.
Tubes with broken nose also is not for use.

Output tubes should be relative matched (min/max difference between tubes is to 10% max)
When replace defective components, set bias on 25mA per one tube, measured on 1 Ohm/ 0,5 - 2W mounted between pin 8 tube socket and chassis with 200mV DVM.

Approximately voltages on output tubes

DC voltage measured between pin and chassics
1 - GND
3 - 400 - 525V
4 - 400 - 525V
5 - -45 to -60V
6 - NC
8 - GND

AC (heater) voltage
6.3 - 6.5V AC measured between pin 2 and pin 7 or
3.15 - 3.25V AC measured between pin 2 or pin 7 and chassics

[mikeboone]

Tiki...

Thanks for the input. I need to make some clarifications. I had some facts confused in my initial post, I was tired and aggrivated.

The amp burned two tubes in that socket before I ever did any work to the amp. When I put the second set of tubes in, the tube in that socket smoked after a few moments of played and destroyed the socket.

I did get the pin assignments wrong when I changed the socket, this has been corrected.

The voltages I measured are without tubes.

All burned up items have been tossed out.

So, now I am at the point of needing to: (1) verify that the OT is good or bad. (2) Get the bias corrected.

I can only get the bias down to 60mA, roughly 30mA each. I checked it with a single tube in each socket and then together.

I am afraid to start playing through it as it may smoke again. It didn't give any warning the first time it smoked.


Voltages:

3 - 555
4 - 553
5 - -51
6 - 564 These pins are connected to a filter cap and to the reverb driver.
2&7 3.5VAC

Please advise...

[mikeboone]

Chuck,

I put a 1VAC signal in at the speaker output and measured 28VAC across the PRI of the OT.

The signal measured 14VAC each from CT to each tap.

Can I safely say that the OT is in good shape??? Is 28:1 too high of a ratio?

Please advise.

Mike

[pdf64]

What secondary tap had the 1V input for the ratio measurement?
What freq used - 250 to 400 Hz is best as that's where speaker's actual impedance = nominal.
But the fact you're getting 14Vac on each pri leg to CT is encouraging.
RG Keen has published a test for shorted turns in a transformer
http://www.geofex.com/FX_images/xform_test.gif
which is a common fault mode that is tricky to diagnose.
So confirm OT is good, screen grid and grid stopper resistors are good.
Advise to fit a fast fuse 330mA in series with each cathode ground return.
And another fuse or 100ohm 1/4 watt resistor in series with the heater winding ground reference, just in case a el34 develops a plate to heater short.
Then power up and see how it responds to a signal - Peter.

[mikeboone]

I measured it again with a 400Hz tone and got 20.6VAC. I am now on the 8 ohm tap. I was using a 1K tone (I am a telcom guy) on the 4 ohm tap. I had it on 4 to match my test cabinet.

I would have to say that is good news. Is there a way, short of the neon lamp method, to test this? I dont have a neon lamp or a way to get one on a Sunday.

Will it behave differently with a higher voltage and current load, possibly exposing a low resistance between windings that hasn't fully shorted???

[vintagekiki]

If amp work correct with 2 tubes and if second set of tubes destroyed the socket, one or both tube of second set is defective. Replace defective tube set with correct tubes.
Do you check base EL34 tube or socket(s) on short circuit between pin2 and pin3 with analog ohm (resistance) meter?

About OT
Load ressistance calculate for EL34
For 2 tubes Raa = 3400 – 4000 Ohm
For 4 tubes Raa = 1700 - 2000 Ohm

Raa = square n x Rspkr
n = square root Raa/ Rspkr or
n = Uaa/ Uspkr
n4 = square root Raa/ Rspkr = square root 2000/ 4 = square root 500 = 22.36
n8 = 15.81
n16 = 11.18

where
n is ratio
Uaa is voltage measured between primary both tap (from anode to anode)
Uspkr is secondary voltage measured between ground tap and speaker (4/ 8/ 16 Ohm) tap

In your case ratio is relativity high
n = Uaa/ Uspkr = 28/1 = 28
Raa = square n x Rspkr = square 28 x 4 = 784 x 4 = 3136 Ohm
This transformer may be used with quartet 6L6 or 5881 tube, because their Raa is 5600 – 6200 Ohm for pair tubes or 2800 – 3100 Ohm for quartet tubes. Also this transformer may be used with one pair EL34.

In basic if measured voltage from CT to each tap is identical or approximately ( to 5%), output transformer is correct. In your case, measured voltage from each CT to tap is 2 x 14V (symetrical) indicate that output transformer is correct.

Which model is your JMC800?

[mikeboone]

Tiki,

This is a 2205, 50W head.

When I re-measured the ratio of the OT, I came up with 20VAC at the PRI with a 1VAC at the secondary. Check mentioned that the ration should be 20:1 and it is. So I think I am OK there.

My major concern now is the bias current. I cant get it below 30mA per tube. I am considering changing the dropping resistor (R51) to something higher to knock down and voltage and give me a lower range.

I played the amp with a 100mV sine wave at various frequencies while monitoring the bias current using a single 1 ohm resistor on one socket, both cathodes are commoned and grounded through that resistor.

I measured 60mV(mA) with the circuit idle. And various freq's 60, 120, 400, 1000Hz, the bias went up to about 80mA at pretty high volumes (above 4). It stayed in that test for about 10 minutes with no problems.

I am feeling pretty confident that the OT is good. Fair assessment???

Mike

[Amp Kat]

I would say that OT is good. You did change those tube sockets right ? Don't want any arching going on there. 30 ma's is starting to get a little cold and approaching cross-over distortion. What is your % # you are trying to get at ? 70 to 80 % seems about right. So lets say a 30 watt tube at 70% = @21 watts / 555 = @37ma's right. So at 30ma's your a little low. Your range seems ok but that's your call how low you need it.

[Chuck H]

Ok Mike, I have a short story for you...

Once I rebuilt a Traynor head into a custom design. The old Traynor eyelet boards are usually wavy as hell. There are usually two. A top one with eyelets and a blank one beneath. I was worried that with my new layout one of the eyelets might contact a metal board mounting screw. So I put a piece of rubber tubing between the two boards to counter the warp. It worked fine. For about ten minutes. Then the amp started to behave very strangley. Multiple symptoms such that I couldn't figure out where to start looking for the problem.

What happened was that the rubber tubing had enough carbon in it to become conductive. My insulated spacer was now a resistor between the B+ rail and other eyelets it was in contact with. The smoke was the givaway...

Point is, your voltages are fine. The OT seems to test good. The bias circuit is there and cold if it's off at all. So what the hell is making that socket kill tubes??? It's not the socket, you replaced that. If anything wierd was afoot there you would have seen it. Plus the amp has the same problem as before you changed that socket. Ok, so it's not the socket. It's not the OT. It's not the bias or B+ rail. And it's only effecting the one socket. So...

Check all the circuitry peripheral to that socket. There could be a componant that seems fine but only fails to perform under operating conditions. There could be a blob of solder or a sliver of wire contacting something it shouldn't. Maybe a lead installed by someone else has too low an insulation rating and is touching the chassis. Or a cold solder joint that behaves only until it's stressed. You need to look for the non obvious answer at this point.

Chuck

[mikeboone]

Thanks neighbor... Hows BR doing after the storm. Houma is recovering slowly but surely.

You know, I forget the little things, the important ones obviously. I just looked at Webers bias calculator and charts and I seem to be right there in the ball park.

Today, owner said he blew the HT fuse just prior to him dropping it off. He only had the amp for a few days. He replaced the 500mA fuse with a 1.5A.

I wonder if the socket was arcing all the while - causing this issue. I have never been this gun-shy about turning on an amp!

I will examine the PCB and wiring for defects.

I just noticed that the 1K power resistor on that socket measures 750K ohms out of circuit. How much tolerance do I have wo work with?