JCM 2000 bias pins maxing at 15mV

[Djizix]

I've replaced the mains inlet pcb and power transformer. Previous owner had a 25a main fuse in it and fried them both. They were relatively simple to repair and now the amp at least turns on, where before it was blowing fuses immediately on power up.

First thing I did was attempt to bias from the pins located on the back. Service bulletin calls for 80 mV and I'm maxing the bias pots out at 15 mV with amp connected to a load and heat up time of approximately 20-25 minutes.

Also, I plugged a guitar in and my ultra gain is humming at only 20% gain. The amp doesn't sound horrible through classic gain but its definitely missing something. Where as my low end is almost non exsistant in ultra gain.

So far I've been told that its likely that my board is conductive between the pins of my output tubes. But by my understanding this common problem results in high bias readings and mine aren't even close. In fact, I'm weary of playing my amp at all for fear of my amp playing too cold. But thats just a hunch, no real knowledge to back that up.

[Randall]

What you are doing at the back connector is reading voltages across a 1 ohm resistor on the cathodes of each pair of power tubes. Since it's 1 ohm the math works out to mV = mA, so in your case we can say your pair of tubes is drawing 15 mA. It's very low, but it won't hurt anything. A current draw of this low usually means the grid is seeing too much negative bias voltage. The higher the negative bias voltage applied, the colder the tube will run (draw less current). These boards usually conduct some of the positive plate voltage to the grid pin, causing it to go less negative and therefore conduct more. In some cases, a lot more until the tube dies or the board burns up.

I'd measure what the voltages are on pin 5 of the EL34s, and I'd also have a look at those 1 ohm resistors. Also, you didn't say if these are known good tubes or not, if they were in the amp when it blew up, they may just be pooched.

[Leo_Gnardo]

I'm thinking, if the amp owner wrecked his PT by installing 25 amp line fuses, there may be other problems lingering in this prize. That PT likely didn't decide to melt down on its own, there's got to be a reason for it. Now there's a new one you are wise to track down & fix the problems. How are the screen grid resistors? If they roasted and increased in value, that would explain an inability of the output tube to move much current. Same for the 270R cement block resistor that knocks down hi voltage to appy to those screen grids. Of course check the bias voltage as Randall suggests, it should be in the -40V region.

If this JCM2000 was made 2006 or before it's suspect to have the dreaded conductive circuit board problem. Far and away the best solution is replace the board. Big bucks, lots of shop time, long wait for an overseas shipment. There's also the solution worked out by our German friends at Hullerum, read here:

The Marshall TSL122 TSL100 thermal bias drift repair page

More JCM 2000 problem areas by Rich Koerner in sunny downtown Union NJ, I could swear there was text to go along with this but follow the arrows & you'll see things that may need attention:

Rich Koerner's Time Electronics Picture Gallery of Repairs & amp Nightmares, A Marshall 2000 TSL-100 Repair

[Djizix]

Every resistor that I've been led to has tested the correct resistance. I agree it wouldn't be a bad idea to try and find a bad one, and that is my next venture when time allows.

On another note though, my tubes passed the continuity test. I'm sure there is more too them being good than just that but I have an option of buying new tubes and seeing if one is just struggling or buying the new PCB and getting it over with to avoid pouring more money into it, to no avail. Obviously prefer it just being the tubes but unfortunately things don't always happen as you prefer. ��

[g1]

Pull one of the power tubes and measure the DC voltages in the socket pins. Report your readings.

[TimmyP1955]

The Marshall TSL122 TSL100 thermal bias drift repair page

I just did the Pin 5 mod to one and the bias seems to finally be stable.

I had previously done another mod - I replaced R77 (common feed to the two pots) with two resistors to feed the pots independently; Replaced R68 (common load on the two pots) with two independent load resistors. (I placed these on the little pot board.) I did this in the hope that the interaction between the pots would be reduced (I am not convinced it was - or wasn't).

Make sure that the grid stoppers are 5k6 - some of these amps have 220k!

[Djizix]

Ok, so I've tested all resistors on the main pcb, and to no surprise there were some not matching up. My schematic reading isn't top notch but I've noticed a pattern of them all being in groups. I can put together that something in that grid isn't working properly. I just don't know how to pin point what exactly. I'm going to go ahead and buy some resistors in hopes that those are what's causing the problem.

Here's the groups of resistors that were off in testing.

R39 100k tested at 31.2k
R40 47k tested at 19.7k
R41 47k tested at 19.7k
R44 22k tested at 18.6k

Make sure that the grid stoppers are 5k6 - some of these amps have 220k!

I believe my amp is an amp that has the 220k in these places. Which if I'm not mistaken are R7, R66, R70, and R10.

And lastly,

R64, R72, R73, R75 all supposed to be 220k and they tested at 170k. And they took like 10 minutes to stabilize, is that normal?

I'm kind of at the end of my knowledge at this point. I have been reading everything I can, but I'm limited in experience so any advice is greatly appreciated!

P.S. To those that instructed me test the pins on my outlet tubes. I'm not ignoring your instruction, simply just don't have the board in at the moment.

[DrGonz78]

Here's the groups of resistors that were off in testing.

R39 100k tested at 31.2k
R40 47k tested at 19.7k
R41 47k tested at 19.7k
R44 22k tested at 18.6k

I believe my amp is an amp that has the 220k in these places. Which if I'm not mistaken are R7, R66, R70, and R10.

And lastly,

R64, R72, R73, R75 all supposed to be 220k and they tested at 170k. And they took like 10 minutes to stabilize, is that normal?

This to me sounds like you might be measuring the resistance while in circuit. For example, you say took 10 minutes to stabilize and I am having trouble understanding what you mean. This will confuse the meter when other components are in parallel and 47k might read 19.7k. Many time to read them correctly you need to lift a leg off the board. Also, is this a DSL or TSL? Please post the schematic to which you are referencing your board version. I am working on a DSL in my shop at the moment and this is the schematic I am using. Seems to match up to what you have mentioned so far. If so some of the resistors that you are testing are not really in the scope of the issues you are trying troubleshoot. Then again maybe I am missing something since I am unsure of the schematic you are referencing. Best Regards.

[Djizix]

It is a DSL. And from what the previous owner tells me, its from 2001.

This link http://drtube.com/schematics/marshal...01)%20iss7.pdf is the schematic I'm using.

And if I'm understanding you correctly, in order to read resistance on, for instance, R40 on this schematic, I need to detach one leg of the resistor from the board? Or do you mean one leg of the meter?

Also, you mentioned that these misreadings aren't in the scope of what I'm trouble shooting. I just assumed that testing all the resistors would be a good way to analyze my problem, judging by the comments I've received thus far. So if they're unrelated would it be a mistake to leave these resistors as is, reinstall my pcb back in the chassis and try some of the tests on my output tube pins, as stated before?

[DrGonz78]

Yes detach (lift) a leg of the resistor off the board to take it out of circuit to get an accurate reading of the individual resistor. Some resistors to be concerned with to start are the grid resistors R7, R66, R70, and R10, like you mentioned. That is is a good starting point and making sure they are updated to 5k6 like in the schematics attached. Also, do some inspections on the bias board itself and make sure those variable resistors (bias pots) are functioning well. At that point we need to go back to testing voltages as G1 pointed out. Testing B+ voltage and negative bias supply voltage to really see what is going on.

Edit: Also don't forget to check out those screen grid resistors as Leo had pointed out which are R60, R65, R74 & R76. First test them without removing them from the board and see if they measure 1k. If they measure higher resistance than expected then you will have to remove them to test further. R64, R72, R73, R75 were the other one's Leo had you check which was good since they all measure 170k in circuit. I don't think they drifted high. I just saw that you might go hunting other resistors that were not related to problem at hand.

[DrGonz78]

R64, R72, R73, R75 were the other one's Leo had you check which was good since they all measure 170k in circuit. I don't think they drifted high. I just saw that you might go hunting other resistors that were not related to problem at hand.

Sorry it is late. Note that R64, R72, R73, R75 are just bleed resistors and that I meant to say check for drift on R71 that 270R resistor Leo mentioned.

[Enzo]

Don't obsess over resistor values. Resistors almost never go low in value, and especially not like from 47k to 19k. They do go high when failing at times. Low resistances are invariably just parallel current paths in the circuit. If you have a legit reason to suspect a particular resistor, you can unsolder one end, but going through the amp unsoldering part after part is more likely to cause more problems than solvve anything.

Screen grid resistors have a tube screen on one end, they are already about as out of circuit as you can get. No need to unsolder them.


When you increase the tube current by lowering the bias, it drags the B+ voltage down. This of course affects the other side tubes. THAt is why the two controls interact. You set one, then that changes the B+ voltage for the other one. So you adjust that one and now the first one changes. You go back and forth a few time until it all centers.


Please stop calling these things JCM2000s, there are numerous models in the JCM2000 line. JCM2000 doesn't specify a model. It is like saying "I have a Chevrolet." Well is it a Malibu? a Corvette? An Impala?

[Djizix]

Ok. I will get the board put back in and start testing as instructed. I'll let you all know what I find. Appreciate it!

[Djizix]

So I tested the 22k variable resistors on the bias board to the best of my ability. I'm having trouble identifying which is positive and which is negative, but I'm fairly sure I have the neutral nailed down. Either way, I alternated the black lead accordingly and my readings were on one side, fluctuating around 11-22k and the other side was 14 ohms turned counter clockwise, and 25 ohms clockwise. Doesn't seem right, but I don't really know.

UPDATE: Got the amp back together, plugged it in, hooked up to a load, standby for 10 min, and powered up. My bias is now reading 8 mV on the left side pins, and 130 mV on the right side pins. Which caused my left two tubes to get a little too hot for my comfort. I immediately shut it off after and haven't resumed testing.

So, if I'm gathering the clues correctly...I, in fact, do have a thermal drift, or at least the same symptom now. Whether or not I caused it while taking the board out, or this is just my first time witnessing it occurring, is beyond me...

So basically, my symptoms changed from being not enough, to way too much on just one side... With or without the proper knowledge of the field, I think this is a pretty easy decision...just not as cost effective as my wallet hoped for.

If you guys have any other suggestions, I'm open minded to new solutions. But I have a feeling my only option is to try and do the drilling pin 5 out routine, and I'm not up to the task. Thanks guys.

[Randall]

DEWALT 3/16 in. Diamond Drill Bit-DW5570 - The Home Depot