No reading when biasing JCM2000 DSL100

[Slinkymcvelvet]

Ok... update, 3 weeks later. I replaced the fried R9 resistor and R6 for good measure, checked the board visually for any other problems, checked that all the screen resistors mentioned above were within spec, removed C46 cap completely and did the ground mod on the output sockets. I installed a brand new matched set of EL34's and biased them to 82ma on each side. The amp worked... for all of about 6 hours at rehearsals then the HT fuse blew again.

Long story short, HT and R9 had blown again, and V8 redplated when I replaced the HT fuse. Removed all the power tubes, ran the amp with Standby OFF and power on (ie amp fully on) and the HT didn't blow. Opened the amp up, replaced R9, checked for arcing, touched up suspect soldering points that were a bit dull, checked spec of all the aforementioned screen resistors which were fine. Reassembled the amp, installed the existing pre-amp tubes, turned it on full again and nothing blew.
A brand new full set of both pre-amp and power tubes are in transit.

Now... rather that try and pick an arbitrary low number out of the air to bias the the new tubes on, I decided to take the plate voltage readings from PIN 3 on all of the power amp sockets. Each reading was 528v. Surely this is too high? By my reckoning this should equate to 30ma per tube when biasing based on the calculations I've seen on here.
Any further thoughts on this would be appreciated before I go installing the new power tubes and ruining them.

 

[chadjwil]

Sounds like you took those plated voltages without the tubes installed. You need to have all the tubes installed when taking that reading.

 

[Slinkymcvelvet]

I did chadjwill. I guess that will entail buying a bias probe then? Is there any way round this?

PS, I only installed the pre-amp tubes.

 

[chadjwil]

You'll need all 8 tubes in there to get the right readings at the plates. You shouldn't NEED to buy a bias probe, you shouldn't be blowing components up like that. There is clearly a bigger problem going on.

Before you install the new EL34s set your bias voltage to the max negative reading you can get (DC volts on pin 5), that will help ensure that you are operating it in a safe region when you put them in. And after you bias it, keep an eye on your bias readings for an hour or two. Play through it for 10-20 minutes then check your bias, do that for at least an hour. If it runs way up after some time that could be the cause of your tube failures and HT fuse blowing.

 

[Slinkymcvelvet]

Ok, before I install the new tubes, I have to make sure that that the DC voltage on pin 5 is at an absolute minimum. So with no tubes installed and the amp powered up, I take the reading from pin 5 and adjust the bias trim pots next to the 3 pronged Molex connector accordingly. Is that correct? If I remember correctly, turning the pots clockwise decreases the reading in milliamps from the molex pins, whilst anti clockwise increases the reading on my amp. Is that also your experience?

 

[Slinkymcvelvet]

Just finished doing what you asked chadjwil, I worked it out for myself. On pin 5 on every tube socket, the bias voltage was reading -40. I turned the trim pots clockwise as far as they could go and now each side, and every tube socket is reading -51. Does that sound about right?

 

[Micky]

Best to take the bias reading from the molex connector...

 

[Slinkymcvelvet]

Best to take the bias reading from the molex connector...

I will be doing that Micky, thanks, but first I want to get the plate voltage by taking a reading from pin 3 and doing the math to get a figure to work to. I want to bias it correctly this time and according to the correct figures. Last time I just turned the bias pots to read 80ma on each side without checking plate voltage first.
It's definately a learning curve and the threads and posts on this forum from you guys are a great help. Thank you.



Backwater. Scotlands' Ultimate Status Quo Tribute.
www.backwater-quo.com

 

[Micky]

Yeah, you will get it.
Tube amps can be relatively lo-maintenance if they are cared for.

 

[chadjwil]

Just finished doing what you asked chadjwil, I worked it out for myself. On pin 5 on every tube socket, the bias voltage was reading -40. I turned the trim pots clockwise as far as they could go and now each side, and every tube socket is reading -51. Does that sound about right?

Yep, that's it. -40 was probably safe enough, but at least this way you know for sure that you're not going to hit the new tubes too hard on initial power up. Just a safety precaution really.

 

[RickyLee]

You can get a quick pin 3 plate voltage check with the tubes/valves installed. Well, for this method you will have one valve pulled and put your voltage meter positive test lead into the socket pin 3 of whatever valve you decide to remove. Then just power on the amp as you normally would for just a few seconds or long enough to see the running plate voltage with the 3 valves only installed, then shut it down.

You can leave the red meter probe in that socket and should not even have to hold it.

 

[Lane Sparber]

You can get a quick pin 3 plate voltage check with the tubes/valves installed. Well, for this method you will have one valve pulled and put your voltage meter positive test lead into the socket pin 3 of whatever valve you decide to remove. Then just power on the amp as you normally would for just a few seconds or long enough to see the running plate voltage with the 3 valves only installed, then shut it down.

You can leave the red meter probe in that socket and should not even have to hold it.

That won't work properly, Ricky. The power tubes present a resistive load to the B+ string, and removing one tube will "un-load" the circuit substantially enough to throw off the plate voltage reading in this instance. Unfortunately, there's no shortcut to doing it correctly. Unless you have a bias probe that also takes plate voltage readings, you have to pull the chassis and do it the "hard" way.

-Lane

 

[Slinkymcvelvet]

That won't work properly, Ricky. The power tubes provide a resistive load on the B+ string, and removing one tube will "un-load" the circuit substantially enough to throw off the plate voltage reading in this instance. Unfortunately, there's no shortcut to doing it correctly. Unless you have a bias probe that also takes plate voltage readings, you have to pull the chassis and do it the "hard" way.

-Lane

Yeah... that was what I was guessing too. So it looks like I'll be doing it "the hard way" when the new tubes arrive.:eek2: That would involve going in the underside of the chassis to take a reading from the pin 3 terminal on the board itself whilst powered up, is that right? I'll need to have a steady hand!

 

[Lane Sparber]

That would involve going in the underside of the chassis to take a reading from the pin 3 terminal on the board itself whilst powered up, is that right? I'll need to have a steady hand!

Yes. Welcome to the wonderful world of amp repair. Keep a steady hand on your probe (snicker...snicker...) and the other in your pocket and you'll be fine.

-Lane

 

[glina]

Once you're done - plug the amp in and relax in the pool

 

[chadjwil]

Great to hear from you again Lane!

 

[Slinkymcvelvet]

Ok folks, I'm now in the process of doing it "the hard way" as Lane has said. Initial readings on pin 3 for plate voltage read 501 v across all tubes. At 25w dissipation rate for EL34-B's, by my calculations that worked out at 32ma. I haven't added 5ma for screen current on this figure, should I do that?

After 5mins having biased each side at bang on 32ma, the plate voltage reads 490v across all tubes, so the calculation now reads 33ma. This I have done, but again haven't added the extra 5ma. The strange thing is that 5 mins later the plate voltage has now gone up to 495v. Any ideas as to why?

 

[chadjwil]

As the cathode heats up after initial turn on the tube becomes more conductive and less resistive which accounts for the voltage decreasing at the plate. Why it would go back up I can't say for sure, but it likely has something to do with your bias. Keep an eye on your bias and see what it does over some time.

And by the way, adjusting the bias will have an effect on plate voltage as well. Just keep that in mind as you're chasing this down

 

[Slinkymcvelvet]

As the cathode heats up after initial turn on the tube becomes more conductive and less resistive which accounts for the voltage decreasing at the plate. Why it would go back up I can't say for sure, but it likely has something to do with your bias. Keep an eye on your bias and see what it does over some time.

And by the way, adjusting the bias will have an effect on plate voltage as well. Just keep that in mind as you're chasing this down

Nearly 50 minutes later the bias is sitting at 33.7ma both sides and the plate voltage is 497v. Isn't that voltage a little hot for a DSL 100?

I'll give it another hour before checking it again. My only other problem is that I can't play anything through the amp at volume at this late hour, it's nearly 10.45pm (GMT) on this side of the Atlantic and I've got early rising neighbours upstairs! :Ohno:

 

[RickyLee]

That won't work properly, Ricky. The power tubes present a resistive load to the B+ string, and removing one tube will "un-load" the circuit substantially enough to throw off the plate voltage reading in this instance. Unfortunately, there's no shortcut to doing it correctly. Unless you have a bias probe that also takes plate voltage readings, you have to pull the chassis and do it the "hard" way.

-Lane

LOL. Yeah, I know Lane. Good to have you back by the way.

I should have added an amount of volts for him to add to that "missing valve" measurement to get him in the ballpark. It was only a "in the pinch" type reading anyway. I see more change in my B+ from varying wall voltage than I would see with one of four valves missing.



EDIT: You DO NOT want to know how I checked my TSL100 plate voltage last night without pulling any valves or the chassis . . .