HT fuse just won't stop blowing.

[MitchK]

This has been an epic journey so far. This is a little long but I'd appreciate it if you all could give me a hand here.

I have a marshall jcm 800 2203, canadian market model. I have another thread about it but I'm just going to give a quick rundown:

1. Amp cuts out for me, no sound, I dime it and it will cut in and out. I do this for about a minute or two trying to figure out what it happening and a tube blows. Turns out the speaker cable had actually died I figured out later on.

2. I pull the dead tube and replace the fuse it took with it. I now run the amp with just 2 tubes in it. I play two shows that night, the amp works for part of the night and then dies on me mid set. I open the amp up the HT fuses (which are 2x 0.5A slo-blo btw) are blown.

3. I replace the HT fuses turn it on and pop, they blow immediately. So I figure I have a bad tube. I replace the el34s in the amp and try out a quad of 6550s.

4. Power the amp up with the 6550s and some new fuses it works for about 10 minutes and then pop. the fuses blow again.

5. I recap the amp, all new power filter caps. While doing this i also happen to come across the fact that the bias pot reads at max 16Kohms when it should be reading 22Kohms. So I also replace this with a new pot.

6. I test the 6550 tubes and they all test good, I go back and test the el34s, 3 of the 4 work but one upon plugging it into my tube tester cause the tube tester to smoke. I throw this tube away.

7. I then power the amp up with a quad of KT88 tubes I have from another amp that I know work fine. The amp turns on and seems ok while idling but at one point makes some not so loud scratchy buzzy sounds. I turn the standby off, wait a second, turn it back on and it sounds fine again. I finish biasing the amp and have it sitting at roughly 20watts static dissipation.

8. I plug in and start to play, the amp sounds good it gives me a feeling better than any woman has ever made me feel (hahahaha) and as soon as I hit a nice chunky chord.... you guest it, the HT fuses blow. I recheck the tubes and they still test the same as before I ran them in this amp.

I'm just lost at this point. I want to check the driver tube plate to power tube grid coupling capacitor based on what I'm reading at the tube amp debugging page (http://www.geofex.com/ampdbug/ampdebug.htm), I've read pin 5 on all the sockets and they are sitting at -37v. When I read one of the sockets it seemed to oscillate between 0 and -37 for a second but I think I just wasn't making good contact with my probe because it seemed to go away, but I'm just so stuck at this point I'll look at anything.

What can you suggest??? I'm just out of ideas! I'm going to go over all the resistors in the amp to make sure they are still to spec but I don't really know how I can go about testing all the capacitors though? I have a replacement choke OT and PT that I was going to use to build a clone of the amp that I can try if necessary but that seems like a huge without looking at everything else first. and above all, I'm sicking of buying 0.5A fuses and watching them constantly blow! Those things aren't cheap at radio shack.

[Mark Black]

Probably a long shot that it could be the same exact problem, but I did once work on a Canadian model JCM800 and it turned out a bias filter cap was leaking in such a way that it gradually dragged the bias toward 0V over time causing excessive current draw. Something of that nature could be verified by monitoring the bias voltage at one of the tubes while leaving the amp on for a while to see if it changes.

Like I said - probably a long-shot...

Good luck!

[ken]

Bum speaker cable? Check your output transformer - you probably have an intermittent winding short from overheating it. It happened to me the same way...

Ken

[Amp Kat]

http://www.antiqueradio.org/dimbulb.htm
This should help with the fuses and I know RG has another lightbulb current tester on his site but I couldn't find it. Almost a must for testing fuse blowing problems. I have had some of these amps that I go up to 1 amp fuses and it never blows again but not any higher than that. Marshalls are pretty common for blowing those 1/2 watt fuses and I agree that the speaker cable would be a concern but I've seen them blow for no apparent reason at all just sitting there and poof. Monitoring the bias voltage also sounds like a good idea and even monitoring the current draw of the tube would be good to find out what's tugging on it.

[stokes]

Like KB says the current limiter/dim bulb tester is an absolute must.It seems to me,from what I am reading you are going between 4 tubes to 2 tubes,El34 to 6550 back and forth and so on.When trouble shooting stick with one tube type,either 6550 or EL34 and use all 4 tubes.Pulling two tubes will require rebiasing as will changing from 6550 to EL34.Switching back and forth is just complicating things.

[MitchK]

Bum speaker cable? Check your output transformer - you probably have an intermittent winding short from overheating it. It happened to me the same way...

Ken

But this is the problem if its intermittent I don't really know how to check it? Aside from just replacing it that is, but thats really a last resort.

I'm going to monitor the bias voltage and see what happens when the fuse blows, but how does this work? the negative voltage goes closer to 0 so therefore the plate voltage increases which means more current being drawn on the B+ and this blows the fuse. correct?

as for the dimbulb test I have a variac and when I power up with that its fine, the lamp slowly gets brighter while I turn up the variac and I've replaced the caps so I know they aren't the problem, its not a dead short (well if it is its intermittent) it takes a while before the fuse is blown.

would it make any sense to hook up an ammeter in place of the fuses and see how high the spike is? I'll put like a 1A fuse in series with the meter just to make sure the spikes don't get too high? I mean I know I may as well just put in 1a fuses and see what happens at that rate but I'd like to know whats going on because if it won't run stock them you know how long is it going to last as it is now if there is some problem that could get worse.

so far on the list of things to test is:

1. check the OT but I don't really know how to test for an intermittent short

2. monitor bias voltage while fuse blows

3. try the amp with 1A fuses



anything else? and can someone point out the bias filter cap for me and the driver tube plate to power tube grid coupling capacitor on the schematic? I think that would be the 2 0.022uf caps that connect to the 5k6 grid resistors (c14 & c16) and the bias filter cap would be the 10uf (c19)??? that connects to the 220k grid leak resistors?

[Don Symes]

so far on the list of things to test is:

1. check the OT but I don't really know how to test for an intermittent short

2. monitor bias voltage while fuse blows

3. try the amp with 1A fuses

for #1 you need to take the OT to a 'Hi-Pot' tester ... I think. Probably take a hair dryer to heat the thing up just before testing? Where you'd find someone who would let you use theirs, I dunno.

#2 should be pretty straightforward. Remember the bias voltage is NEGATIVE.

#3 NO. The fuses are blowing because something is bad wrong. Just raising the current limit can start a fire and/or do seriously expensive damage.

To find the bias cap, look for the electrolytic with its positive terminal connected to GND. There may be a couple of these in a Pi filter arrangement .

One last thought - have you checked the socket tension on these grid pins? On a hot socket, they may open up just enough to let the grid drift high.

Hope this helps!

[R.G.]

I'm sicking of buying 0.5A fuses and watching them constantly blow! Those things aren't cheap at radio shack.

It's frustrating, isn't it?

The frustration disengages your ability to think logically though.

You gotta get it to quit blowing fuses so you can debug it. If you don't do this, you're sunk. So try one of these:
(a) a light bulb limiter
(b) a power resistor calculated to limit B+ current to a safe max value for the PT, and inserted into the HT fuse slot.

With either of these, you can now get it to fail and when it does, measure what's happening.

It's either oscillation or loss of bias when something heats up, but the "something" will be forever lost until you can pin it down without a fuse blowing.

[MitchK]

alright a couple quick questions:

first off with running the amp through my variac, aren't there going to be issues with running the amp with all tubes in at say 50 volts input?

and can you go into a bit more detail on the method of using a power resistor in place of the fuses? how would I calculate this? take the regular HT voltage, lets say 450v divided by 500ma makes for about 1.1k ?? how would I determine the wattage?

And won't the presence of the resistor make the plate voltage drop every time the current draw drops? I'm a little confused here.

basically I would be replacing with the resistor and then of course just watching the bias voltage on the grids to see if they drop at any point?

any information on testing for oscillations?

thanks a lot guys!

[Enzo]

I'd be worried about arcs. Hipot testing is a great idea, take your OT in for test. ANy motor shop ought to be able to handle it, and I don't think it would cost a ton.

Also possible arc on a socket, look them over - over and under.

[MitchK]

http://www.geofex.com/ampdbug/outtrans.htm

What about doing the "advanced" tests on this page? seems like a pretty similar idea to the Hipot test?

Also I know the chain of events began without the speaker cable cutting out but when what seemed to have caused the current problem was playing the amp for an extended period full blast with just two tubes.

and once again word on running the amp through the variac, is this going to be ok even with the tubes in it?

[MWJB]

Variac Q - yes, this will be OK. Start at around 1/3 wall AC.

Fitting 1ohm cathode resistors to ground from the power tube sockets pin 8 will allow you to monitor bias current at pin 8 (in mV) whilst playing. It'll fly around to high readings under normal circumstances but might give clue at the point of failure?

You don't say what kind of plate currents you are running at the moment?

Before removing the OT, i'd check it in situ first. Use a bare 1/4" jackplug to apply 0.5VAC to the speaker jack (secondaries) with your variac, monitor with a voltmeter (dialling in 0.5VAC migt be a bit fiddly but be patient). Use a 2nd meter to read AC voltages, end to end, at the OT primaries (no tubes fitted). Divide the primary's end to end reading by 0.5VAC, then square it, then multiply by output impedance and you should get primary impedance for your tube type, divided by 2 (1800ohms-ish?).

You can also measure from one end of the primary to the centre tap, then compare reading to the other end of the primary to the centre tap to see if one side is shorted.

-37vdc at pin 5 with KT88's? Is it just me or does this sound a bit low?

[MitchK]

Variac Q - yes, this will be OK. Start at around 1/3 wall AC.

Fitting 1ohm cathode resistors to ground from the power tube sockets pin 8 will allow you to monitor bias current at pin 8 (in mV) whilst playing. It'll fly around to high readings under normal circumstances but might give clue at the point of failure?

You don't say what kind of plate currents you are running at the moment?

Before removing the OT, i'd check it in situ first. Use a bare 1/4" jackplug to apply 0.5VAC to the speaker jack (secondaries) with your variac, monitor with a voltmeter (dialling in 0.5VAC migt be a bit fiddly but be patient). Use a 2nd meter to read AC voltages, end to end, at the OT primaries (no tubes fitted). Divide the primary's end to end reading by 0.5VAC, then square it, then multiply by output impedance and you should get primary impedance for your tube type, divided by 2 (1800ohms-ish?).

You can also measure from one end of the primary to the centre tap, then compare reading to the other end of the primary to the centre tap to see if one side is shorted.

-37vdc at pin 5 with KT88's? Is it just me or does this sound a bit low?

Can you elaborate a little more on the equation there? I've really really tried to get the variac to 0.5vac but the closest I can get is 0.643, can you give me the formula so I can just change it, I think I've already got it but I'm confused about the divided by 2 part and I don't know this relates to the primary impedance for my tube type?

[MWJB]

OK, I don't know exactly what primary impedance your OT is wound for, so I would guesstimate that it is wound for between 3.4K and 4K per tube (being an EL34 design, anyone who knows for sure please chime in). You have 2 tubes at each end of the primary, so I guess the the figure you want is between 1750 and 2000ohms. (If its 1695, for instance, don't panic, we are looking for a ballpark reading, if something is really wrong it will show up).

So if we work backwards:

1800 (our nominal target primary Z) /4ohms (voltages with the 16ohm tap will be much smaller and trickier to determine)= 450

Square root of 450 = 21.21

21.21*0.643VAC = 13.68VAC from end to end of the primaries. (13-14VAC with 0.643 at the 4ohm secondary would appear to be within tolerances, if you were using the 16ohm tap I'd expect to see 6.8VAC-ish, end to end on the primary with 0.643 on the secondary ).

Working forwards...

Voltage@Primary, end to end/voltage applied at secondary (ideally 0.5VAC)...

Square the result...

Multiply by output impedance at that tap = Primary Z.

[MitchK]

JESUS H CHRIST.

haha sorry guys just had to get that out. So this thread has been abandoned for a little bit but well this JCM 800 has been through a lot and I'd just make my day to have it back up and running.

So a short as possible recap:

amp got connected with no speaker load and cranked, this blew a tube and some fuses.

replaced the fuses and play the amp with only 2 tubes in it but cranked to 10. works for about a half hour and then cuts. HT fuses are blown

replace the HT fuses turn it on and it blows them again. turns out a tube is dead.

replace the ht fuses and a new quad a 6550 tubes (no el34s available, bad move for the purposes of trouble shooting but what can you do), works for about 10 minutes and the fuses go again.

Replaced all the power filter caps.

find out the bias trim pot is not giving as much as much resistance as it should so I assume the tubes had been over biased and this killed them, but upon testnig them with a tube tester they are ok. I replace the trim pot anyways and giver a try it seems to be fine until I turn it up a little whack out some full chords and the fuses pop again

finally I replace the OT thinking the OT must have gotten somehow wrecked from the original issue of operating it with no load. works ok reasonable volumes, I cranked it for a second just to see if the fuse would blow and it seemed ok. tried it again, this time with a guitar that had a "boost switch" for a little more gain and of course it again crapped out when i turned up to playing volume.

I am basically ready to take this to get serviced. I' started to take apart the original OT to rewind it so I guess I'll put that back together and put it back in. but is there ANYTHING anyone can recommend that I check before I send it in? its not just about paying to get it fixed, I feel defeated. I need something to play though, I'm ampless at the moment. ANYTHING though folks, ANYTHING!