Marshall JCM 900 fried! Need help :)

[J M Fahey]

And then reading bias voltage at the junction R26/27 .
Not contradicting Stan at all, but trying to know first what do we have available.

He suspects not enough bias voltage reaches actual grids, which is entirely possible, I would first want to know whether said bias voltage actually exists to begin with

Either a leaky bias capacitor (C13/14?) or a wrong bias voltage dropper (C15?) may conspire to bad bias.

........................................................ .

Note: measure bias voltage at the resistor junction without power tubes, just to play it safe.
Also measure bias voltage at each of the tubes grid pin, what Stan suggests.
It will show lower than normal, because the meter input current drops voltage across the bias resistors, but it's a fair indication, specially about path integrity.

Still no answer

As of:

Be careful, it's over 300VAC there and is not referred to ground but floating.

if you want the longer explanation (besides the warning which is absolutely real), is that technically each end is ground ... and 340 VAC , the other one the same ... and they switch roles 100/120 times a second.

So when it's ground , no problem but each end is 340V RMS above ground 50/60n times a second.

As deadly as if it were there all the time.

And the otherb point is that you can't consider one end safer than the other, both are deadly.

That's why I ask you to be careful about the red probe and the black one too.

Many subconsciously look with 4 eyes at the red one and grab it with the right hand (unless you are a lefty), and are not that concentrated on the other hand.

FWIW an Industrial Safety "11th Commandment" is: "it's not the big accident which will kill you, but the stupid unimportant one"

I already have 3 VERY skilled friends neatly packaged in fine examples of polished wood furniture, plus a legally blind one, because of accidents so stupid you'll say "you must be joking".

Oh well.

[sibo]

First of all, I had a very busy week so no time were left for these kind of activities x(
I did, however receive those brand new 50+50 caps but I will change them once we pin out what the problem really is so the new ones don't get burnt - unless you want me to replace them..

What I ask in 3) is the AC voltage across Red to (Red?) wires which feed main HV rectifier bridge D2/3/4/5 .
Be careful, it's over 300VAC there and is not referred to ground but floating.

And then read DC voltage across main filter cap (50+50x500V)

And then reading bias voltage at the junction R26/27 .

HV rectifier bridge (red to red): 116mVAC
This is measured at the point where the red cables enter the power section at the far most south-east.

Main filter caps
Green to blue cable: 347VDC
Green to yellow cable: 491.5VDC
This is measured on the one to the south. I did get my new multimeter and now when I measure it seems values aren't as jumpy as before.

Bias voltage (correct me if I measure it wrong here)
R26-27(common point) -> Ground: 4mVAC
It counted downwards from about 1VAC after power on.

How do I measure bias voltage at each tube grid pin? I recently measured between pin 2 & 7 on all EL34 sockets. Is that what you're after?

If there is a problem with bias, one single test point can determine if that is the source of the problem, testing noting else, measure the control grids, if it is right, move on, forget all the rest of the bias supply. If it not there, fine, you have narrowed the problem to a very simple circuit that can be tested with any VOM DMM or VTVM.

Thanks! How do I do that? I haven't gotten around to learning all those shorts but that's different types of meters, yes?

I already have 3 VERY skilled friends neatly packaged in fine examples of polished wood furniture, plus a legally blind one, because of accidents so stupid you'll say "you must be joking".

Oh well.

Sorry about that. I am waiting for my second daughter to be born so I will do everything I can to prevent any kind of injuries. I hope you'll understand when I ask things that may seem to be very basic and logic to you, but perhaps aren't to me

[pontiacpete]

"Bias voltage (correct me if I measure it wrong here)
R26-27(common point) -> Ground: 4mVAC
It counted downwards from about 1VAC after power on."

Turn your meter on to DC, you should have a negative voltage reading,,, if it's there.

[J M Fahey]

HV rectifier bridge (red to red): 116mVAC

Impossible (with the amp ON, of course).
I expect bover 300VAC there.
To be more precise, I expect measurements across the red wires which leave the transformer and with Standby ON .

Correction, what is in blue is wrong.
Saw one HVAC wire leaving the transformer clearly labelled as RED, didn't see a label on the other so assumed it was also RED (by the way an unofficial convention denoting high voltage/danger) but on rechecking the muddy schematic now notice that the rightmost wire has a circle drawn in its path (a wire tie/junction?) with what might be interpreted as BL (blue?) and the wire after the standby switch also seems to be labelled BL(ue?) so check whether 2 Blue wires reach the rectifier bridge , measure AC voltage across them with Standby on and post here.

This is measured at the point where the red cables enter the power section at the far most south-east.

Geographical labels are useful in maps and nowhere else.
In a schematic they are as useful as saying "the transformer which is closer to the kitchen".

Main filter caps
Green to blue cable: 347VDC
Green to yellow cable: 491.5VDC

Since no colours are shown there in the schematic, this is meaningless for us.
Yes, we may imagine what you are measuring.

Guess what?
I don't want imagination and deadly voltages mixed in the same sentence.

This is measured on the one to the south.

South?

I did get my new multimeter and now when I measure it seems values aren't as jumpy as before.

Bias voltage (correct me if I measure it wrong here)
R26-27(common point) -> Ground: 4mVAC
It counted downwards from about 1VAC after power on.

Bias is a DC voltage.

How do I measure bias voltage at each tube grid pin? I recently measured between pin 2 & 7 on all EL34 sockets. Is that what you're after?

No, but leave it there.

I am leaving this thread with 2 suggestions:

1) for Sibo: get a good Tech.

As of learning, fine, do it, but don't mix that with repairing this amp.

2) for all others: don't know you but if after 108 (one hundred and eight) posts we STILL don't have the bias measurement, which should be the first one made, then I suspect we are going nowhere.

Or we might reach somewhere, say, after 1000 posts?

And no, you are not helping sibo at all.

At best, all he'll pick is little bits here and there, which when applied somewhere else won't fit anyway.

Not even to another fried JCM900, go figure.

[Mickey]

Thanks both of u, I will replace all of them. Just out of curiousity - how can you know that C29 & C30 needs replacing? What do they do? Excuse these questions that may seem stupid but I'm novice.

Probable cause of them being too hot? Worn tubes?

S

Hi, caps don't usually smoke or get overheated unless there is another issue (say a higher than stated working voltage). Resistors can and do overheat because of excess current passing through them or they have changed in value. If the gain went high and then distorted that could be due to an increase in anode voltage which eventually destroyed that particular tube.
Remove most of the tubes except the output ones, take some measurements and compare them to the ones on the schematic. If this is within say 10% tolerance then fit the phase splitter/driver and so on untill you get to the stage with burnt or overheated components.
If you find that the phase splitter stage is working as per spec substitute each of the other unplugged ECC83s to see if any are defective.
Now if you're concerned about getting zapped by the HT (you should be worried) find a 100K to 220K resistor preferably 5W or higher and attach an alligator clip in seies with about 6" of hook up wire to either end and connect this from B+ to ground, you will probably find that you can leave this connected while you work on the unit, and although it won't discharge the caps completely it will make the amp much more service friendly.
Good luck and keep us posted. Mickey

[sibo]

Hi, caps don't usually smoke or get overheated unless there is another issue (say a higher than stated working voltage). Resistors can and do overheat because of excess current passing through them or they have changed in value. If the gain went high and then distorted that could be due to an increase in anode voltage which eventually destroyed that particular tube.
Remove most of the tubes except the output ones, take some measurements and compare them to the ones on the schematic. If this is within say 10% tolerance then fit the phase splitter/driver and so on untill you get to the stage with burnt or overheated components.
If you find that the phase splitter stage is working as per spec substitute each of the other unplugged ECC83s to see if any are defective.
Now if you're concerned about getting zapped by the HT (you should be worried) find a 100K to 220K resistor preferably 5W or higher and attach an alligator clip in seies with about 6" of hook up wire to either end and connect this from B+ to ground, you will probably find that you can leave this connected while you work on the unit, and although it won't discharge the caps completely it will make the amp much more service friendly.
Good luck and keep us posted. Mickey

Thanks for helping out but all of the "turn it in to a tech" and "rookies shouldn't repair amps"- kind of posts got me realizing it was a mistake asking for help here. When teaching someone you have to be able to lower yourself to their level of knowledge and the only one here that managed to do so was Audiotexan and since he got ranted due to a slower/less effective approach he's now gone. I know all of you took your time trying to help but this just doesn't work out as I simply don't find it funny/interesting anymore. Amp will end up getting trashed in a music video or so
We all have different approach but our goal was always the same so I want to thank each and everyone who helped out! <3

This tread is now dead.

[52 Bill]

Thanks for helping out but all of the "turn it in to a tech" and "rookies shouldn't repair amps"- kind of posts got me realizing it was a mistake asking for help here.

Sorry that you feel this way. Perhaps you could contact Audiotexan privately and pay him to help you long-distance troubleshoot the amp away from the forum.

With only a few exceptions, my experience here is that most everyone posts things that are meant to help advise, teach, train and direct others to find a solution to a problem. There are lots of different ways to do things and even more ways to say things. In the world of web forums I find that unlike a few other musical equipment related sites, the general tone here is supportive and helpful.

Troubleshooting an amplifier or an automobile or a toaster requires a certain amount of logical thinking. What is the problem, what could cause the problem, what specific measurements will help to narrow down the source of the problem, etc. Sometimes too many people suggesting too many things make it hard to do, but you need to try and stay focused. Listen to all sides and then pick one logical path to follow to reach your goal.

Good luck with whatever you decide to do. And if the amp is still overheating it might make for a very cool pyrotechnic effect in the video.

[Jazz P Bass]

I for one, have never considered MEF to be a site for totally inexperienced individuals.

That's is how I take it.

To actually assist someone that has done all that they can in a proficient manner is a good thing.

To teach someone how to read a volt meter?
Meh.

[Audiotexan]

Thanks for helping out but all of the "turn it in to a tech" and "rookies shouldn't repair amps"- kind of posts got me realizing it was a mistake asking for help here. When teaching someone you have to be able to lower yourself to their level of knowledge and the only one here that managed to do so was Audiotexan and since he got ranted due to a slower/less effective approach he's now gone. I know all of you took your time trying to help but this just doesn't work out as I simply don't find it funny/interesting anymore. Amp will end up getting trashed in a music video or so
We all have different approach but our goal was always the same so I want to thank each and everyone who helped out! <3

This tread is now dead.

I made some mistakes (at least one was crucial) of my own here, along the way, and it was a learning experience for me in several ways as well.

1. in post 74 where I recapped... your voltages were still way high.
I *should have* caught that. But blinded by a couple different things, I didn't.
That was a gross error on my part. And I should have stopped things cold in their tracks right there (troubleshooting wise -not helping wise).

Had it registered that you were still showing over 500vdc in those checks...I would have gone into more basics (possibly much to some's chagrin).

At any rate, as a rule of thumb, it's a MASSIVE clue (that I still missed the slap in the face of) that you should NEVER have a higher voltage reading than the highest value cap ("blue cup" as you called it IIRC lol) in the amplifier.
In your case, the 500vdc PSU mains caps. When you came back with 538v...I should have known instantly that something was awry. And I'm STILL *beyond* embarrased to have missed that. And it's only that going back several times and reviewing this thread several times over the past month that I realized how badly 'hell bent' I was on helping, that I let it override logical progression, and obvious warning signals. So again, Juan was right in a respect. I wasn't helping in the manner I had hoped.

Process and steps are important (CRITICAL even) but only IF you let the results 'register' as you progress. So while I can do this easy in RL, Bill's absolutely right, and it's an even finer 'art' to being able to convey it across the internet, much less across languages/geo-boundaries.

There are some fundamentals that you just have to have in place, part of which is having the proper test tools, and equipment to be able to have the ability to diagnose. And I believe in this case in particular, without those, you're going to have a very hard time.

Getting accurate readings is at the forefront. And without being able to have known inputs (say for example, having the amp on a variac so you can regulate the input to a 100% consistent voltage, while trying to take readings) and then having a True RMS meter, so that you don't have to do the 'math you don't know' (.707 x "your actual" reading to get an RMS value)...it's going to be hard at best.

We can spot burnt components, and guide you through the basics all day long, but if you don't have the hardware, and basics, we can only help so much.
No matter how much we desire to help. And I trust you'll understand this, as you KNOW how badly I wanted to help. But I'm not up to speed as some of the vets on here in that department, and it's a skill I'm quite obviously still learning. So I appreciate your kind words, and following along safely. As at least you're still safe and healthy.

I do apologise for letting you down though. I learned quite abit in this thread myself however. Possibly more than you. So regardless, I appreciate you being here, and *trying* to learn, and wanting to learn. But perhaps, as was said by some others, might try jumping into shallower waters first.

Keep this on the back burner. It's worth holding onto (obviously), and you still have the hopes/desire to fix it. You'll get there if you keep the attitude of wanting to learn, and putting the effort behind it. Personally, you're off to a roaring start!

Just run a 'few laps around the park' before you try to tackle the damned New York Marathon brother. lol

All the best Sibo!


BTW: Just like RL man, don't put any weight to the attitudes that *truly* oppose you, or you let them win. And don't confuse those that are just saying 'walk before you run'...There are MANY fine people at MEF, and some are just cutting to the chase out of sheer 'repetition'. That's just human nature. Once you've seen the same thing so many times, you tend to get jaded by it and 'put out/annoyed' and that shows through over their insight/wisdom. Which is a shame, but, we're all human. But there's a ton of experience and wisdom here!
Keep at it!

[sibo]

I made some mistakes (at least one was crucial) of my own here, along the way, and it was a learning experience for me in several ways as well.

1. in post 74 where I recapped... your voltages were still way high.
I *should have* caught that. But blinded by a couple different things, I didn't.
That was a gross error on my part. And I should have stopped things cold in their tracks right there (troubleshooting wise -not helping wise).

Had it registered that you were still showing over 500vdc in those checks...I would have gone into more basics (possibly much to some's chagrin).

At any rate, as a rule of thumb, it's a MASSIVE clue (that I still missed the slap in the face of) that you should NEVER have a higher voltage reading than the highest value cap ("blue cup" as you called it IIRC lol) in the amplifier.
In your case, the 500vdc PSU mains caps. When you came back with 538v...I should have known instantly that something was awry. And I'm STILL *beyond* embarrased to have missed that. And it's only that going back several times and reviewing this thread several times over the past month that I realized how badly 'hell bent' I was on helping, that I let it override logical progression, and obvious warning signals. So again, Juan was right in a respect. I wasn't helping in the manner I had hoped.

Process and steps are important (CRITICAL even) but only IF you let the results 'register' as you progress. So while I can do this easy in RL, Bill's absolutely right, and it's an even finer 'art' to being able to convey it across the internet, much less across languages/geo-boundaries.

There are some fundamentals that you just have to have in place, part of which is having the proper test tools, and equipment to be able to have the ability to diagnose. And I believe in this case in particular, without those, you're going to have a very hard time.

Getting accurate readings is at the forefront. And without being able to have known inputs (say for example, having the amp on a variac so you can regulate the input to a 100% consistent voltage, while trying to take readings) and then having a True RMS meter, so that you don't have to do the 'math you don't know' (.707 x "your actual" reading to get an RMS value)...it's going to be hard at best.

We can spot burnt components, and guide you through the basics all day long, but if you don't have the hardware, and basics, we can only help so much.
No matter how much we desire to help. And I trust you'll understand this, as you KNOW how badly I wanted to help. But I'm not up to speed as some of the vets on here in that department, and it's a skill I'm quite obviously still learning. So I appreciate your kind words, and following along safely. As at least you're still safe and healthy.

I do apologise for letting you down though. I learned quite abit in this thread myself however. Possibly more than you. So regardless, I appreciate you being here, and *trying* to learn, and wanting to learn. But perhaps, as was said by some others, might try jumping into shallower waters first.

Keep this on the back burner. It's worth holding onto (obviously), and you still have the hopes/desire to fix it. You'll get there if you keep the attitude of wanting to learn, and putting the effort behind it. Personally, you're off to a roaring start!

Just run a 'few laps around the park' before you try to tackle the damned New York Marathon brother. lol

All the best Sibo!


BTW: Just like RL man, don't put any weight to the attitudes that *truly* oppose you, or you let them win. And don't confuse those that are just saying 'walk before you run'...There are MANY fine people at MEF, and some are just cutting to the chase out of sheer 'repetition'. That's just human nature. Once you've seen the same thing so many times, you tend to get jaded by it and 'put out/annoyed' and that shows through over their insight/wisdom. Which is a shame, but, we're all human. But there's a ton of experience and wisdom here!
Keep at it!

Thanks! I really don't take it personally as I really don't think it was anyones intention to "get to me" and if so - they sure didn't .
I'm sure they had their reasons to say when something's getting out of hand and I can only see this through my end so I wouldn't possibly know.

On a sidenote - A while back I did the exact opposite of what you suggested, I changed filter caps, inserted power tubes (got ready to outrun the explosion and fired her up. Worked excellent on channel A for like 5 minutes but as I switched to ChB there came a tick and the burnt smell returned.
Guess that narrows it down for those of you interested in what might've been wrong in the first place.

[audiopete]

I think the real goal here for you is to learn to read schematics so you can look at the malfunctioning parts of your amp and trace the entire circuit to find the problem. Being told how to hook up a DVD player is different then understanding where the wires go and why. The only way you are going to get good at troubleshooting (besides experience), is by learning schematics. They are the roadmap to your circuits and allow you to trace and isolate issues till you find the culprit(s). Good luck.

Here's a good book to read to learn more about tubes in easy to understand format - http://archive.anthemav.com/OldSitev1//pdf/tot1.pdf

As an aside, if you are wondering what it's like to take a jolt from charged caps, it's not that bad, provided the amp is unplugged. They don't have a lot of capacity (current flow). I'm not saying it couldn't be bad, and I'm NOT suggesting you try it, but I've been shocked by caps and it was no worse than a shock from the wall outlet in terms of pain or feeling. Again, keep in mind the amp was unplugged from the supply. If it is live and operating then that is a whole different story. You will keep getting shocked until you break contact or a fuse blows. Of course there is a difference between AC and DC, and again I am not suggesting you try it, but know that it's not supernova time if you get shocked by a cap in an unplugged chassis. I did have my other hand behind my back at the time. You don't want current goin from one hand across your chest to the other and stopping your heart in the process. Even microamps can upset a heart beat.

Power supplies can isolate caps from discharging, so be sure to check all the caps when discharging. That's how I got jolted

[audiopete]

On a sidenote - A while back I did the exact opposite of what you suggested, I changed filter caps, inserted power tubes (got ready to outrun the explosion and fired her up. Worked excellent on channel A for like 5 minutes but as I switched to ChB there came a tick and the burnt smell returned.
Guess that narrows it down for those of you interested in what might've been wrong in the first place.

Maybe look into the switching system for burnt components. Makes no sense that switching signal would burn out something in the preamp signal path. Check that the switching ICs (M5201) all have +15VDC on pin 8 and -15VDC on pin 4 and that the control pin 1s are getting switching signal when you stomp the footswitch. Check the switching circuit itself against the schematic to confirm voltages. Check points C (negative side of C29), B (anode side of D7) and E (cathode side of D69) in the switching circuit (top right of preamp schematic) for proper switching operation when you stomp the footswitch.