Diagnosis of Marshall 8100 Valvestate head problems

Hi guys & gals,
I need technical help!.....I had a problem with a Marshall 8100 Valvestate head, blowing fuses. After checking power supply, bridge rectifier, etc...it was found that power transistors TR8 & TR10 were shorted(see enclosed attachment for scematic)....
After removing TR8 & TR10 from the board, the amp now turns on and has sound, volume, tone, etc, from preamp all operating fine from the FX loop out, which I verified by plugging this signal into an external amp....so that was good news to this stage.....
I was working with someone on the Marshall site, who has been excellent with helping me through this, but he hasn't been able to continue with the diagnosis due to time constraints.
I'm fairly new to electronics(DIY), have done some testing of components, soldering, DC voltage checks, etc.....BUT I really don't understand the complicated circuits of an amp like this from reading a schematic. I can understand the symbols, values, and find them on the board. I just don't know how each part affects the other parts of these circuits....thus my dilema in correctly diagnosing why I am having trouble trying to check throughout for other possible components that may be bad also, before putting new TL072's back in......I'd like to learn more about the testing and eliminate blowing the new ones if something else is wrong......which appears to be the case based on polarity problems as shown below....these are tests that were requested from my friend in the Marshall forum. I'm testing blindly, not knowing where to look for the problem:

This is the scenario:

1) TR8 & TR10 still out of the board
2) Board is still out but connected for powering up
3) I left off by checking voltage, referenced to gnd at various points to check for other potential problems and getting wacky polarity and uncertain of values to expect.....

Here goes...
1) voltage at R98(-40.7v), R99(-40.6v).....so that's not good
2) (power rails +VE, -VE) at R84/R85(+40.4vv), at R77/R75(-41.4)....good here..?
3) check at R102 (+15.4v & +41.3v), at R101(-15.4v & - 41.3v).....good here..?
4) check at R97 (+2.12v), R117 (+2.12v)....not good here..?
5) check at IC8 TL072 pin4 (-15.5v), pin 8 (+15.5v)....good here..?
6) check at IC8B TL072 pin5 (+2.11v), pin6 (+1.19v), pin7 (+13.57)...not good here..?
7) check at R25/C18 (+1.19v)......opp side of C18(0.0v)no input...good here..?
8) check at speaker out (+2.09v)...good here..?
9) check at IC8A TL072 pins 1,2,3(0.0v)....no input...good here..?
10) check board at C(rails) at TR8 (not installed)(+40v) & TR10(-40v)....good here..??
11) check BCE at TR9(still on board)(all read-40v)....not good here..??

Turned off, waited, then turned back on...now observed "polarity changes"(voltage values approx. same)
1) voltage at R98 & R99 (both now +v).....so that's not good
4) check at R97, R117 (both now -v)....not good here..?
6) check at IC8B TL072 (pin 5,6,7 now -v)...not good here..?
7) check at R25/C18 (now -v)......opp side of C18(0.0v)no input...good here..?
8) check at speaker out (now -v)...
11) check BCE at TR9(still on board)(now all read+40v)....not good here..??

Tried 3 more times and got same as last test throughout (R25/C18= -v)......?????????
Tried again, polarity back to original throughout (R25/C18= +v).......??????
Tried again and again and polarity changes at will throughout(R25/C18= +-v)......?????????
What could be causing the polarity changes at input R25/C18, etc...or am I thinking backwards.?
Does TR8 & TR10, not installed, have anything to do with this polarity weirdness...or is it all in IC8 that is causing all the havock....?

Please suggest where and what to test from here...with explanation of HOW do we determine where to look next.........Thank You!!! ..

When you have power transistors shorted on a high power discrete transistor amp, it's time to take it to a professional. And hope they don't charge you too much extra because you messed it up and made their job of fixing it so much harder.

I'm sure that's not what you wanted to hear, but it would have been so much better if you took it to a pro amp guy right from the start.

As it is a feedback circuit, you are unlikely to get consistent readings unless everything is working correctly.

I'd replace ALL the transistors in the power amp (the second page of the diagram), plus D4, D5 and R96, R118 R119 & C60 at a minimum.

After removing the old transistors, check for any resistors that look to be discoloured, & replace as required.

It would not hurt to test all the resistors with a multimeter to ensure they are within 10% or so of the stated values. While doing that, check all the solder joints to ensure they look smooth and clean & none have cracked away from the PCB or resistor lead.

Make sure you use heatsink compound and any insulators on the power transistors are properly refitted to prevent shorts - test with a multimeter before switching the power on.

rjenkins said:

As it is a feedback circuit, you are unlikely to get consistent readings unless everything is working correctly.

I'd replace ALL the transistors in the power amp (the second page of the diagram), plus D4, D5 and R96, R118 R119 & C60 at a minimum.

After removing the old transistors, check for any resistors that look to be discoloured, & replace as required.

It would not hurt to test all the resistors with a multimeter to ensure they are within 10% or so of the stated values. While doing that, check all the solder joints to ensure they look smooth and clean & none have cracked away from the PCB or resistor lead.

Make sure you use heatsink compound and any insulators on the power transistors are properly refitted to prevent shorts - test with a multimeter before switching the power on.

Click to expand...

This is the type of response that I'm hoping to get from some professionals on this site. I would like more feedback regarding WHY certain things should be replaced, checked, etc....I've already tested the components you've mentioned and they all appear to be OK.
The only bad components found thus far are TR8 & TR10......prior to replacing them, I wanted to be able to test and diagnose all potential components that could affect their demise, and learn more along the way.....this is a learning project that I hope will turn on someday. I started with an amp that did not even turn on, tested many components, found that those transistors were shorted, after testing transformer, DB1, many components, changed broken switch, etc......and got this operating right out to the FX out.....now I'm learning and repairing the power section with the components that actually are the problem, rather than just changing components that test out to be OK. My attempt is to isolate the actual bad components in each circuit.........Thanks for all your support and helpful suggestions.....
I'm also open to any other suggestions that may be productive to reach my ultimate goal of learning more and finally to get a repaired amp done by my own hands, with the help of the forum......
I've been successful at getting 5 other amps working, and all the fixes turned out to be fairly simple in the end......with many hours of testing and ruling out all the components that are not bad and zeroing in on the bad component or solder joint that caused the problem......it's ultimately science, but this one is just a bit more complicated and need everyone's help that feel qualified to assess the situation and guide me to real understanding....to progress my abilities for future projects....

hi wolf i can relate to and admire your sentiments, i have repaired a few amps myself mainly dj stuff its usually the output stage or power supply or on mixers the faders.and cd players the lasers,,
having stuff repaired is not really cost effective nowadays. £50 an hour is not unusual then theres carriage both ways,and of course the less people have stuff repaired the less repairers there are.
i am amazed by how helpful the various manufacturers are , you would think they would be a bit uncooperative, but quite the opposite
regards
dougal

Thanks tibbles...and you're right on about the repair costs, especially in this world of "just throw it away and get a new model", but I'd really like to tackle this thing and need knowledgeable assistance.....Marshall is not very helpful with something like this when you're in the USA, so I NEED the forum to get this accomplished.........I need the WHY and HOW to proceed from here...

You already got the help. I suggested a good option.

And Rjenkins suggested a good professional repair technique, probably the exact technique I would have used whan I was a pro amp repairer. But I would have added the technique of powering it up after the reapair with some current limiting devices in place, until the testing was done.

HOW is what you have already been told.
WHY is because that is the best possible way to go about it.

OK, points well taken to this stage.....and I hear you, but let's say I had a different mission, as a project.
Let's try a different starting point for communication as Hypothetical only

1) If I had a full supply of all the components necessary to build this entire amp from scratch and tested every piece individually and found that only 2 parts were bad, being TR8 & TR10(put 2 new ones back in the supply to avoid the 2 bad parts, now confident that I have 100% good components to start, would it be logical that if all components were installed properly, exactly to plan, with no poor solder connections, or bad traces, etc.......should the amp operate....?

Hi Wolf,

the problem is that semiconductors that test OK using a multimeter may not work as they should in circuit. The junctions can be damaged by overload or long term heating, resulting in further failures.

The cost of the rest of the component set is probably low compared to the power devices and the inconvenience of a repeat failure.

If you replace all the parts that could have degraded or 'aged' (like the power resistors & some capacitors, plus all semiconductors) you should have a few more years of problem-free operation.

From long experience, repairs done by people who only replace the obviously duff bits frequently do not last long, where 'full set' replacements are generally reliable.

rjenkins said:

Hi Wolf,

the problem is that semiconductors that test OK using a multimeter may not work as they should in circuit. The junctions can be damaged by overload or long term heating, resulting in further failures.

The cost of the rest of the component set is probably low compared to the power devices and the inconvenience of a repeat failure.

If you replace all the parts that could have degraded or 'aged' (like the power resistors & some capacitors, plus all semiconductors) you should have a few more years of problem-free operation.

From long experience, repairs done by people who only replace the obviously duff bits frequently do not last long, where 'full set' replacements are generally reliable.

Click to expand...

There appear to be 3 answers:
1)OK, so the short answer is Yes........ if nothing else is truly wrong....in an otherwise perfect situation....
2)Maybe.... because you can't truly trust your junctions, nor test each part with a multimeter to be sure of truly actual condition...
3)N0... because it may startup, but blow again, now or later, due to unknowns, age, etc.....bottom line being reliability & risk factors...

With all that in mind....I still would like a teacher for learning about the interaction of circuits....in this case an understanding of the power circuit of any parts any way directly related to TR8 & TR10.......and what they actually contribute to the circuit to fulfill the needs and create audio power in this stage of the circuit........I know that's probably a lot...but I'm disabled, have plenty of time, and plenty of patience, even if it's just a very small piece of knowledge at a time.......if anyone has the time/patience to actually be a teacher.........

Hello Wolf,

There are parts directly related to the current protection of the TR8 and TR10.
For TR8 the resistors R116, 117, 118 and transistor TR12 and D5 make a current limmiter.
For TR10 the resistors R96, 97, 115 and transistor TR11 and D4 make a current limmitter.
The "quiesing current" is set by the resistors R73, 74 and transistor TR9.

Greetings,
Bertus

bertus said:

Hello Wolf,

There are parts directly related to the current protection of the TR8 and TR10.
For TR8 the resistors R116, 117, 118 and transistor TR12 and D5 make a current limmiter.
For TR10 the resistors R96, 97, 115 and transistor TR11 and D4 make a current limmitter.
The "quiesing current" is set by the resistors R73, 74 and transistor TR9.

Greetings,
Bertus

Click to expand...

First and foremost...thank you Bertus for sharing your knowledge and happy to work with you once again....I appreciate your common sense approaches to better understand circuit design, which will be helpful for my progress, albeit at a snails pace...slow, but still possible to get to my destination if I don't get "stepped on" first.......
I'm studying these circuit fragments, see the 2 limiting circuits, and the quiesing circuit, which after trying to get definitions, appears to maintain a normalization or balance of current between the first 2 circuits mentioned....if I'm way off, please explain...
What is the function of R98 & R99 in these 2 circuits to further alter the output circuit....?
Is there a procedure for isolating & testing individual sections of the output circuit, once individual components are checked for obvious problems to get a feeling if a certain section is doing it's intended function...??.....as I have a polarity switching issue which is explained at the beginning of this thread that is very confusing to me....

Hello Wolf,

The "strange" behavior with the swinging voltage has to do with the fact that there is no good feedback to the input of the opamp.
This has to do with the absense of the power transistors.
The main feedback goes through R61.

For a little background information:
The transistors TR6 and TR7 make a level translator for the positive side,
as the output of the opamp can only go ±15 Volts and you want to have a output swing of ± 40 Volts.
The transistors TR5 and TR4 do the same for the negative side.

As for transistor TR9 the voltage accross C-E should be about ( (R73 + R74) / R73 * 0.7 ) = (1800 + 680) / 680 * 0.7 = 2.55 Volts.

Greetings,
Bertus

PS here is also a link to some amplifier basics:
http://sound.westhost.com/amp-basics.htm

bertus said:

Hello Wolf,



The "strange" behavior with the swinging voltage has to do with the fact that there is no good feedback to the input of the opamp.
This has to do with the absense of the power transistors.
The main feedback goes through R61.

For a little background information:
The transistors TR6 and TR7 make a level translator for the positive side,
as the output of the opamp can only go ±15 Volts and you want to have a output swing of ± 40 Volts.
The transistors TR5 and TR4 do the same for the negative side.

As for transistor TR9 the voltage accross C-E should be about ( (R73 + R74) / R73 * 0.7 ) = (1800 + 680) / 680 * 0.7 = 2.55 Volts.

Greetings,
Bertus

PS here is also a link to some amplifier basics:
http://sound.westhost.com/amp-basics.htm

Click to expand...

Now this is VERY useful info and some new clues for my study and potential understanding.....this might keep me busy for a bit, as I wade through this new territory......

I now physically "see" that the combination of components & subcircuits mentioned would have the main feedback through R68 back to IC8B, and that cannot function as intended without TR8 & TR10 on the board...

This probably is really basic & stupid, but trying to learn from scratch, I've been lost in finding "signal flow" in schematics, with no electronics education. A "lightbulb finally went on in my head". You just "pointed" me to the actual "signal", which has been the missing ingredient for even beginning to understand what's going on to get a sound altered & amplified. The +V & -V are the needed voltage rails required to power the multiple circuits to get the desired "signal" on the main line(centered line in schematic) through to the output jack positive lug(pin)........then to ground(sleeve)......simple concept, but totally baffled when one is not aware...............THANKS....I was chasing what I perceived to be signal, all over the schematic, instead of current flow, interactions, and feedback, powering circuits that alter your signal from the guitar input.....I just explained my newfound knowledge to my wife and she thought of a good "analogy" of dominos, set up by real pros, starting, splitting, reconnecting, scynchronizing, etc. to get to some final point of last domino being my amplified sound.......that's what happens when you don't "step on" the snail............just guide him slightly when needed and let him crawl at his pace to the destination...in this case...KNOWLEDGE.......great teacher!!!............

I do have one calculation question at this point. I initially assumed that * = X(multiply), but your final 2.55V, when refigured, the 2nd half of the equation calculates as R73(680) /(divided by) 0.7
Is that what you intended...? *=(divided by)...?
1800+680=2480, 680/0.7=971.429, thus 2480/971.429=2.553=2.55V, thus both symbols...... / & * = divided by, and x = multiply.....correct..?

I will have to study to see if I can figure out how you get the starting voltages and how your equation is developed.....I'll see if I can do that.
Thanks, Ron

Hello,

Let me phrase the calculation a littele different:

It is a voltage divider of R73 and R74.
The voltage accross R73 will be the B-E junction of the transistor TR9 (about 0.7 Volts).
When the Voltage accross R73 is 0.7 Volts you can calculate the total voltage of the divider.
So the total voltage will be:



Greetings,
Bertus

bertus said:

Hello,

Let me phrase the calculation a littele different:

It is a voltage divider of R73 and R74.
The voltage accross R73 will be the B-E junction of the transistor TR9 (about 0.7 Volts).
When the Voltage accross R73 is 0.7 Volts you can calculate the total voltage of the divider.
So the total voltage will be:

Greetings,
Bertus

Click to expand...

Sorry to go backwards, but I just edited my last post as you were doing this.........have to go take care of my Mom..see ya later...

Hello,

No worries. As long things are clear.

Bertus

Thanks for clearing up that equation.....I read it wrong first time.

Does my previous post make sense as far as the need for DC current through all these various components/circuits being the outside forces necessary to alter and amplify the input signal throughout the amplifier, with feedback circuits basically designed to keep "control" of certain aspects of each circuit to get our desired amplifier characteristics throughout each stage of the amplifier(tone, distortion, frequency response, FX, amplification, etc....)...is this a fair generalization to begin with in gaining an understanding of amplifiers, design, and making sense of schematics.....?.......

Hello,

As you might have noticed the behavior without the power transistors is unpredictable.
You will need a closed loop to see what is happening.

For information on the powertransistors I attached the datasheet.

As for "standard" amplifiers, they amplify each frequency the same amount.
For tone control, the amplification is frequency dependend.
The most extreme version of tone control is the graphic equalizer.
This has a lot of filters that can be put to amplify or attenuate the signal.

As for "effects", these are special for musicians.
They can consist of various parts like modulators (phase and amplitude),
clippers (hard and soft), delay lines (for echo and reverb).

Greetings,
Bertus

Hello Ron,

In how much detail do you want the preamplifier part explained?

If you want every opamp can be described.

Greetings,
Bertus