JCM 900-4100 preamp question

[stratrat_57]

This is for "Enzo", or anyone else who may know......
I'm in the middle of sorting out a problem with my 900-4100, and slowly discovering (I hope!) how everything in the preamp section works together when functioning properly. This preamp section is the most confusing, convoluted signal path I have worked with, a hybrid mix of old school tube design and modern solid state op-amps. Also, sections of this preamp are not obvious in there function.
Q: What is the function of BR2? It's obviously tied to IC6. Is it to set some reference voltage?
Q: How are these op-amps, IC1,4,6 and 8 generally adding to the signal strength? When tracing the signal, what would be seen in a normally functioning amp?
My 900 has a very low signal/output. The best description I have is that of a 5watt amp! Both channels affected, clean channel is dirty. Have done all of the usual stuff, effects loop by-pass, new tubes, bias check and re-check. I've compared in-circuit signal values between the 900 and my 800-2204. At the input, my guitar signal is .050vac. In the 800 it climbs upward through 1-3vac at v1b, then at the cathode follower it is 40vac into the PI. This amp is working fine.
The 900 on the other hand starts at .050vac input and never climbs above 6vac at the cathode follower, all the way through the preamp. I just don't know what NORMAL values look like on this amp. I'm runnin' out of Tylenol!!

[techineer]

This is for "Enzo", or anyone else who may know......

Q: What is the function of BR2? It's obviously tied to IC6. Is it to set some reference voltage?
Q: How are these op-amps, IC1,4,6 and 8 generally adding to the signal strength? When tracing the signal, what would be seen in a normally functioning amp?
My 900 has a very low signal/output. The best description I have is that of a 5watt amp! Both channels affected, clean channel is dirty. Have done all of the usual stuff, effects loop by-pass, new tubes, bias check and re-check. I've compared in-circuit signal values between the 900 and my 800-2204. At the input, my guitar signal is .050vac. In the 800 it climbs upward through 1-3vac at v1b, then at the cathode follower it is 40vac into the PI. This amp is working fine.
The 900 on the other hand starts at .050vac input and never climbs above 6vac at the cathode follower, all the way through the preamp. I just don't know what NORMAL values look like on this amp. I'm runnin' out of Tylenol!!

BR2 is a hard limiter. The gain of IC6 can be adjusted from about zero to around 100, and its output is applied through R58 and C36 to BR2 and D10. If the signal at the right side of "C36" (the misidentified 4.7K resistor) goes more positive than about 1.5 volts (3 diode drops), then the upper left diode, D10 and the lower right diode conduct and limit the signal to about +1.5 volts. On the negative alternation, the other two diodes and D10 conduct to limit the signal to -1.5v. This is where the distortion is developed. This limited signal is passed on to the Boost Pre-EQ and IC4.

ICs 1,4,6 & 8 are odd birds... these are dual op-amps that act as an analog switch. Pin one selects which half of the IC is ON. In this amp, the pin 1 control line is connected to point C of the switching control circuit over at the right side of the diagram. When point C is HIGH, the signals applied to pins 2 and 3 of the IC are amplified and fed to the output pin 5. When point C is LOW, the signals appled to pins 6 and 7 are amplified and fed to output pin 5. Each half can have its gain set by the values of the external resistors. These op-amps are the key to channel switching in this amp.

From your description, I would suggest looking carefully at the control circuit for point C. If that appears to be functioning properly, then you must trace the signal through the channel paths, following the channel switching logic through the correct amps. The problem could be in one of the M5201s, but might also be in a 1458.

You HAVE checked the +/- 15volt supplies, right?

One quick check that might help.... since all of the op amps in this preamp have negative feedback, they should all have 0V at the output with no signal. If any one is more than a few tenths of a volt, suspect that circuit. Also check for hot chips with a finger.....

Good luck....

[stratrat_57]

Thanks techineer,
Thats the nitty-gritty info I was looking for, and with my minimal ss knowledge I would probably have never figured out on my own.
Yes, indeed I have checked the op-amp +/-15vdc power supply, and all is well with it. I will next check the "no signal" status of the pin 5's...as you've suggested. I also checked the switching voltages at pin 1 during chA/ chB switching, expecting to see volts/ no volts condition, but did not read this way? I will try again to verify this, since it seems it would be that simple....
I still suspect a simple problem with this pre-amp, I'm a firm believer in Ocsams Razor. Probably a high resistance solder joint in the signal path somewhere. All the pre-amp and power tube voltages appear nominal. Someone also suggested the OT, since these 900's use sub standard transformers. Just wish this thing was a Plexi!

[Enzo]

Isolate the problem.

Both channel are affected equally, so if power supply is OK, I'd look after the preamps. I'd consider V2, both channels run through it after the preamps.

But back to isolating the problem.

Plug a cord into the FX send and connect that to the input of some other amp. Now play. Does the sound from the other amp sound OK, or does it have the same symptoms? Likewise, pluig your guitar into the FX return jack. Does that sound OK or weak and distorted?

That tells us if the trouble is the powr amp or the front half.

If the preamp out signal is bad, you have a few IC stages and V1. I'd plug a good 12AX7 in V1 just to see if it is that simple. Failing that, you have three 5201 circuits and the 1458 output. At this point, rock the amplifier to crash the reverb springs. If they come out strong, then reverb output IC7 5201 is probably good and can be stolen to replace one of these preamp ICs for troubleshooting.

Always quick and often fruitful is going down the row of ICs and check each output pin for unwanted DC offset. I am not worried abiout millivolts, I am looking for severl volts or more. That is pin 5 of the 5201s and pins 1 and 7 of the 1458s. And there is one TL071 in the reverb circuit, and output is pin 6.

I don;t suspect IC6, since that would only affect the dirt channel. Of course if there is more than one thing wring, who knows.

APply a signal and see if it makes it through each stage. I tend to suspect the ICs that connect to the outside world before I suspect the inner ones, but that is only a rule of thumb.

DiD you check the B+ at the tubes?

I know you already covered some of these ideas.

On the other hand, if the preamp out is OK, then the back half of the amp needs attention. That comprises the FX return, the reverb stages, the master volumes, and the power amp proper.

Again, if it is in this half we want to further isolate the problem.

Stick a test signal in the FX return. Oh you can use the front if you want, I just prefer injecting signal as close to the problem as possible.

You already swapped the tube V2, but did you check its B+? Are all the tube element voltages OK? For that matter are the heaters lighting up like they should?

Earlier I suggesting crashing the reverb springs. Well? Strong or affected with the disease? If strong, that means everything from IC7 onwards is working and we are left with something in the IC5a and V2a area.

If you don;t get a strong crash, then the trouble is after the reverb. In cases like this I often just use my meter probe as a source of hum signal to inject. If i touch the grid of V2b, I should hear it out the speaker. Have the masters up some. Leaving the meter ground probe off makes a larger hum signal. How about at the grids of the phase inverter V3, get hum? And check the element voltages on that as well.

And IC8 is the final stage before the phase invereter. Touching pin 5 with your probe is like touching the grid of the PI through a cap. If you can inject hum at pin 5 of IC8, but not at the plate of V2b, then IC8 is suspect.


There, that ought to be something to chew on.

[stratrat_57]

Wierd.....the reverb appears to work in both channels but I can't seem to "crash" it by banging the cab. I hear it crash mechanicaly in the tank, but not through the speakers.....Keep on diggin'!
My mantra for troubleshooting other things is "know how it works normally before you try to find whats wrong with it". It's really nice to have knowledge of "normal" voltage ranges and values for the equipment your working on. Since the 900 seems so different from other more traditional tube amps, it's tough for me to assume what "normal" should be.

[techineer]

Have you checked the reverb coils for continuity?

Plug an RCA cable into each jack on the tank and put an ohmmeter across the pin/shell at the other end. The reverb coils should measure a few ohms to a couple hundred, depending on which coil you are measuring. It's fairly common for the wires inside the tank to break between the coil and connector.

If those are good, then it sounds like the problem is in the reverb recovery circuit. A quick test is to turn up the reverb controls, then disconnect the cable plug from the 'OUT' jack on the tank. Touch the center pin of the reverb cable. If you get a loud buzz, the recovery circuit is probably OK.

[Enzo]

Well, let's remember that his complaint is low output, not a reverb problem.

[stratrat_57]

Update: With a patch from effects send to input of my JCM 800, the signal seems strong and quite loud at moderate volume settings. I will proclaim this section of the pre-amp, before the effects loop, normal.
With my guitar plugged into the return jack, the sound is still weak and strangled sounding. The MV needs to be maxed. Now I'm not sure how it should sound with an input at this stage of the amp, but I'm betting it should sound stronger, and the problem lies in the second half of the pre-amp or the power amp section.
Can I use the Direct Out jack to isolate the entire pre-amp from the power amp? No one has mentioned this in any posts that I could find. Just patch the D.O to input of JCM 800? That would narrow it down sugnificantly!
I hope it's not my O.T, but if so, a good opportunity to install a better one.
Cheers

[Enzo]

If you want to know how a guitar ought to sound plugged into the FX return, just try it on that other working amp.

[stratrat_57]

Enzo, my JCM 800 does not have an effects loop. Can I use the D.O jack on the 900 to input of my 800-2204?

[olddawg]

You are making this much harded than it needs to be. Just plug a line level signal source (one channel of a bench amp tape send, and Ipod, a signal generator, the pre-out on another amp, etc) into the the main amp in or effects return of the JCM900. If it isn't loud and clean the problem is in the output section. BTW, have you run a cable between the effects send and effects return jack? Those jacks are notorious for causing problems.

[stratrat_57]

On the JCM 900, the effects loop does not devide pre-amp from power amp. I've allready done what you have suggested, thank you. An input to the effects return still has the second half of the pre-amp to go through, then it enters the power amp just after the tap off for the direct out line. If I could use this output to run to a functioning amp, THEN I could seperate pre-amp from power amp. I,m not making this more difficult, I'm trying to make my job easier by isolating this thing section by section, as others here have suggested. Still no answer on the D.O?

[Enzo]

The DO is an output, so why not just try it? The DO is the signal entering the PI.

If the reverb crash is not strong but the reverb sound comes through clear, that is good enough for me. At least usually. If the reverb signal comes out clear, that means IC7, V2b, IC8 must be working.

The reverb return jack is just an inoput like others. Apply a test signal there if you like.

In the second half of post #4 I suggested a few things. The loop return through PI input section of the amp. Either by signal injection or signal tracing (audibly or scope), follow the signal path through IC5a, V2a, V2b, IC8. Where does it die. I won;t get pedantic over whether that section is part of the preamp or not.

And if for some odd reason the reverb comes through OK, but the dry signal doesn;t, then I'd want to check R20 2M2 resistor for opens.

[olddawg]

Here's a link to the owner's manual: http://www.drtube.com/schematics/marshall/jcm900.pdf
The Direct Line Out is just that. A line level preamp signal out. If it cleanly and loudly drives another external power amp (main amp in, tape monitor in, power amp input, etc on another unit) the pre-amp section is good.