JTM45 Build

[martin manning]

Ouch is right! The unequal widths look like an overdriven PI, with lots of crossover distortion, which would be consistent with the bias change. The upper (or lower?) half is phase shifted though, which seems odd.

Moving the tone controls changes the frequency content downstream, which includes noise in addition to the 1kHz input. I don't know what BW you had there, but it looks like there is something much higher riding on the waveform.

[Littlewyan]

well it must be higher than my hearing as with the amp in the head cab its dead quiet. The only signal I found with the scope with nothing plugged in was a 8Khz one. But I would hear that surely. Also as it still occurs even with the NFB disconnected would that not put it to the tonestack or earlier in the circuit?

Edit: The issue occurs on the normal channel as well just not as bad. So I think we've narrowed it down a bit.

[martin manning]

Also as it still occurs even with the NFB disconnected would that not put it to the tonestack or earlier in the circuit?

Not if it is happening inside the feedback loop. Edit: ...and there isn't enough FB of HF to kill it.

[Littlewyan]

Well I just prodded every wire in the amp and nothing crackled or anything. I'm not sure what the next move would be :/

[Reeltarded]

33k stops and it's done. You will never know they are there. It's not evil. It's called for. Hypersonic treble is a tone killer. No sustain and no audible cues as to why. Just falls off into 'uuuub'.

I think the amp is screaming ultrasonics.

Your Express is a magic trick. This amp is pretty typical. The Express has created an expectation that may never happen again in your whole life.

Dunno what else to do.

[martin manning]

Another strategy is clipping a cap across the PI (in addition to the fizz cap), or across one or the other of its anode load resistors, to see if you can make something happen. You might learn where the problem is without having to solder/unsolder anything. A couple of tiny croc clips soldered directly to the leads of a 500pF-1n cap would make this easy, but be careful.

Miles are you saying add grid stoppers to the preamp stages at the sockets?

[Littlewyan]

I just added in the additional 10k dropper on the HT, partly to help out the pre amp valves as the voltage is a tad too high for them and I read this can help, but it didn't anyway! Shortened one of the power valve grid wires, same issue. Looks like I may have to resort to grid stoppers.

One interesting thing I found was on the schematic for Marshall's JTM45 Offset special amp, they put a 100pF Cap across the anode of V2A. Perhaps they ran into a similar issue?

I really don't think grid stoppers will solve the issue though as I can make it disappear by turning up the Middle control? I'm leaning towards it being a ground issue based on that.

[Reeltarded]

When you dump the mids the treble rises because there is headroom available for that RF.

It's RF.

Yes, Martin. I don't own a Marshall that doesn't have 33-68k on the input stages, at least 33k on the second stage, and I even have one with 10k on each side of the PI.

When I saw that pic of the kinky treble it looked like what it sounds like when you turn up a superlead next to a naval air station. "Rgr CN tower, niner-two-two clear and rolling. Have a good day.".

[Littlewyan]

Yes but even new Marshall reissues don't come with grid stoppers and they work? I used to have one, but if I have to resort to it then so be it.

I could still shorten the other power valve grid wire and another thing that comes to mind is the 1Ohm Bias Resistors. Would they perhaps cause an issue?

Also I just plugged the amp direct into the speakers instead of using an attenuator, turned everything up on full and it hissed a lot. I don't know how normal that is with old Marshalls, but the main cause of hiss seemed to be the Presence Control. The Treble control affected it but not very much.

[Reeltarded]

hmm.. yeah, they sound like a waterfall dimed all controls. 3 100w amps dimed is about as loud a hiss as a 1500w hair dryer in your face.

[Littlewyan]

I always wondered how Hendrix was able to stand in front of three 100W amps going through 24 speakers. I think my 35W Express through 2 speakers is ear blistering!

I wonder if the 250pF Treble cap is bad, either that or the 560pF Cap bypassing the mixer resistor. The treble cap being bad would explain why the issue is less prominent when I turn the middle and bass up as the signal now has another path. Whereas with them turning down and just the treble up it has no choice but to go through that cap. What do you think?

[joeboo88]

Ryan. Hope you get this sorted out. If you need any measurements from my JTM let me know. Will the JTM 50 sound much different from the jtm45?

[Littlewyan]

Thanks Jooboo88. They aren't much different, mine just runs EL34s and has 470K mixers.

Just tried replacing the 250pF Treble Cap but still got the same issue . So strange that it doesn't occur when playing chords on the low E but if I hold an A chord and whack it then it happens.

Tonight I'm hoping to get my brother to play guitar whilst I poke around inside the amp.

Edit: http://www.amparchives.com/album/Marsha ... 48375.html

Going by this JTM45, Metroamp's JTM45 and suggestions I have created a to do list.

1. Move Presence wire under the board
2. Move Normal Volume wire under the board
3. Check Pre Amp Socket pins are making a good connection
4. Check for small voltage on the coupling capacitors
5. Move wires around whilst brother plays through the amp
6. Replace the 560pF Bypass Cap
7. Shorten V2 Cathode Wire
8. Move jumper leads for tone pots to above the pots
9. Up the PI fizz cap to 100pF
10. Add grid stoppers
11. Try a 100pF Cap across the anode of V2 (Marshall did this on the JTM45 Offset Reissue)
12. Replace bias diode
13. If all else fails check into mental asylum

I really cannot see anything else it can be.

[rp]

It's wired up so nice that you are forced to go through all the annoying and usually futile basics first - bad tube, bad mica, wanky filter or coupler, wrong resistor, cold solder, etc. before you mess up your nice wiring. But most of the time it's not so easy as a wrong value resistor somewhere and you are going to pull your hair out. My problem when I used boards was I used a non standard chassis and layout (Traynor butchering days) and I got oscillations, which will really take the wind out of your sails when it's your first build attempts. You are using a standard Marshall chassis, parts, and layout, it has worked for zillions of originals and clones - so hang tough, it'll come together. Might want to stick it on a shelf for a few days.

You mapped out all the voltages, right - made a nice clear chart? You printed out a schematic and checked off every part?

[Littlewyan]

You're right RP. I checked the amp with the schematic tonight and everything is in the right place. Checked the voltages, please find voltage chart attached, everything looks ok.

Anyway I was checking for leaky coupling caps and found that when I measure the treble cap or the 0.022uF Cap that goes to the other side of the treble pot I read 0.02V or if I set my multimeter to 2000m I read .005V. I also measure exactly the same on the PI 0.022uf Coupling cap. I believe any DC on the other side of a cap is bad news so we have a bad cap here I think. I think its the PI Coupling Cap as if I turn the treble down to 0 the voltage goes down. I'll swap this and see.