JTM45 build presence oscillation

[casey73]

I just finished building a JTM45 clone. I got the chassis from AES because they had the chassis and front/back faceplates for a great price. I got most of the parts I didn't already have on hand from AES, including a Hammond OPT and choke. But I had to get the power tranny from Mojo because it was the only one that fit the chassis. I took my time and it fired up first try. I've tried it with VA KT66 and VA EL34 power tubes. It's my first JTM45 experience and it lives up to the reputation as far as tone. I went with the low voltage power tranny (300-0-300) and get around 390 VDC on the plates. Playing a Les Paul with Dimarzio 36th Anniversary PAFs it starts breaking up before noon.

There are two issues I'm having. One is when I turn the presence up past around 7, I can hear a faint high pitched oscillation depending on where the tone controls are set. The other issue is when cranked up I can hear some ghost notes in the background. I went with original specs and I know modern amps have more filtering than the JTM45 did. I didn't scrimp and used the best parts I could buy, Cliff, CTS pots, F&T filter caps, Sozo, Carling switches etc. The presence feedback resistor is connected from the board via a short jumper from the turret to the 16 ohm tap on the OPT impedance switch.

Any tips on layout and how to address the ghosting issues?

[Chuck H]

Regarding the oscillation. Lead dress is CRITICAL with this style of amp. Make sure the presence circuit lead from the impedance switch is routed away from the preamp leads. Make sure grid leads are not longer than they need to be. Make sure OT leads (primary AND secondary) are also not longer than necessary, twisted together and routed away from the preamp.

Regarding the ghost notes. Double check your wiring for the power supply. If the power supply is wired correctly with good working caps it shouldn't ghost. You may have a wiring error that is preventing the necessary filtering and/or decoupling.

[Justin Thomas]

All I can say is, switched/reversed plate leads on an OT are not "guaranteed" to cause wild howling or screeching. The bad effects can sometimes be subtle. Hey, it's free, quick, and easy to check...

Justin

[bob p]

assuming that you've signal traced the amp, what's the frequency of the oscillation and which stages have it?

[nsubulysses]

I was gonna say can you remove V1, then V2, then V3 to start honing in on what stage is oscillating. Seems like it would be visible on a scope, or if you don't have one just set the amp up to the edge of oscillation and listen through a speaker (probably easier anyway) and start chop sticking some wires around to see if moving any of them makes it better or worse.

[pdf64]

There are a few schematics described as JTM45; please provide a link to the exact one used.
Some of these don't have power tube grid stoppers; if that's the case with yours, suggest that something in the range 1k5 to 10k is fitted to each power tube control grid socket terminal, 5k6 being the classic value with these.
As noted, good practice with balanced systems such as heaters and power amp signals is to tightly twist the wires carrying the balanced signals.
Some detailed photos of the chassis internals may help with remote diagnosis.

[cjenrick]

you must have done something right because that JTM 45 is famous for the Marshall Squeal.

maybe not turn the Presence up so much?

is the amp in the cabinet? does it have a steel bottom plate?

how far from the amp were you when testing for squeal?

get some chopsticks and move OPT leads around while squealing.

maybe raise the ends of the grid stoppers up in the air like they do on a lot of marshall amps.

try changing resistor values in the Presence circuit, sometimes you see two different values printed on the schematic, this could mean they ran into the same thing at the factory and revved the value.

try a lead from the speaker frame to the amp chassis.


OT: want the Blues Breaker sound? sport down for some brown base Genelex tubes for insane midrange detail like you have never heard in your life! yeah there $$$ but they are the dope. just a question of how bad you want good sound.

[mozz]

As was said, these are known to squeal. One schematic it puts DC on the presence pot, there is a mod for that. Just built one last year and it squealed above 7-8.

[alexradium]

i tried recently a replica with KT66 and mercury Magnetics irons,stock circuit,and sounded perfect at any volume,no squeal no noise.
For sure you have to check out the feedback wire from the board to the presence pot,keep it further away from the preamp,but having so much feedback,27k on a 16 ohm tap,is also a problem ,it could lead to oscillations,i'd try a 100k and maybe even a lower tap just in case.
Post pics of the build.

[nsubulysses]

As noted, good practice with balanced systems such as heaters and power amp signals is to tightly twist the wires carrying the balanced signals.

When you say power amp signals do you mean primary and secondary wires of output transformer or something else??

[pdf64]

I meant that any wires carrying balanced signals, ie identical signals of opposing polarity https://en.wikipedia.org/wiki/Balanced_audio should be twisted, in order to minimise their susceptibility to interference and their radiated field.
So in a regular p-p tube amp, the wires between phase splitter outputs and power tube control grids, and those between the power tube plates and OT primary; and the heaters.
Hope that makes sense now, apologies for phrasing it poorly before.

[nsubulysses]

No worries thanks for the answer. I never thought to twist the PI --> screen grid wires. So now I'm a little smarter

I also wondered if you possibly meant the HV power supply node wire and the ground return wire from cathode of gain stage to cathode of power supply cap

[Chuck H]

I never thought to twist the PI --> screen grid wires.

Me either. I always connect the PI outputs to the control grids

I don't imagine twisting the screen wires is necessary. And I've never twisted my PI output leads, but it makes sense. A lot of guitar amp desgns include a snubber cap across the PI outputs, but some don't. I think the twisted leads could contribute more stability than the cap with less HF attenuation. Just the thing if your non snubber clone or design is misbehaving

[pdf64]

Can the ghost notes be described in more detail (eg akin to a ring modulator) or better, clips of the issue be recorded and made available?
Is the issue worse/same/better with EL34 (compared to KT66)?

[soundmasterg]

Every amp I've ever seen that was a Marshall or a similar clone such as an early Traynor will oscillate like that when the presence is turned up. Some do it when the treble is turned up too. I couldn't see anything on a scope the last time I worked on an early Traynor for someone and it was doing that. I just followed best practices for lead routing and was able to minimize it. Each circuit is different though so if could link to the schematic that you used, that would certainly help in troubleshooting.

Greg