JMT45 kit build crackle and hum

[jude]

Hi Folks,

I have completed a JTM45 kit build from Mable Audio. The amp works ok but there are a couple of problems -

1. the more serious problem is that the bright input channel 1 constantly crackles and this goes up in volume when the channel 1 loudness is turned up without any guitar plugged in, the 'normal' channel 2 is fine, white noise only.

2. less serious but annoying, what sounds like ground hum only when a guitar is plugged in that mostly goes away if I touch the input jack of the guitar lead or the output jack to the speaker - I realise this may be EMI but possibly also a grounding problem

I have attached a photo of the guts of my build + some audio, the hum goes away when I touch the input jack and the layouyt

JTM45 Sound Test.m4a

. I have tidied up the filament wiring since taking the photo. All the resistors have been checked with a multimeter and the electrolytic caps with an ESR tester. Continuity of the wiring and turret solder joints with a multimeter.

I have not tried removing or swapping tubes but they are all new. It has 3 12AX7B pre-amp tubes and 2 KT66 power amp tubes. Everything is new but I guess it could be a bad tube. I have biased the tubes.

This is my first build and I only have a some electronics experience so looking for expert advice.

Any advice appreciated. Cheers Jude

[pdf64]

1/ try reinserting the V1 valve into its socket a few times, to try to clean any crud away from the contacts. If that doesn’t help try another 12AX7 in V1, one that you know works nicely in the V1 / input stage of another amp, ie a known good valve.
If that doesn’t help the bright channel’s input stage CC anode load resistor may need swapping.

2/ where have you got the ground wire from the IEC inlet connected to? But that type of symptom isn’t unusual, eg don’t your other amps do it?

More, better lit photos would be useful.

[jude]

Yes, thanks I did try cleaning the contacts already and also the sockets and tubes are new.

I have been searching the other amp forums and a few Metropoulos JTM45 builds came up in the one thread with a very similar symptom on the tone stack. Mine is obviously to do with the volume for channel 1, different pot but same crackling noise. For the three Metropoulos JTM45 builds the problem was the 250pF cap so I am guessing it could be the 500pF cap. I will try alternatively removing the V1 and V2 pre-amp tubes to see if it goes away to isolate the problem. It would be handy to have an LSR tester but I do have an oscilloscope so maybe I can test it that way or just desolder the 500pF cap or the 100pF cap on the channel 1 volume pot.

Ground wires are soldered to two lugs attached to the bolts of the main power transformer inside the chassis. Sorry new to amp builds so the IEC is? cheers

[pdf64]

You seem to be under the impression that new valves will be perfect. That is not the case. Indeed any part can fail in a variety of modes at any time. But that is especially true of valves, hence them being connected to the circuit via a socket, rather than being soldered in place.
Were the valves advertised and sold to you as being tested and warranted for low noise etc?
Clearly, as the kit is new, everything in it should be new, yet there is a noise issue whose level is controlled by the bright channel vol control. Why shouldn’t it be due to a bad valve?
My reading of your first post is that the noise stops if the bright channel vol control is turned to minimum; if that is the case, then the issue has already been isolated to the bright channel input stage, any further isolation tests are pointless.
By IEC socket, I meant the mains inlet socket, ie that the power cable from your wall outlet plugs into.
The bolts that pass through the lamination stack of laydown style transformers and retain them to the chassis are not a good choice for also mounting chassis connection eyelet lugs, because the lamination stack may tend to compress over time, leading to a poor chassis connection for the lugs. There is an additional problem with aluminum chassis, as an insulating oxide layer can develop over the surface of the chassis over time. Hence it is best to use dedicated, high tensile fasteners with star washer, spring washer and locking nut for connections to the chassis, especially so for the only safety critical connection in an amp, ie the ground wire to the chassis.

Note that the noisy tone stack caps are typically silver mica types subject to significant VDC. The noise possibly being due to DC leaking through them. The cap on your volume control might be silver mica, but it should have no VDC across it.

[pompeiisneaks]

pdf64 is on the ball as usual, that sounds to me like a tube too. Swap V1 and V2 and see if the noise is a little 'less' prominent, as the first stage is usually the most sensitive to noise/problems. Or just put in a new tube in V1 as suggested.

The second issue is common with guitars plugged in. If the noise is not present with no guitar cable connected, it's coming from the guitar itself, and/or your environment seeping into the guitar.

~Phil

[norburybrook]

I think those Cliff jacks are switching Phil so with nothing plugged in the signal is grounded so there will be no noise as there's no signal. Plugging just a cable in will open the jack socket and you'll hear the amp.

pdf64 is correct about the IEC mains input. I think it's good practice to have a separate ground close to where the IEC plug is and leave an amount of slack in the earth connection just in case the amp fell of a cab and it pulls the socket out, the ground will still hopefully be connected.



M

[pompeiisneaks]

I think those Cliff jacks are switching Phil so with nothing plugged in the signal is grounded so there will be no noise as there's no signal. Plugging just a cable in will open the jack socket and you'll hear the amp.

pdf64 is correct about the IEC mains input. I think it's good practice to have a separate ground close to where the IEC plug is and leave an amount of slack in the earth connection just in case the amp fell of a cab and it pulls the socket out, the ground will still hopefully be connected.



M

Understood, every amp I've ever built had a switched tip to earth design when nothing is connected. That being said, plugging a guitar into it and getting noise means the noise can be coming from the guitar. I.e. when I plug single coils in or have a crap cable, I often get tons of noise, but when I swap to a humbucker there's next to 0 noise. This is the point I'm making. Noise has multiple avenues of ingress into an amp.

I guess an easy way to test this would be to not put a jack into the socket but instead use a chopstick to force the tip of the jack down to separate it from the ground switched connection, and see if the noise is identical, in which case you've ruled out a guitar or bad cable.

~Phil

[ViperDoc]

You could also try isolating your preamp ground buss. It appears that your preamp buss is connected all the way through to the bias section of your board and on the layout as well. After the phase inverter, your power tube grid leak resistors and on through the power section ought to be grounded separately, is my understanding. Also, cliff style jacks are not grounded to the chassis, so make sure your preamp buss wire is chassis-grounded, and only at one point.

EDIT: I noted an improvement on an 18-watt build using the advice above, but after reviewing the original JTM45 schematic, it appears it suggests a continuous grounding scheme. I'm not sure it would be recommended these days, but the powers that be on this forum will know more about that than I would.

[jude]

Thanks everyone for your input and I have taken note of the suggestions.

I have swapped the pre-amp tubes, V1 with V2 and then V1 with V3 to no avail, the crackle on input one remains and become louder as the volume is turned up. I also removed the 100pF and 500pF caps independently and the crackle is still there. Checked all the wiring to V1 and V2 again and can't see anything wrong. So it seems to me it's not any of the 12AX7B tubes, maybe a bad socket at V1? I checked the resistance on the volume pot 1 and it increases steadily to 1M ohms as the volume is turned up.

My electric guitar has single coils and toggling to the out of phase positions does reduce the hum although I still think there's a grounding problem. I inserted a chopstick into each of the channel 2 inputs, the upper input gave a very loud hum when doing this while the lower input did hum but was very soft (low volume). So perhaps we could discuss the intricacies of grounding later?

Any feedback on the crackle problem would be appreciated. Change the socket at V1? thanks

[pdf64]

Try replacing the bright channel input stage’s 100k anode load resistor.

[angelodp]

Reflow the B+ cap wires. I just had this on a build. I had a low hum and vibration in the amp. You could feel the vibration with your hand on the chassis. I though I had inspected all the solder joints, but two of the big cap solderings were insufficient. I reflowed them and problem gone.

[jude]

I did test the 100k resistors in turret for both the bright and normal and both seem ok, close to 100k. I guess the next thing to do is check for continuity on the V1 socket, tube side to turret side.

Thanks for the tip re hum on the big caps, I did check for continuity and seems good. All the non-electrolytic caps seem ok in circuit, no continuity, have resistance - as far as I can tell this way.

[pdf64]

How exactly did you test them?
I’m not aware of a test that can identify whether a resistor or indeed any part is noisy or not, other than to replace the suspected part for a new, hopefully good, one.

[jude]

I tested them with a multimeter, not making any judegement there but just to comment that it seems ok. I realise readings in circuit are not reliable and if you recommend replacing it I will. I guess since the problem has been isolated it should be in that part of the circuit. Thanks I will give it a go.

[pdf64]

When fault finding, there doesn’t tend to be much point in testing something for the sake if it, there should be a rationale to do so, eg in this case you have an expectation of what the result should be.
So when I asked how exactly you tested them, I meant what parameter you were testing for, and how you went about it, eg what range on your multimeter, where did you put the meter probes, what were the circuit conditions?

The point is that multimeters don’t have a range for ‘noise’ per se.
The noise heard will be a uVAC, but that is far below the resolution of almost any multimeter.
In this case, our ears and the volume control reveal that the noise was coming from the bright channel’s input stage, and a valve swap indicated that it probably wasn’t due to one of the triode valves in the first 12AX7 being unsuitable.
So, what parameter of the resistors were you testing, how did you go about it, why did you think it beneficial to test it, what result were you expecting?

I’m asking you this stuff as I suspect you are under a misapprehension regarding the nature of the issue / the fault finding process.