Sessionette - highly sensitive to EMI

[dai h.]

(sorry, just a non-expert guess) loose (something changed so possibly a nut loosened over time) RV 10 (A250k) pot body becoming a significant impedance over which RF is C coupling into the feedback loop?

[Mick Bailey]

The pots ground fine. With IC6 removed and pin 1 grounded the noise is still there, so the problem is after this. I've had to park it to get some other repairs done, but it still doesn't make sense to me.

[Helmholtz]

Is C44 good (low ESR)? Wondering why only grounding its negative end stops the noise.

[nickb]

Is C44 good (low ESR)? Wondering why only grounding its negative end stops the noise.

Mick removed and bypassed that cap in the first post so it can't be that.

I see the focus in on the signal path but I wonder if the noise is getting in the ground side of things. If that is the case then I would expect the noise to come and go dependent on where you ground the negative end of C44 to. That process might lead you to a point in the ground path that had gone high impedance, at least at high frequencies.

[Helmholtz]

Mick removed and bypassed that cap in the first post so it can't be that.

Sorry, missed that info. Still I would try a tantalum cap in this position to see if it makes a difference.

I see the focus in on the signal path but I wonder if the noise is getting in the ground side of things. If that is the case then I would expect the noise to come and go dependent on where you ground the negative end of C44 to. That process might lead you to a point in the ground path that had gone high impedance, at least at high frequencies.

Good point!

[dai h.]

The pots ground fine. With IC6 removed and pin 1 grounded the noise is still there, so the problem is after this. I've had to park it to get some other repairs done, but it still doesn't make sense to me.

Ah well. Been reading about RFI/EMI issues but only understand it in general terms (avenues of ingress, countering with filtering, shielding, device/topology susceptibility, etc.) without much real life experience solving problems.

Mick removed and bypassed that cap in the first post so it can't be that.

I see the focus in on the signal path but I wonder if the noise is getting in the ground side of things. If that is the case then I would expect the noise to come and go dependent on where you ground the negative end of C44 to. That process might lead you to a point in the ground path that had gone high impedance, at least at high frequencies.

Was thinking a bit of that (some AC/DC ground connection went open or iffy). Also, I gather it has insulated jacks and no high frequency AC shunts (capacitors) immediately at IN/OUT points to chassis (I understand this could mean higher susceptibility via the (significant at RF) ground impedance of cabling).

(sorry for the interruption, carry on...)

[Mick Bailey]

Further investigation shows noise is also getting picked up by the screen of the reverb cables. With my original phones used for testing I though this was a problem fixed and there was insufficient screening, but not so with the new test phone. I think nickb has identified something I'd overlooked.

This reminds me very much of noise on a digital ignition system I designed where I thought noise was propogated by a switching IGBT. It took me months to identify the noise was the ground reference being pulled up - something called 'ground bounce'. Because my scope was always referenced to ground I was always looking at what the probe was seeing relative to ground. Time for a different approach.

Changing the cap for low ESR made no difference, neither did a tantalum.