Ghost Sound LondonPower StandardPreamp

[Strelok]

Hi all!

We all know the song "Ghost Town" by "The Specials".
I have a problem with a ghost sound, but it is not coming from that town.

I built a Standard Preamp a long time ago, using circuit card and the
layout from the kit.
The gain channel has an extra sound that is a shrill and thin version of the gain channel.
It does not listen to any tone controls, neither on the clean nor on the
gain channel.

When I turn up the volume of the gain channel the sound remains the
same, same volume, but the normal sound comes up like it is supposed to.
So of course you cannot hear it anymore but it is still there.
I can tell because annoying buzzing is much worse in that preamp, so
those frequences get amplified more.
I have to turn op the volume (Pump Up the Volume) way too high to swamp
out the ghost.

A while ago I built another Standard Preamp and it has the same thing.

I don't have this issue with single channel preamps I have built, like
the Vintage and High Gain variations and the single channel "Y".
They are also on circuit card.

What can this be?

I remember somehting about electro-magnetic coupling, but I'm not sure
where I read that.
That would occur near tone controls or channel switches.

Or maybe I just dreamed that.


Kind regards,

Strelok

[K O'Connor]

Hi Guys

The predecessor of the LP-PRE with a PCB was GPK Guitar Preamp Kit with an eyelet board layout. is that what you built, Strelok? If so, the decoupling could be improved and that effects the overall layout.

Ghosting is often attributed to bad filter caps and that might be the case here. It is also related to layout with respect to supply+ground feeds and whether gain stages share a filter cap. Ghosting can also be via induction or electrostatic coupling via proximity of signal paths - so another layout related problem.

Where a DPDT switch or relay can be used for channel switching, this brings in-phase signal points close to each other with the risk of oscillation. It can also allow signal bleed-through, a common problem with high-value value controls wired with 2-conductor shielded cable. The fix for these kind of problems is to use separated muting elements placed where the signal already is so the signal does not have to be moved elsewhere. Another fix is "double muting", particularly effective when jfets are the mute elements.

So,the ghosting cure will likely come down to layout and decoupling.

[Strelok]

Hi Kevin,

Well thank you very much!

Yes, I built that one, with the kit you mentioned, with no local decoupling.
The alternate circuits work fine, but they are single channel.

Soem yeras ago I built a new version that has local decoupling but it does not make a difference.

I remember you saying "it's all I use" about shielded cable with not one but two conductors.
If I recall correctly

Sometimes you can't get passed around that, but I used seperate cables.

I'll remove the clean channel completely, just to see what happens.

A while ago I tested with a separate tone stack box I built.
If I remember this right the problem went away then.
But only if I used the volume control that was in the box, when I used the vol on the amp the problem remained.

Strange I have this on both preamps.

What jfets do you recommend? I have trouble finding the right ones, in all my collected documents only obsolete ones are mentioned
Or very expensive ones.

Strelok

[makinrose]

Strelok,
I think KOC is right in that your problem likely lies with layout and/or decoupling. A few photos of the build might help illuminate what your problem is.

[Strelok]

Hi guys!



Thank you so much for this!



Here are the pictures.


















V1 has one wire going to the PS terminal point, V2 has too.



I could try to add decoupling.





Have a fantastic weekend all!!!



Strelok.

[nauta]

Hey man
What kind of construction is that?
I don't see any filter caps so isn't that the problem?
Happy new year

[Strelok]

Hi Nauta!

Yes that would be magical, without a Power Suppy!
It uses the 48VA supply as designed bij KOC.
Happy New Year

[K O'Connor]

Hi nauta

The construction is using eyelets on a rigid fibreglass epoxy card, similar to Fener's old methods from the 1960s to 1980s except Fender used flexible card material.

Yes, there are no specific filter caps on the preamp card as it is built and as it was designed - definitely a flaw! - and that is the reason for the ghost notes. Strelok's reference to the power supply states the obvious that there are filters "somewhere", just they are not physically on the preamp board - adding some will fix the problem. (unless he uses Illinois Capacitor electros )

[makinrose]

Are the input jacks isolated from the the chassis? If they are not they could be causing a ground loop....

[Strelok]

Hi all!

Yes, the jacks arte insulated.

I think it has to do with two cables going parallel a long way.
There is antother preamp with the same problem but that one has proper local decoupling. It also has cables running parallel.
Not good, Kevin mentioned that in this thread.

On the original layout Kevin uses 3-conductor cable, quite a difference.
I will try that.

There are more preamps without local decoupling that run fine, one even from the same Power Supply.

So, who knows that is it!

Thank you so much all!