KOT weird volumepot behaviour

[PMowdes]

Hi all

I've made a pcb based on Matsumin's KOT schematic and i've got a strange issue which i didn't have on my vero version.

Whilst it works and the channels switch it seems as though both volume pots are working in parallel, so that they are interactive and i only get full volume when both pots are cranked up.

I've tried to decipher the original post on Matsumin's website but the google translate doesn't do a great job

Do you think that there is some reverse flow going on???  can i solve this with some polar caps as condensers following the volume pots??

here's the schematic for reference



Thanks in advance

Phil

[BuGG]

Yep, when Vol-A is active the signal is flowing back into the wiper of Vol-B and to "ground" (Bias)...

What you need to do is use the 3PDT to switch Lug 2 of the volume pots instead of Lug 3.  You should be able to patch up your PCB to achieve this by omitting one of the 1uF caps and adding a couple jumper wires.    

If you need any help shoot me a message or post a pic of your layout.

[PMowdes]

Thanks Bugg

But the more i look at it the more i'm confused.  I've been dissecting the vero layout and it appears to work exactly the same way as the matsumin schematic.  

The switch is before the pots and switches lug 3, lug two goes to the output via a 1u non-polar cap and lug 1 goes to bias

I've built the vero and it works fine?

Audio probe time i think.

[Ciaran Haslett]

Are your Vol 1 lugs definitely going to vref and not the more common ground?

[PMowdes]

Ciaran Haslett wrote

Are your Vol 1 lugs definitely going to vref and not the more common ground?

Yeah i'm pretty sure they do.  I found a missing junction with the capacitor running from bias to gnd in the power section.

Trouble is i've been tinkering and i can't remember if it was there in the board version or not.

I'll have another look tomorrow with fresh eyes.

[Travis]

I think Bugg has it right. In the original schematic, one pot will be like a variable filter while the other is in use right? If you do it as Bugg shows, the wiper of the switched out pot won't be able to influence the in-use pot, it will appear as a 100K fixed resistor to vref