This is not just another gainclone

[fdegrove]

Hi,

Peter,

Apparently this cable is thought to be cause of trouble by a few people:

GOERTZ.

Note also that highly capacitive LS cable can render quite a few solid state amps unstable, regardless of brandname of both cable and amp.

BTW, just out of curiosity, how much of the sonic results would you attribute to the choice of passive components, percentagewise?

Cheers,

[Peter Daniel]

Originally posted by fdegrove

BTW, just out of curiosity, how much of the sonic results would you attribute to the choice of passive components, percentagewise?

Between 30-40% . This amps can be made to sound terribly (although it's not easy to make) and they may also sound very impressive. You figure out how much percantage it is

We've been optimising the final configration through many experiments and we tried many different parts. It seems like some of them work much better than others. What Brian is selling in a kit now, is exactly the choice we figured works the best for this amps. It took us almost a year to reach this stage.

I'm still making it p2p in Amp-1, but future models will use PCBs. They are just too convenient to be ignored and if made properly, I don't see deterioration in performance. It took Brian and his idea of a kit, to change my mind about PCBs. I kinda like them now

[Peter Daniel]

As to Goertz, I'm still using copper foil in one of my systems, but I run two wires separately, so capacitance is not an issue. It seems to work well this way.

[fdegrove]

Hi,

It took Brian and his idea of a kit, to change my mind about PCBs. I kinda like them now

No need to tell you where mine would end up I reckon...

Thanx, P.

[A3rd.Zero]

It has been suggested that the best value for a gainclone pot is 25k. Would it be ok to use a 100K pot with a 33k shunt resistor will that be alright? Thanks

milo

[Peter Daniel]

25k for NI GC is good as suggested input impedance for that circuit is approx 22k. I've been using 50k pot with 22k shunt resistor across and it was OK too. I wouldn't really recommend 100k, but if you don't have other choice it will work as well. Just make shure that it's shunted and max input imp that the chip "see" between +IN and ground is approx 22k or so.

[A3rd.Zero]

Ok, so lets say that I use a 100K pot with an ~18k shunt resistor. That gives me a top resistance of ~15.25k where your 50k 22k combo is about 15.27k. Will it be particularly twitchy or fine in its adjustment because I am using a 100K pot?

Thanks

Milo

[Peter Daniel]

No, it's not twitchy at all. You may even go for higher combined resistance, the only inconvenience is increasing DC offset at the output. So if you'll get too much of it, just reduce input impedance. That's all.

For instance, with 50K pot without shunt resistor, the offset was reaching as high as 350mV. Adding shunt resistor, decreases that value to more reasonable, 60mV, or so.

[A3rd.Zero]

Ok great thankyou.

<sheepish silly question that should have been asked before all this>

A shunt resistor is a resistor in parallel with the pot right?

</sheepish silly question that should have been asked before all this>

Thanks again

Milo

[Peter Daniel]

The shunt resistor is the one from +IN (on a chip) to ground. If you mount this resistor between input pin on the pot at ground, the dynamics will suffer. In other words, the resistor has to be mounted "after" the pot.