Any interest in GC input buffer and power supply regulator pcb's?

[Sheldon]

Didn't fry it after all. Just another wrong connection. All sorted now. I had the gate and source connections reversed, and one other wrong connection. I'm getting 5-6 ma current through R1 and R2 (measuring .85v across R1 and about .77v across R2). Is that ok, or should I drop the resistor values a bit? Otherwise good. I get +24.2 volts and -24.4. Now on to the buffer.

Thanks, by the way, for the very detailed notes and clear schematics. I'm at the very beginning stage of relearning some electronics. The theory takes a bit of work, but the hardest part is getting the practical tidbits. Yours are great.

Sheldon

[Pedja]

Tnanks.

You may once try to push Zeners harder but 5-6mA is yet quite good and you won't get wrong if you leave theam as they are for now.

[Sheldon]

Pedja,

Or anyone. If you are using the jfet buffer with a 1k resistor in the filter, couldn't you just short the buffer output after the resistor to ground during warm up? Or what value would be required to protect the buffer. Even if a relay were employed, it would be out of the circuit during run, so shouldn't hurt sonics?

Sheldon

[Pedja]

Yes, it can be done that way.

[SRMcGee]

Forgive the novice's ignorance, but I am about to purchase a bunch of PCBs from BrianGT (LM3875TF) to construct some integrated Gainclone amps. Are these boards compatible with the ones that Brian is selling, and will these boards provide an active buffer stage suitable for use with the passive preamp (probably a DACT CT2-100K attenuator and Greyhill selector) I'll be installing along with Brian's PCBs in the chasses?

[Sheldon]

Originally posted by SRMcGee
Forgive the novice's ignorance, but I am about to purchase a bunch of PCBs from BrianGT (LM3875TF) to construct some integrated Gainclone amps. Are these boards compatible with the ones that Brian is selling, and will these boards provide an active buffer stage suitable for use with the passive preamp (probably a DACT CT2-100K attenuator and Greyhill selector) I'll be installing along with Brian's PCBs in the chasses?

I think that the buffer should be able to handle any reasonable source, as that's it's function - provide a stable high input impedence for the source and a good low output impedence. But I have a very related and equally ignorant question. I'm also using Brians board and resistor values. I've built the buffer circuits and am now realizing major gaps in my knowledge of filters. Pedja, Brians board has a non-inverted input. It has a 220 ohm resistor in series with the input, followed by a 20k resistor to ground (I assume that the chip has a high input inpedence, so the impedence seen by the buffer would be 20k). How would you adjust your cap values (or do you?) to get the desired hf roll-off? Thanks for your patience with us beginners.
Sheldon

[Pedja]

Sheldon,

With 22k the slope will go lower and roll-off at 20kHz will be about 0.2dB higher.

But you can always use 10k there instead, no?

If you stay at 22k you can use 910R instead of 1k to have about the same slope as with 1k-9k(or 10k) combination.

But again, I can not comment the filter’s consequences on the non-inverted GC, have not tried and a bit of experimenting won’t be bad anyway.

Pedja

[Sheldon]

Originally posted by Pedja
Sheldon,

With 22k the slope will go lower and roll-off at 20kHz will be about 0.2dB higher.

But you can always use 10k there instead, no?

If you stay at 22k you can use 910R instead of 1k to have about the same slope as with 1k-9k(or 10k) combination.

But again, I can not comment the filter’s consequences on the non-inverted GC, have not tried and a bit of experimenting won’t be bad anyway.

Pedja

Thanks Pedja,

I guess the 20k shunt resistor value was originally selected to work with a pot directly tied to the front end of the amp, and acts as a voltage divider with the pot to set input sensitivity and impedence, which changes with pot position? With the buffer, it makes sense that I can change that value to anyting that the buffer itself can drive. I think I''m gradually getting it. I'll see what resistors I've got on hand. Almost there. I'd like to fire this thing up today.

Sheldon

[Sheldon]

Originally posted by Sheldon
Pedja,

Or anyone. If you are using the jfet buffer with a 1k resistor in the filter, couldn't you just short the buffer output after the resistor to ground during warm up? Or what value would be required to protect the buffer. Even if a relay were employed, it would be out of the circuit during run, so shouldn't hurt sonics?

Sheldon

Originally posted by Pedja
Yes, it can be done that way.

Ok, now I'm going to prove that a little knowledge can be dangerous. I've got some extra j309's. How about drain to buffer output, source to ground and gate to a cap and resistor to ground, powered by =15 volts from the buffer reg.? If I understand correctly (a very dubious assumption), output from the buffer should flow to ground until voltage on the gate comes up enough to shut it off. Time would depend on the cap and resistor values. Any chance?

On the grounds- I finished the buffers. I left the 20k shunt resistor and used two 15nf polystyrene caps in parallel for C2 and a 4.7nf polystyrene cap for C1, separated by a 914R. I have a ground wire that connects the ground terminals of the buffer regs. I have attached the caps to that ground and had planned to connect that, in turn, to the signal ground on the chip (Brian GT's non-inverted clone). I was also going to connect the buffer input ground to the same wire. Bad idea? Should I use a separate ground for the regs and signal paths? Or are the reg. grounds low enough current to not interfer with signal?

Great thread, by the way. Lots of good info from everyone. Hope I'm not taking it too off track.

Sheldon

[troystg]

Hi All-

I know I am opening a can of worms here but I have to ask the question. Please excuse me if it has been asked before.

I have been using the Marchand XM-1 crossovers for years with no issues. Since they have a unity gain from in to out, couldn't they also be used as a buffer for the gainclones?

Of course they would also introduce a filter but if you were using a sat/sub system that would be a good thing.

So would the XM-1 make for a decent buffer or is the circuit design too dated?

Thanks, Troy