Just wondering if it could me made simpler and less costly by removing a few FET's. Also wonder how it would sound with resistors in place of FET's?
Pretty inexpensive experiment to try it yourself and find out. The original was painstakingly optimized for low cost, see the BAF video where Nelson quotes Steve Martin "And it only cost ten buxxx". So as ZM recommends, build two the NP way and then, build two more your own "simpler and less costly" way. Listen and critique.
I have a feeling that if you use balanced XLR inputs, which NP bent over backwards to provide for, you'll be able to IMMEDIATELY hear the difference between the two approaches. One has common mode rejection, the other has caprine scats. If you speak yiddish, bupkis.
I have a feeling that if you use balanced XLR inputs, which NP bent over backwards to provide for, you'll be able to IMMEDIATELY hear the difference between the two approaches. One has common mode rejection, the other has caprine scats. If you speak yiddish, bupkis.
Trying it with resistors in place of the CCS is worth doing.
When the J113 are available from Mouser for $0.40 at qty (1), with just under 48,000 in stock, there isn’t a compelling cost savings. These aren’t $8.00/ea K170s.
When the J113 are available from Mouser for $0.40 at qty (1), with just under 48,000 in stock, there isn’t a compelling cost savings. These aren’t $8.00/ea K170s.
Thanks 6L6, that is a good price for the FET's. Some people say the simpler the design the better and I remember that one of Nelson's first DIY designs was just one FET!
Trying it with resistors in place of the CCS is worth doing.
When the J113 are available from Mouser for $0.40 at qty (1), with just under 48,000 in stock, there isn’t a compelling cost savings. These aren’t $8.00/ea K170s.
CCS improves the PSRR of this circuit significantly.
The problem with J113 is the huge variation of Idss between devices and between batches.
This makes designing already difficult if you want to achieve a certain performance.
And definitely a lot of sorting, which Nelson has done for you in this case.
Cheers,
Patrick
My experience over the years is that you don't know what Idss you get if you order another batch 6 months later.
But maybe they have their process under better control now, maybe not. Nelson will know better.
The advantage with Toshiba (e.g. 2SK209) is that they are pre-sorted into specific Idss ranges.
So it is easier to cope with in design. I know for certain what to expect.
Patrick
But maybe they have their process under better control now, maybe not. Nelson will know better.
The advantage with Toshiba (e.g. 2SK209) is that they are pre-sorted into specific Idss ranges.
So it is easier to cope with in design. I know for certain what to expect.
Patrick
2SK209 will be OK for the diff pair. And they will perform better there.
They won't stand the dissipation in the current source.
But J113 is OK for those positions.
We have a custom adaptor board which turns SOT23-3 into TO92.
Size is 5x5mm.
Costs extra even if you can find a shop willing to make them.
Patrick
They won't stand the dissipation in the current source.
But J113 is OK for those positions.
We have a custom adaptor board which turns SOT23-3 into TO92.
Size is 5x5mm.
Costs extra even if you can find a shop willing to make them.
Patrick
Thanks for that picture. It is now hanging on the wall of my office.
Operating the VAS in class A has its benefits👍The schematic seems to be revised; compare to the version shown in the BAF-2022 video. The biggest surprise, to me anyway, is the disappearance of PNP's emitter resistor. In Nelson's article "DIY Opamps" on his PassDIY site (link), his opamp circuits that used a PNP for the second gain stage, did have a 10 ohm emitter resistor like the one shown in the BAF-2022 video. But now, poof, gone.
_
Hello 6L6,
referring to your post #151, I made a little pcb (DIY Front End 2022) for me with both possibilities - all parts in SMD and / or in THT.
Only exception are the three big caps (C1,C2,C3). So I can choose if I prefer SMD or THT or what is available,,,
I'll see how it works. Will be delivered next few weeks...
Don't ask: I will not offer these boards or the gerbers. I respect the efforts of Mr.Pass and the shop with his helpers!
Cheers
Dirk 🙂
referring to your post #151, I made a little pcb (DIY Front End 2022) for me with both possibilities - all parts in SMD and / or in THT.
Only exception are the three big caps (C1,C2,C3). So I can choose if I prefer SMD or THT or what is available,,,
I'll see how it works. Will be delivered next few weeks...
Don't ask: I will not offer these boards or the gerbers. I respect the efforts of Mr.Pass and the shop with his helpers!
Cheers
Dirk 🙂
Attachments
Hello Patrick,
you are absolutely right. If using SMD-JFets, then the railvoltages have to go down. Or another pcb with cascoded SMD-JFets.
Have a nice evening
Dirk 😉
you are absolutely right. If using SMD-JFets, then the railvoltages have to go down. Or another pcb with cascoded SMD-JFets.
Have a nice evening
Dirk 😉
- Home
- Amplifiers
- Pass Labs
- DIY Front End 2022