John -
I only wanted to off-load PMA from too much simulation work. Sure, let's see what PMA can do.
If you need my help, let me know, and I will do the simulations.
My email is also available.
Sigurd
I only wanted to off-load PMA from too much simulation work. Sure, let's see what PMA can do.
If you need my help, let me know, and I will do the simulations.
My email is also available.
Sigurd
john curl said:
Bob Cordell said:
Hi Pavel,
Perhaps you missed this question, as I haven't seen a reply.
I've since checked the section in Self you referred to, and it is really only very cursory treatment of complementary VAS achieved with a unipolar input stage. You're right, he is not a fan of the use of a complementary VAS. I am.
I also checked your web site and did not see any elaboration on your opinion of using a unipolar input stage to drive a complementary VAS.
Anyway, if you could elaborate on specifically why you think that approach is inferior to the full complementary dual-differential input/VAS arrangement I would appreciate it.
Cheers,
Bob
PMA said:
Notice that while the overall distortion is higher, the distortion products are monotonic.
Isn't this what Hiraga and others have said is preferable?
So how about the same using Ge transistors?
se
PMA said:
No? Mmmm. I'm pretty sure that's what Hiraga advocated. I'll see if I can find something more specific.
Even products are higher. Odd products are all about the same.
So do you have any Ge SPICE models?
se
He is a pretty good audio designer. He first discovered differences in wire, and wrote the first real article with measurements included back, in the 1970's. I met him in 1978 and he even sold me a little phono preamp, Would you like to buy it, or copy it? You never know, it might be near perfect. It is built like a bird cage in construction.
SE, you can follow his directives if you wish, but PMA and I would prefer to reduce the higher order distortion to virtual unmeasurability, even at the cost of slightly more 2'd or 3'rd harmonic, if necessary.
This follows earlier work from the 1930's-1940's, as described in 'The Radiotron Designers Handbook' that makes more complete sense to me.
SE, you can follow his directives if you wish, but PMA and I would prefer to reduce the higher order distortion to virtual unmeasurability, even at the cost of slightly more 2'd or 3'rd harmonic, if necessary.
This follows earlier work from the 1930's-1940's, as described in 'The Radiotron Designers Handbook' that makes more complete sense to me.
john curl said:
I don't know. The "built like a bird cage in construction" has me at least a little intrigued. Can you manage a photo?
Well in the sims at least, the highest product, the 7th, is the same in both instances so there's no apparent advantage in that regard. So I guess it just remains whether or not it would be preferable to have the products fall off monotonically with even order dominance or not with odd order dominance.
Wasn't there ultimately a monotonic weighting applied? That the third should be lower than the second, the fourth lower than the third, the fifth lower than the fourth, etc.?
se
Steve Eddy said:
Steve, I am not interested in doing Ge simulations. My work - my time, not yours.
Also, I have to apologize to those I do not answer, my free time is really not unlimited to be spent all in high-end chat.
Unlike me, PMA. 
I think the example would be more clear if and when you can go back to fets as an input. There, the differences should be more pronounced, expecially with high drive.
I agree, Ge, would be interesting, but you have to get good models first. Let SE do it, he is the resident expert around here, certainly he can SPICE something together.
I think the example would be more clear if and when you can go back to fets as an input. There, the differences should be more pronounced, expecially with high drive.
I agree, Ge, would be interesting, but you have to get good models first. Let SE do it, he is the resident expert around here, certainly he can SPICE something together.
PMA said:
Fine. I never claimed it was mine. No need to get so defensive.
I just thought it would be interesting as some are of the opinion that germanium transistors sound better than silicon transistors.
se
john curl said:
Well, as you say, you'd first have to get good models.
Better perhaps just to build.
If you can recommend a complementary pair of germanium transistors that would be a good candidate for this, I'll see if I can't hunt them down. If SY's up to it, I could build it and send it on over to him for distortion analysis.
se
Lots of OT posts deleted. If someone wants to start a separate thread on wire and diodic effects, feel free.
SY said:
FASCIST PIG! 😀
So if I build up a double complementary germanium input stage, would you care to run a distortion analysis on it?
se
- Status
- Not open for further replies.