Well firstly DeForest was one of his first names, and although named in honor of Lee de Forest there is no relationship.
Secondly 'Spook'. Come on Nigel...
That's called in a rush to do what I should be doing. Dyslexia seems common in engineering. If you look up the good doctor he is named after him and some say related.
https://en.wikipedia.org/wiki/DeForest_Kelley
Sid's hi fi was Model 9 in triode ( g2 to anode ). Floating LM317 at circa 450V regulator ( he was very insistant to let that be known ). Thorens TD160 with Grado PU ( Joe was his friend ). And Song Discman as battery PSU was a big deal to Sid. Speakers were Quad ESL 63 with no covers. In terms of total system distortion it is the lowest I had ever heard by a very long way. I remember Sid saying as he had no idea on my views that he had recently been asked to change capacitors in a Model 9 and to his total surprise could hear a difference. The best bit was " Nigel, you have to believe this ". The most stupid thing I never asked Sid was if he knew Albert Einstein. They were more or less neighbours if I'm right. It seems totally unlikely he didn't. Sid told me he wasn't that gifted except he never gave up. History knows that to be the real gift. He also said between he and Dick Sequerra you got one decent engineer.
Something else. When Marantz made some generic Model (?) Ken Isiwata never bothered to contact Sid. Sid said he would have been glad to help and wouldn't have wanted money. With Sid it would be very easy to mistake him for a hobbiest as he was so modest about anything he talkled about. As I was taught " You never know who you are talking to ". That has been part of my learning. My three next door neighbours knew very well, Margret Thatcher, Stephen King ( I was said to be as bad as him except I sure knew my Bible, the neighbour was a Bible basher and I retain factoids ). More importantly Roy Rogers plus 4 legged friends. That was because I asked.
Sid to me. " Nigel, I hope one day my transistor designs become respected. I suspect all they need is capacitor upgrades. Problem is we were taught it couldn't matter and I saw no reason to question that". He lookd very sad to think he wasn't noted for them. You could see it in his eyes. Please note the " I saw no reason to question that ". How many here like the fact that no one asks them to question things ? Know things? Yes. Question things ? I think our Royal Society says " Take no one's word for it ".
Do video op amps work in all active phono stages? Nothing seems to say they wouldn't ( noise perhaps and you need to read the graphs more than usual ). The low output voltages assume a second stage. 75 uS with gain followed by 3180/318 with lets say NE5534 or better. Some work at 32 V so could be the second stage. All inverting if MC ( twice inverting ideally ). My Lyra PU will cope nicely when 22R input loading. It seems just as happy with 47K! It's designer is a DIY member.
Again late to do some plumbling for a friend who's moving house so do the correctons please.
Sid to me. " Nigel, I hope one day my transistor designs become respected. I suspect all they need is capacitor upgrades. Problem is we were taught it couldn't matter and I saw no reason to question that". He lookd very sad to think he wasn't noted for them. You could see it in his eyes. Please note the " I saw no reason to question that ". How many here like the fact that no one asks them to question things ? Know things? Yes. Question things ? I think our Royal Society says " Take no one's word for it ".
Do video op amps work in all active phono stages? Nothing seems to say they wouldn't ( noise perhaps and you need to read the graphs more than usual ). The low output voltages assume a second stage. 75 uS with gain followed by 3180/318 with lets say NE5534 or better. Some work at 32 V so could be the second stage. All inverting if MC ( twice inverting ideally ). My Lyra PU will cope nicely when 22R input loading. It seems just as happy with 47K! It's designer is a DIY member.
Again late to do some plumbling for a friend who's moving house so do the correctons please.
You can easily buffer the opamp (also make the whole thing class A if you want) and end up with superb noise performance and >30 dB overload. If you want >80 dB with a cart connected, use some JFET's in front of it. See Syn08's website on how to do it.
We are just having a discussion about phono stages over on the Source forum. While I don't decry your many highly rated designs and awards in this area, you can't go making blanket statements like the one above re opamp designs. They are simply not true. Splitting an EQ stage for any reason when using opamps is suboptimal - noise, overload and distortion. For tubes - agree - you have the voltage swing, but loading effects mean you have to use high EQ values (transformer impedance matching may be a way out here, but that's added complexity and cost). And, I have yet to read about or see a test report where a tube phono stage bests the best solid state can offer.
Separately, EAR had a phono stage where Tim dP proudly spoke about it having '42 tubes'. The mind boggles.
Unfortunately there's equally no guarantee that higher distortion, higher noise and lower overload margins present better subjective impressions. The devil is in the detail in these matters.
However, if we are talking 'bout that 0.00003% THD opamp, I think for RIAA its bested by the one that does about 0.001%. The noise current on the former is too high.
More L8r ;-)
However, if we are talking 'bout that 0.00003% THD opamp, I think for RIAA its bested by the one that does about 0.001%. The noise current on the former is too high.
More L8r ;-)
I got a fix for that. Whenever I want to listen to music, I connect the scope to the amp and spend my time watching the screen. In fact, lately I don't even connect up the speakers anymore - too much bloody noise!
Jan
One idea I rather like although had no sucess with was a voltage rail boosted op amp. My feeling is that overload margins for phono stages is all imagined. A little thought about how a pick up works and it's damping would suggest it's like a Spitfire going supersonic ( The Soviet Bear TU 95 is a study in that as it exceeds what was thought possible ). I have tested this to crazy limits and found at worse the distortions of a good valve record player.
The conjecture is if we took a fantastic op amp that would allow boosted voltage rails I know in my heart the rails would never be used, thus do no harm. For people who insist passive EQ and op amps won't work it might keep them happy to look at it. I possibly would use 75 uS passive with gain ( 5 to 60 to allow MM/MC ) followed by active 3180/318 gain 20 at 1 kHz. Then a 2uS pasive output filter. I don't suggest the circuit in the link is used. It is an example of how simple it could be. I will post this in the other forum.
http://joebrown.org.uk/images/DualPSU/BootstrappingOpAmps.pdf
The conjecture is if we took a fantastic op amp that would allow boosted voltage rails I know in my heart the rails would never be used, thus do no harm. For people who insist passive EQ and op amps won't work it might keep them happy to look at it. I possibly would use 75 uS passive with gain ( 5 to 60 to allow MM/MC ) followed by active 3180/318 gain 20 at 1 kHz. Then a 2uS pasive output filter. I don't suggest the circuit in the link is used. It is an example of how simple it could be. I will post this in the other forum.
http://joebrown.org.uk/images/DualPSU/BootstrappingOpAmps.pdf
Last edited:
Boosted rails might partially nullify any voltage-variable (hence: nonlinear) capacitance on the opamp input pins.
Or just be a macho guy and build a discrete amplifier running on ± 35 volt rails. JFET input, Class A OPS @ 30mA standing current, 100V/usec slew rate, the works. You could even use the Sansui cascode bootstrap (JFET-JFET) to nullify any voltage-variable capacitance on the input transistors.
Or just be a macho guy and build a discrete amplifier running on ± 35 volt rails. JFET input, Class A OPS @ 30mA standing current, 100V/usec slew rate, the works. You could even use the Sansui cascode bootstrap (JFET-JFET) to nullify any voltage-variable capacitance on the input transistors.
Reality check:
Plenty OPA's deliver 1ppm on bog standard +-15V Rails. Clearly no problems with non linear capacitance.
More to the point, if low distortion does not correlate with subjective impression, why are you even concerned about this?
Higher voltages on discrete designs mean you can approach IC type distortion performance with less circuit complexity - which IMV is not a bad reason to use that approach.
Plenty OPA's deliver 1ppm on bog standard +-15V Rails. Clearly no problems with non linear capacitance.
More to the point, if low distortion does not correlate with subjective impression, why are you even concerned about this?
Higher voltages on discrete designs mean you can approach IC type distortion performance with less circuit complexity - which IMV is not a bad reason to use that approach.
Last edited:
the question is, will you, John ever acknowledge anything done in recent decades in monolithic op amps
Groner makes specific in-put Z linearity measurements of recent op amps
johnc124 has even helped with Groner's work:
http://www.diyaudio.com/forums/anal...pedance-jfet-input-op-amps-4.html#post4129482
TI has new OPA164x with extremely linear input C:
Groner makes specific in-put Z linearity measurements of recent op amps
johnc124 has even helped with Groner's work:
http://www.diyaudio.com/forums/anal...pedance-jfet-input-op-amps-4.html#post4129482
TI has new OPA164x with extremely linear input C:
Attachments
Yes. I've played around with high V supply rail circuits. I see jcx is on the case anyway. You can always bootstrap the rails as well.
But to my earlier point, why bother if you feel ultra low distortion is worse sonically?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.