john curl said:
I did already, low frequency distortion at GROSS SPL's like 150-160dB not relevant to nature recording. You still show no evidence at understanding that circuit I posted. Now I will leave this conversation.
Scott, you can leave the conversation, but does your technique nullify the problems of the nonlinearity of the microphone element in any way? And does it bypass the passive cap loading problem, because it is an active cap? Maybe a negative Z active cap? And if so, what effective value? Perhaps you are on to something, but do you "Throw the baby out with the bathwater"? because you 'may' pull more distortion from the mike element with your circuit. And finally: Why didn't Erling Fredreksen do this, as he knows enough to have tried it?
Bird calls are not what I care about. I care about loud instruments like the human voice and trumpets.
Bird calls are not what I care about. I care about loud instruments like the human voice and trumpets.
PMA said:
Very interesting OP stage and compensation. The specs look good,
have you had a chance to get some?
BTW did you try the new National OPA's like LM4562?
cheers
Terry
Terry Demol said:
OPA211 and LM4562 together in action:
http://www.synaesthesia.ca/Auxiliary-circuitry.html
Scroll down the page. Its perfect from every objective perspective I was able to look.
Of course, they are opamps, so they simply can't sound good enough for the golden ear brigade 😀
syn08 said:
Thanks Syn08,
Any reason for preference of 4562 over 211 in front end, or was that
purely due to fact that it is dual?
I think you have a slight error on that page, WRT 1632, did you
mean 'lack of even harmonic cancelling' instead of 'odd'?
cheers
T
I appreciate some of the input on op amp suggestions. The 'footprint' is wrong on several that have been recommended, but they look OK, and even very good.
I would like to point out that it is not very much of a challenge to USE an op amp. Unfortunately, op amps are limited in what they are made of, their physical size, the idle current, and perhaps other factors. This does not make them not OK for audio, but it is usually found that discrete designs sound better.
This can be for a number of reasons: Topology restrictions, lack of complementary jfets, class B output stages, thermal non-linearity, voltage restrictions, etc, etc. That does NOT mean that they are not cheap, convenient, and compact.
I might remind several here, who like to take cheap shots, that I have an IC design myself on the test bench. I also have an all discrete design in process. Think of it like a Volkswagon and a Porsche. There WILL be performance differences, but nothing that will get you to the store faster, or make you more comfortable on a long trip. In fact, the Porsche might be a big pain in heavy traffic, eat gas like it was almost free, and really tough hauling more than 2 people.
However, if someone got on a website and said:"I love my Volkswagon. It is all the car that I or anyone else will ever need." You, most probably, would laugh at them.
It is the same between good discrete and IC's in analog audio.
I would like to point out that it is not very much of a challenge to USE an op amp. Unfortunately, op amps are limited in what they are made of, their physical size, the idle current, and perhaps other factors. This does not make them not OK for audio, but it is usually found that discrete designs sound better.
This can be for a number of reasons: Topology restrictions, lack of complementary jfets, class B output stages, thermal non-linearity, voltage restrictions, etc, etc. That does NOT mean that they are not cheap, convenient, and compact.
I might remind several here, who like to take cheap shots, that I have an IC design myself on the test bench. I also have an all discrete design in process. Think of it like a Volkswagon and a Porsche. There WILL be performance differences, but nothing that will get you to the store faster, or make you more comfortable on a long trip. In fact, the Porsche might be a big pain in heavy traffic, eat gas like it was almost free, and really tough hauling more than 2 people.
However, if someone got on a website and said:"I love my Volkswagon. It is all the car that I or anyone else will ever need." You, most probably, would laugh at them.
It is the same between good discrete and IC's in analog audio.
Terry Demol said:
Yes, I did. I am attaching my test circuit. OPA211 outperformed LM49710/4562 in noise, 211 has about 10dB less noise. Also, LM noise tends to increase for LF. For me, 211 is much better chip, than LM. The LM is quite sensitive to demodulation/detection, and is worse in PSR EMI. For me, not the chip to use.
In case that John does not appreciate opamp discussion, for obvious reasons, we can continue in
http://www.diyaudio.com/forums/showthread.php?s=&threadid=122129
Attachments
Sorry to have brought up OPamp IC:s. I just felt that there are
alternatives to the AD797 for first stage in RIAA amps.
Some ICs are not that bench-friendly as you say, John, so I make a proto PCB for becnh work, too,
or make an adapter to DIL8.
The AD745 I mentioned, has the SOIC16 SMD package, and me and a buddy made an adapter PCB for it so it can be plugged in into any DIL8 space.
To use and OPamp is of course easy,
but to make them
"sing"
is another ball game,
and is as hard as to make a discrete circuit "sing". Maybe harder....
The analogy between VW and Porsche makes me get the message, but there are also cars like Volvo:s and SAAB:s that are in between these two extremes.
I do like German cars.....have one in the garage.
But let's get back to the Porsche level of audio.
Sigurd
alternatives to the AD797 for first stage in RIAA amps.
Some ICs are not that bench-friendly as you say, John, so I make a proto PCB for becnh work, too,
or make an adapter to DIL8.
The AD745 I mentioned, has the SOIC16 SMD package, and me and a buddy made an adapter PCB for it so it can be plugged in into any DIL8 space.
To use and OPamp is of course easy,
but to make them
"sing"
is another ball game,
and is as hard as to make a discrete circuit "sing". Maybe harder....
The analogy between VW and Porsche makes me get the message, but there are also cars like Volvo:s and SAAB:s that are in between these two extremes.
I do like German cars.....have one in the garage.
But let's get back to the Porsche level of audio.
Sigurd
Terry Demol said:
LM4562 is dual and comes in DIP. I was looking to swap a few opamps and see if they make a difference (they do). Also, LM4562=$4 OPA211=$7, which accounts for $20 savings for that little board. Though, OPA2211 (the dual OPA211) will most likely be my standard dual opamp for audio as soon as it will we available to the masses. Also, OPA827 is a FET input marvel that will be available soon (engineering samples available now from TI).
Thanks for spotting the error, I'll fix it ASAP.
PMA said:
What do you mean ?
Dynagroove was made when ingeneers realized how a pick-up really works with vinyl and the dynagroove is nothing else than
a compensation for wellknown problems with stylus shapes and works best with spheric tips and was a very good solution to deal with the 19 Khz resonance of the cutting heads in those times.
One of the results was the DL 103 with the original spheric tip.
If you listen dynagrooves therewith , this cartridge sound really pretty good for the money .
Other stylus shapes create new problems which dynagroove cannot compensate .
BTW : My TT a Goldmund Reference with the T3F Servoarm and i think its a good TT which make many things audible , altough my ears are no more the same like 20 years ago . Thus my interest on the Blowtorchthread , this preamp must be really F 1 .
syn08 said:
This is not importat, from my point of view. I am trying to find best component/opamp for a specific task, and several $ saved does not make a difference. Especially when saving $ makes worse sounding product.
Groove-T said:
I know what is Dynagroove. I thought you were mentioning the same issue in your post.
PMA said:
I know you don't like the National LM4562 and LME49710; from where I sit, I was unable to find a shred of objective/measurable of evidence against these opamps (for the application I was looking for).
For a MC/MM preamp I would definitely go with OPA211.
I am sorry but I don't think the alleged RF sensitivity (BTW I was unable to measure anything like, using an industrial test jig) makes any difference to audio performance. Its just that OPA211 has, as you mentioned, lower noise and slightly lower distortions.
PMA - I am with you here.
If there would exist an OPamp that would sound as good as the best discrete circuits, I would gladly pay 100 USD for one of these OPamps.
After all, this is the Blowtorch thread, and cost is (almost) irrelevant.
Sigurd
If there would exist an OPamp that would sound as good as the best discrete circuits, I would gladly pay 100 USD for one of these OPamps.
After all, this is the Blowtorch thread, and cost is (almost) irrelevant.
Sigurd
PMA said:
john curl said:
both of these can be effectively addressed with composite/multiloop op amp circuits as advocated by Walt Jung
all op amps under consideration are class A for the currents required to drive the input of a buffer/gain stage that can be discrete class A or cfa op amp
cfa op amps with -100 dB distortion specs at 1 MHz into 25 Ohm loads may be a few dollars more, discrete buffer stages could be < 50 cents in parts at a somewhat worse stage performance but still improve the composite amp performance
Yes, many IC 'problems' and compromises can be improved by addition of extra components. Think about the old fashioned carburetor. It was a fundamentally flawed component that people spent decades modifying, improving, adjusting, etc. We used to swap out carburetors like people here swap IC's.
Yes, it is true. I put a dual carburetor assembly on a 1954 Chevy 6 in 1958. It didn't help much, but taught me a lot. Lived with dual SU carbs on a AH Sprite for 5 years. Added a 2 throat Weber for a single throat (something) on my 68 Saab Sonnett that I drove for 23 years. However, by 1974, fuel injection became available to me, and I got my first. It had problems, could only be adjusted by an expert, but it had subtle advantages in operation.
For me, IC's are much like carburetors. They are fundamentally compromised from the beginning, and all you can do is work around their limitations.
Now why do I say this?
While Scott is trying to reduce IC operating current (and proud of it) and still remain within very good operating specs, I am at the same time INCREASING current of HIS output stage by the only way that I can, by adding extra components. Who is going to have a better audio circuit, Scott or me?
Please remember Scott's microphone circuit. He is using a NEW op amp with even LESS operating current. Is this design philosophy going to succeed, subjectively?
Should we switch to IC's, unless there is VERY GOOD REASON? One of the last vestiges of discrete design is in the microphone electronics. It is very difficult to change it, once designed, and while many examples of commercial microphone electronics are pretty marginal, is it really better to change to an IC ? What real advantage is there? Are there any drawbacks? Is it really a lower noise approach? What about capsule loading and its real needs for low distortion operation? These are still the real questions that have not been answered.
I regret that my 'tone' has been arrogant. It is in my nature to be so, after spending 40+ years almost exclusively dedicated to audio design from microphone to loudspeaker.
I feel like the 'old professor' although I do not expect to be treated like one, but a little more respect between colleagues would make things flow easier.
Yes, it is true. I put a dual carburetor assembly on a 1954 Chevy 6 in 1958. It didn't help much, but taught me a lot. Lived with dual SU carbs on a AH Sprite for 5 years. Added a 2 throat Weber for a single throat (something) on my 68 Saab Sonnett that I drove for 23 years. However, by 1974, fuel injection became available to me, and I got my first. It had problems, could only be adjusted by an expert, but it had subtle advantages in operation.
For me, IC's are much like carburetors. They are fundamentally compromised from the beginning, and all you can do is work around their limitations.
Now why do I say this?
While Scott is trying to reduce IC operating current (and proud of it) and still remain within very good operating specs, I am at the same time INCREASING current of HIS output stage by the only way that I can, by adding extra components. Who is going to have a better audio circuit, Scott or me?
Please remember Scott's microphone circuit. He is using a NEW op amp with even LESS operating current. Is this design philosophy going to succeed, subjectively?
Should we switch to IC's, unless there is VERY GOOD REASON? One of the last vestiges of discrete design is in the microphone electronics. It is very difficult to change it, once designed, and while many examples of commercial microphone electronics are pretty marginal, is it really better to change to an IC ? What real advantage is there? Are there any drawbacks? Is it really a lower noise approach? What about capsule loading and its real needs for low distortion operation? These are still the real questions that have not been answered.
I regret that my 'tone' has been arrogant. It is in my nature to be so, after spending 40+ years almost exclusively dedicated to audio design from microphone to loudspeaker.
I feel like the 'old professor' although I do not expect to be treated like one, but a little more respect between colleagues would make things flow easier.
In case there is a PO for an MC pre design, and purchaser is not willing to produce with discrete components, and does not want to hear about JFETs at all (I have to admit that I understand why), there is not much chance but OA. The best possible OA for this position, as it is "reference line".
- Status
- Not open for further replies.