Favorite op-amps for RIAA preamp?

[jcx]

a long time since I looked at phono carts, but noise limited amplification is my daily concern designing strain gage amplifiers

700 Ohms DCR seems low for a mm, the coil R is in series with the coil L and both are paralleled with the winding C and recommended load C to give a resonant circuit at mid audio frequencies, the cart noise is dominated by the DCR Johnson noise and any op amp Inoise flowing in the cart impedance

700 Ohms is well above the LT1028/AD797 noise optimum, among bipolars lower bias devices with more Vnoise but less Inoise would be better, OPA227/228 is a closer match for 700 Ohms

mm carts are also reputed to be highly allergic to Idc which saturates/magnitizes the Iron - so input coupling caps are highly recommended with bipolar input op amps - and when making any impedance measurements

from a noise perspective I wouldn't consider anything other than the fet input AD743/745 for mm cart, even at 700 Ohm DCR
the OPA627/637 are the next best fet input op amps I know of noise wize with a high reputation in audio apps

[Conrad Hoffman]

The new LME49710 opamps came in yesterday, and this morning I checked them out both on the bench, and with a few LPs. My basic philosophy is that below certain limits, everything sounds the same, but the fact is that different opamps, especially in low level front ends, RIAA filters, and cable drivers, sound different. By differential measurements, one can see signal differences. Using SA, one could probably learn more, but I haven't done that yet. Oh, the LME49710s... First, the noise level is lower than anything else I've used. Neither the LT1028 (non-optimal anyway), nor the OPA627, nor the AD845, nor the 5534, was as good. In fact, the noise level of my preamp dropped by a factor of 4. The noise level is so low I'm having a heck of a time getting the residual hum from the ps transformer down where I want it, and distortion measurements are too contaminated with hum to be meaningful. The inverse RIAA box and cabling introduces a lot of it as well. Understand that this is hum you still have to crank the volume level way up to even hear, but it gets in the way of the lowest level measurements. Offset is also extremely low, but I did select the front end parts for minimum offset just because I could. Sound-wise, I'm probably not the golden eared boy to ask, but I wasn't happy with the LT1028, 5534, AD845, or OPA627. They were either harsh or flat sounding. The LME49710 is near to perfect. Brushes on cymbals sound correct, female vocals like Joan Baez Diamonds and Rust sound exactly the way you think they should, and bass is neither thin nor exaggerated. Now I just have to fool with the grounding, shielding, and ps to get the residual hum down, then change out the remaining silver mica caps. Hopefully those things will make it sound better, not worse! No doubt, time will illuminate whatever failings the LME49710 has, but right now I'm pretty amazed at how good it is.

[Conrad Hoffman]

Ok, fixed some grounding issues and reduced the hum a bit. What I really need to do is ditch my power supply and build something with a toroid, or just go battery power. Never thought it would matter, but at this level it does. Going through an inverse RIAA box, the THD+N is about 0.007% @ 1khz. My guess is it's mostly N, not THD. Spent most of the afternoon listening to LPs, and the sound is still very pleasing. I re-recorded (digital) some tracks that I had also done with the old preamp, and compared them. The differences are real, mostly the improvement in the high end- more shimmer and musical, than HF noise, but it's subtle. Still have to pull the board out and change the caps.

[MikeBettinger]

Originally posted by jcx
from a noise perspective I wouldn't consider anything other than the fet input AD743/745 for mm cart, even at 700 Ohm DCR
the OPA627/637 are the next best fet input op amps I know of noise wize with a high reputation in audio apps

Based on Conrads comments I'm going to go ahead and order the National chips. I'll try both, but I'm guessing the newer designs will be the keepers. I thought I'd like to try either the AD743 or the OPA627 for my supplies but find neither Newark or Digi-Key stocks them.

Can someone recommend a source for either of these; or a newer (availible) FET input design still in a DIP package?

Thanks, Mike.

[Conrad Hoffman]

Once more into the chassis... I replaced all the caps with polypropylenes- nothing fancy, the little brown 2% parts from Digikey, selected by me to about .5% or better, and replaced the one non-metal film resistor with a metal film. The residual hum was cured by twisting some wires and using a ground strap between the chassis and the chassis of the inverse RIAA box. Apparently even a short RCA cable isn't up to the task for this low amplitude a measurement. THD went down to 0.004% at 1Vrms out, and the RIAA is within two tenths of a dB. THD+N is still inversely proportional to level, so it's still likely noise, not actual THD. It's too late for much listening, but IMO the thing sounds fantastically smooth and musical for the brief time using it. Like Dan in 60 Minute Man (Jerry Lee Lewis and others), the new National chips appear to be all they say.

[MikeBettinger]

PCBs ordered, National chips ordered, I think I have everything else.

Now I wait for UPS and time to put it all together. It's like bringing a child into the world, without the 20+ year commitment.

Regards,

Mike.

[redrabbit]

Johnson Noise


http://www.gellerlabs.com/JCan%20Parts%20and%20Kits.htm

=RR=

[jackinnj]

Originally posted by redrabbit
Johnson Noise


http://www.gellerlabs.com/JCan%20Parts%20and%20Kits.htm

=RR=

Subject of the N&V article --

[drdagor]

Seems like several of you had good projects going. What results did you get with the new National preamps, and can you share circuit ideas?

I'd love to listen to the third act of this little radio drama.

[gmphadte]

Just happened to read some part of the thread

I recall seeing datasheet of op77 (maybe IIRC, but that is not important)
I read that these have same slew rate for positive and negative sides of the signal, which made them suitable for audio.
Audio was mentioned in their application list.

That made me realize that op27/37 were not good for audio
What unmentioned parameter makes many opamps non-audio, mfr only knows.

Gajanan Phadte