• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Retro - A fully symetrical phono stage with RIAA filter

How about a version using inductors?

Since this is all the rage.

Would be easier to implement with op amps.

One does have a hard time returning to digital from analog but what do you do for the records that are only available on CD?

I am not one to speak, since getting a turntable set up after two years of abstinence I have yet to re-connect the digital stuff. I am sure I will eventually. But then that is what I said six weeks ago. It is getting close to six months now ...

Allen Wright's missing time constant could certainly make a better sounding phono amp.
 
Russ,
Bravo on your fully symmetrical phono pre. Why 99% of people treat this balanced symmetrical signal source as unbalanced is beyond me.

We do it with tubes and zero NFB, but yours is a good as I've seen with ICs.

But you could add a tweak to the RIAA curve that more accurately mirrors what goes down on the disc:

http://www.vacuumstate.com/fileupload/SP_15_Article.pdf

See #4, "Add the missing time constant".

Regards, Allen (vacuum State)

Thanks Allen.

I will certainly take a look.

Cheers!
Russ
 
Misc. comments:

"DNR" -- I questioned because that is not a standard technical abbreviation as far as I know. Much like the strange use on this forum of the term "trafo" to mean transformer. Who invented this?

EQ: Hagerman and Wright advocate the so-called "Neumann pole" - debunked multiple times, most recently in Stereophile a few months ago. Lipshitz' paper is the definitive work on the topic -- and you need to get DC to 10kHz correct before you even start to worry about the Neumann pole pros/cons. I can send a copy of Lipshitz if you need it. Walt Jung's Audio Amateur article was on an inverse network. (It's amazing how so many youngsters are ignorant of such basic information. A few minutes with Google should point you to enough references to start researching. A few years ago an applications engineer from TI published some info about RIAA EQ where he stated that it was impossible to design such a circuit mathematically and the only way to get it right was by fiddling with component values in SPICE -- obviously did zero research on the topic)

"balanced differential" is perhaps a misnomer for a phono cartridge. With the exception of strange devices like the Decca, the L and R coils on either MM, MC or MI are floating in 'electrical space' if not specifically tied to chassis ground (one channel usually to the cartridge/cable/chassis ground but in your application this would be lifted). If not connected to ground they are not 'balanced' as they are not symmetrically referenced to circuit ground and are not 'differential' until connected to a differential input stage.

I need to ponder the latest schematic with the two FDAs. There seem to be more 'moving parts' with odd connections that I need to ponder. So, Russ, are these FDAs OPA1632 or THS4131 ? :)

Keep up the good work, Russ. I enjoy both analog and digital, but must admit there's a level of tweaking and DIY that can be done with Vinyl that can't be done as readily in digital. I.E., a DIYer can actually build a phono playback setup from scratch (save for cartridge and the actual drive motor); you can't do that with a CD player.
 
yes, I know it works fine. My comment regards the sloppy use of the terms in someone's posting. Of themselves the L and R cartridge coils are floating with respect to the world, until connected to something else. Balanced implies that the two pins move in equal and opposite directions relative to the (ground) reference, but that's not true until connected to something. And really, it's that 'something' that forces them to behave in a balanced manner. The coils are only 'differential' if they are connected to a differential input stage.

This may seem like technical nit-picking, but misunderstanding the reality can lead to incorrect approaches. See Bill Whitlock's papers on his "In Genius" circuits. He shows how many approaches to balanced (line) inputs yield poor results because they are designed with a misunderstanding of what is going on.

That said, your circuit looks fine. I still need to ponder the various extra moving parts. Are you planning on providing a more detailed description of the functions of various circuit elements?

FDAs: yes, either p/n should work identically :)
 
Hi Brian, The extra moving parts section you are probably most curious about is the servo section.

The other two blocks are pretty standard FSA fare, with the notable addition of the "H" network to increase gain.

The servos actually do two things. They keep the outputs differentially at 0VDC. They are also summed and taken as negative feedback at VOCM to keep the output common mode at 0V. This will come in handy for people how want to just take one end's output for SE use, but that is just a nice side effect of the overall goal of nulling the DC. That op amp will be a dual JFET type for very low offset and high input impedance, but once again, I am very open to suggestions. :)

Cheers!
Russ
 
Last edited:
Russ,

I hadn't had the time to analyze it much - just printed it the other night before bedtime. I figured a servo or perhaps some CM bootstrapping scheme ala "In Genius".

Low offset FET-input amp. Sounds like an OPA827 or OPA627 to me, or possibly the high-grade flavor of OPAx132. OR AD8610 if you don't mind reduced power supply voltages
 
Russ, a few more thoughts.

I'm not sure one of the FDAs is the best choice for your first stage. Although the noise at high frequencies is low, the OPA1632 and/or THS4131 have pretty high 1/f noise corners both for voltage noise and current noise (see page 5 of OPA1632 datasheet). I think this will become an issue especially with MC cartridges. You may not find the "sound emanating from inky blackness" to borrow a phrase from the high-end audio press.
 
Hi Brian,

Yes, I thought about that too. But the background seems to sound pretty "black" to me even with the front end opa1632 in trans impedance mode. :)

I am still thinking I will try both version. The fully symmetrical trans-impedance input version, and the instrumentation amp input version.

May I should call them Low-Z and High-Z. :)

Here is what I am using right now. I think the High-Z is the version I will order first. As shown it has 32db gain on the input and a total gain of 40db. This has been plenty of gain for my HO-MC cart.

I have not used the servo part yet because I have been lucky enough not to have any crazy offset. But I am pretty sure it should be fine.

Last chance for input before I order some for myself and a few beta-testers.

Cheers!
Russ[/ATTACH][/ATTACH]
 

Attachments

  • schematic.pdf
    34.5 KB · Views: 940
Last edited:
Success!!!

I have a pair of Retro-Hi Z running right now. :)

I Am listening to the Dave Holland Quintet right now as I type.

Since I am using a HO MC cartridge I have the gain at about 40db at 1Khz. I find that I could actually decrease that a bit and still have enough output for my setup. I Will probably make it 34db.

The servo is working nicely with the output staying at < 500uVDC differential.

I don't like to critique my own stuff. But it sounds great to me. :D

I am running mine dual mono. With one PlacidBP(and trafo) and one retro per channel.

Here are pics of the completed Retro modules.

I could not wait for my Mouser order to get here so had to parallel some Rs to get the values I needed. :)

I also would normally use .1% Rs, but I had time to carefully match 1% Rs. This proved very much good enough.

Cheers!
Russ
 

Attachments

  • Retro 001_small.JPG
    Retro 001_small.JPG
    214.9 KB · Views: 1,598
  • Retro 002_small.JPG
    Retro 002_small.JPG
    270.3 KB · Views: 1,569