A simplified universal differential or single ended phono preamp - diyAudio
Go Back   Home > Forums > Source & Line > Analogue Source
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Analogue Source Turntables, Tonearms, Cartridges, Phono Stages, Tuners, Tape Recorders, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 5th April 2017, 05:09 PM   #1
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
Default A simplified universal differential or single ended phono preamp

This thread split off from the Blowtorch part II thread. April 20th 2017

I promised Patrick that I would post the final schematic of this circuit so here it is. I have a lot of JFET's in bags here and there but I didn't get inspired until I realized how simple the common-mode feedback would be to do with nothing but FET's. There is nothing that special here, a simple undegenerated diff-pair, folded cascode, and buffers. J11 provides common-mode feedback and for a differential signal has no "work" to do but keep the common-mode DC level around 0.0V, so I don't even consider it in the signal path. So we have a purely current driven RIAA network with all the time constants between two nodes. The AC coupling in the output is optional, I don't like servos so what.

Almost everyone I show this to starts adding more supplies, extra cascodes, current sources, etc. that's not the point. Bob Cordell published a phono pre-amp in Linear Audio and I started looking at this as removing things from it until it stops working (Muntzing).

The output buffers are a little side experiment. I chose the FET's in a way contrary to just about what everyone else does. I did NOT match the Idss's, instead I sorted them such that a single standard value of source resistor would set the operating current at the same value. I picked 5mA and used a homemade curve tracing jig to sort 25 each N and P FET's. So now when they are connected together the input to output offset is 0, no trims. It turns out that this had less un-trimmed distortion (in sim) than selecting equal Idss's and trimming for just offset (story for another day). Worked fine when stuffed all four were < 5mV. I picked some reasonable matches for the rest of the pairs. I had to snub the output of the buffers with 680pF.

Notes on the amplifier:
I chose the LSK489 FET pair (thanks Paul Norton and Linear Systems for their support of the DIY community) because it had higher Vp and lower transconductance (gm) which gives a wider range of linear input and keeps the RIAA network's impedance level from getting too low. Bob used the LS844 which has the same gm at 1mA and chose 20dB midband gain but his load resistor value does not reflect that(Bob ???). I chose 26dB midband gain and used the 3mS gm at 2mA from Bob's app note which should be the correct value. R7 is chosen for around 1.0ma per side of quiescent current, that and a goal of 46dB of DC gain gives you a target for R19 and R3. At 2.2mS gm for the FET at 1mA two 90.9K resistors were chosen and the rest of the RIAA values fell from that. The caps were all chosen from a large NOS stash of TRW 1-3% polystyrene hence 9.375n was actually an exact available value. The resistors are all 1% Dales with R20 two standard values in series, while C4 was two caps in parallel.

Since all LSK489's have nearly the same gm at 1mA R7 was a trim on the second channel for gain, and it was easy to get nearly .1dB match. The lowish midband gain means a low noise line stage is necessary if you want the SNR limited by the input FET noise alone over the entire audio band, if you take A weighting into account I'm not sure this matters much at all. For first listening I will use a low noise in-amp at a gain of 14dB to bring up the midband gain to a more standard 40dB and give me a single ended output since as yet I don't have a differential input system.

R18 and R17 are redundant offset trims, the idea was that by skewing one and bringing the offset back with the other one might be able to null the residual second harmonic. The distortion was so low I would not include both again.
The three green LED's set the current in the folded cascode to around 1.75mA this is not super critical it just wants to be enough. R4 could be a trim, it sets the common-mode voltage but I just computed the value and put it in.

First light:
Distortion at 1K with .025Vrms in, ~-87dB (.0045%) Nothing much other than 2nds and 3rds (almost the same as Bob's). The noise was input FET limited at 3-4nV/rt-Hz (out to the last octave or so) fine for MM/MI which was the goal. MC would require some rework.

RIAA accuracy is pretty good (the drop at 20k is Windows brick-walling my soundcard at 22.05k in the background) and the low rolloff is half due to my added high pass at the output. Mostly +-.1db out of the box. Picture is both channels (offset is random and does not indicate the gain mismatch). The RIAA response used a 1/3 octave multi-tone (30 tones) signal pre-emphasized at the same .025mV rms in which will have a crest factor more like recorded music. No IM noise floor was observed.

This is a work in progress, the recent availability of complimentary pairs of FET's from Linear Systems means that it may be possible to realize this entire signal path with LSK489's and LSK689's. I know this is JC's thread but it is the lounge after all. Comments like, "We did this fully complementary 40yr. ago", etc. are inevitable, I've learned there is no point in asking them not to come flying.

EDIT - The supplies are +-15V and the bypassing is left up to the builders personal practice.
Attached Images
File Type: jpg phonopre.jpg (132.2 KB, 967 views)
File Type: jpg phonoRIAAcombo.jpg (106.3 KB, 923 views)
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."

Last edited by scott wurcer; 5th April 2017 at 05:23 PM.
  Reply With Quote
Old 5th April 2017, 05:40 PM   #2
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: Northern Iliinois
Thanks Scott for the schematic and detailed explanations.
Playtime begins tonight ...
Now, where did I put those TRW caps ...

mlloyd1

Quote:
Originally Posted by scott wurcer View Post
I promised Patrick that I would post the final schematic of this circuit so here it is.
  Reply With Quote
Old 5th April 2017, 05:43 PM   #3
DPH is online now DPH  United States
diyAudio Member
 
Join Date: Dec 2008
Location: Portland, OR
Scott -- how does it do on overload? I'm not in the mode of a RIAA, but the CM feedback mechanism and the overarching elegance of the circuit makes for a fun idea to play with (to this amateur).

And of course, thanks for the circuit, and, IMO more importantly, the explanation.
__________________
Happy DIYing, Daniel
(This space is occupied by the artist formerly known as Derfnofred)
  Reply With Quote
Old 5th April 2017, 06:00 PM   #4
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
It behaves like Bob's for the most part, I'll look at clipping of the actual build since I have relied on sims for that. I expect very soft compressive behavior especially for tick and pops.
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 5th April 2017, 06:17 PM   #5
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: The great city of Turnhout, BE
Blog Entries: 8
Thanks Scott for a voice of reason and rationality - had a bad day with the trolls on diyaudio..

Jan
__________________
Music is dither to the brain; lets me think below the usual chaos - me
Linear Audio Vol 13 is out! Check out my Autoranger and SilentSwitcher
  Reply With Quote
Old 5th April 2017, 11:02 PM   #6
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
What is the CMR vs freq and the +/- PSRR vs freq?


THx-RNMarsh
  Reply With Quote
Old 5th April 2017, 11:12 PM   #7
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
Not bad actually, better than you would think. Put this on a 5V battery and a Silentswitcher, time for some DBT if you can stand it.
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 6th April 2017, 04:59 AM   #8
wayne is offline wayne  United States
In the hills
diyAudio Member
 
Join Date: Jan 2002
Location: Northern California
Scott What do you see without the 680 pF? Pretty neat circuit.
__________________
wizards assistant
  Reply With Quote
Old 6th April 2017, 12:29 PM   #9
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
There was some 100MHz fur probably gate stoppers would also get rid of it but that would involve cutting and soldering and I already killed one 75n poly cap by just brushing it with a hot iron.
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 6th April 2017, 01:29 PM   #10
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
Quote:
Originally Posted by DPH View Post
Scott -- how does it do on overload?
As good as I expected, more than 25Vp-p at clipping and at 20Hz there is a little droop after clipping from the common mode loop or the coupling cap, nothing pathological. Your woofers would be long gone before this. THD at 250mV rms in is still <0.5% and still no 7th's . FET's just can't make high order harmonics.
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Differential to single ended recommendation (PCM179x) sfasf Digital Line Level 20 20th March 2013 01:11 AM
single-ended to differential? gentlevoice Digital Line Level 16 17th December 2008 06:18 AM
testing a single ended or differential topology Stefanoo Pass Labs 11 26th October 2007 05:59 PM
Converting from single ended parafeed to differential dsavitsk Tubes / Valves 9 12th July 2007 06:46 PM
Differential to single ended... Bas Horneman Tubes / Valves 21 2nd April 2003 02:05 PM


New To Site? Need Help?

All times are GMT. The time now is 01:39 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki