Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

Bob Cordell's VinylTrak
Bob Cordell's VinylTrak
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 6th April 2020, 12:44 PM   #131
Bob Cordell is offline Bob Cordell  United States
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by PSchut View Post
Hi Bob,
With great interest have I been reading the VinylTrak article in Linear Audio and what you added on your home page.
I know it is already a reasonable old article (8 years) but in the time of vinyl playback it is still fresh.
If it is alright I would like to ask you a few questions?
1 Why not use a single LSK389 for the MM stage instead of the LS844? (I don't see the advantage of the 844 in this circuit)
2 would paralleling multiple LSK389 not work for MM, the cascode input stage should minimize input capacity.
3 Why use a current mirror as a constant current source for the input stage instead of a reference based current source? (is that to control startup behavior?

Thanks for a great article.
Peter
Hi Peter,

Thanks for your kind words. These are good questions.

The MM stage does not really need the extremely low noise of the LSK389, and I opted for the lower capacitance of the 844 in this appication. It was strictly a judgement call, and a 389's capacitance can be mitigated by a cascode, especially with a driven cascode where a copy of the signal is effectively applied to the bases of the cascode transistors. Also, my choice these days would be the LSK489, which is similar in characteristics to the 844, but benefits from better processing to reduce noise. The LSK489 provides a very good combination of low noise and low capacitance. Its transconductance is similar to that of the older 844.

Paralleling multiple LSK389 for the MM stage would lower the input voltage noise, but still increase the effects of capacitance, even with cascode mitigation. I have not done the analysis, but with that many LSK389s in parallel, combined leakage current shot noise might begin to come into play. Shooting for extremely low input voltage noise in the MM preamp in the real world where the source is an MM cartridge is not necessarily that effective. This is because the cartridge resistance contributes thermal noise anyway. Also, the typical 47k cartridge load resistor contributes thermal noise. The latter can be mitigated by synthetic cartridge loading techniques in some designs.

I think a reference-based current source would do just fine in that application. I think my choice there was fairly arbitrary. Precision and drift of the current value is not an important factor.

Cheers,
Bob
  Reply With Quote
Old 6th April 2020, 02:47 PM   #132
PSchut is offline PSchut  Netherlands
diyAudio Member
 
PSchut's Avatar
 
Join Date: Mar 2004
Location: Tilburg
Hi Bob,
Thanks for the fast reply.
1. makes sense thanks for your explanation.
2. ok, yes indeed the higher MM impedance is more dominant, but I would have hoped that it is less than 47k. an average MM cartridge is 500 -1500 Ohm? Still a lot higher than MC
3. Ok thanks

About 17 years ago I designed a phono amplifier for Classe Audio (that never went into production) with also 2 individual input stages.
The MM had a 2SK389 (I would think a close equivalent to the LSK389) input stage with a gain of 10 (differential stage with no feedback) and an MC stage identical to the MM but with a MAT02 and a gain of 100.
The following stages were full differential opamps circuits with a passive RIAA stage in the middle.

I like your approach better with a passive 75ÁS stage followed by an active 318/3180ÁS stage. The opamps don't need to work as hard and this is better for headroom.

As I am designing a new one I wonder if the quad LSK389 would be better than a MAT12 topology.
choices choices.
Thanks again,
Peter
  Reply With Quote
Old 6th April 2020, 03:28 PM   #133
Bob Cordell is offline Bob Cordell  United States
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by PSchut View Post
Hi Bob,
Thanks for the fast reply.
1. makes sense thanks for your explanation.
2. ok, yes indeed the higher MM impedance is more dominant, but I would have hoped that it is less than 47k. an average MM cartridge is 500 -1500 Ohm? Still a lot higher than MC
3. Ok thanks

About 17 years ago I designed a phono amplifier for Classe Audio (that never went into production) with also 2 individual input stages.
The MM had a 2SK389 (I would think a close equivalent to the LSK389) input stage with a gain of 10 (differential stage with no feedback) and an MC stage identical to the MM but with a MAT02 and a gain of 100.
The following stages were full differential opamps circuits with a passive RIAA stage in the middle.

I like your approach better with a passive 75ÁS stage followed by an active 318/3180ÁS stage. The opamps don't need to work as hard and this is better for headroom.

As I am designing a new one I wonder if the quad LSK389 would be better than a MAT12 topology.
choices choices.
Thanks again,
Peter
Keep in mind that, say, a 1000 ohm MM cartridge resistance will introduce a full 4.1 nV/rt Hz. A 47k cartridge load will introduce the amount of the 47k's open-circuit thermal noise divided by the voltage divider formed by 47k against the cartridge impedance.

I used the quad JFET LSK389 approach to demonstrate that a JFET-input MC preamp could provide low-enough noise for the MC application. I did this partly because JFETs are more forgiving of EMI ingress that might impair the sound. Overall, I just like JFET input stages :-).

The quad of LSK389s is, of course, a rather expensive approach.

Cheers,
Bob
  Reply With Quote
Old 6th April 2020, 05:02 PM   #134
PSchut is offline PSchut  Netherlands
diyAudio Member
 
PSchut's Avatar
 
Join Date: Mar 2004
Location: Tilburg
indeed and yes, I gathered you like JFET's
Take care
Peter
  Reply With Quote
Old 7th April 2020, 07:14 PM   #135
PSchut is offline PSchut  Netherlands
diyAudio Member
 
PSchut's Avatar
 
Join Date: Mar 2004
Location: Tilburg
Quote:
Originally Posted by Bob Cordell View Post
Keep in mind that, say, a 1000 ohm MM cartridge resistance will introduce a full 4.1 nV/rt Hz. A 47k cartridge load will introduce the amount of the 47k's open-circuit thermal noise divided by the voltage divider formed by 47k against the cartridge impedance.

I used the quad JFET LSK389 approach to demonstrate that a JFET-input MC preamp could provide low-enough noise for the MC application. I did this partly because JFETs are more forgiving of EMI ingress that might impair the sound. Overall, I just like JFET input stages :-).

The quad of LSK389s is, of course, a rather expensive approach.

Cheers,
Bob
Hi Bob,
I thought I need to simulate this so after 15 years I entered spice again. I used to use MicroSim PSpice but that doesn't run anymore and found on your website the link to LTspice.
Installed, got it running, a bit of get used to, I still don't like the FFT as it shows -330dB of S/N on a generator.
But I got a diff pair up and running and I am now comparing MAT02 to LSK389C. Also the effect of cascode in this stage looks realy reliable.
The funny part is that they both the MAT2 and LSK389 are part of the fresh install (dated only a week ago) but the LSK389C is broken, the LSK389A and B work.
And yes I get a nice -82 dB of S/N with 3 pieces parallel.

Do you have an idea what could be wrong with the "C" version?

Thanks, and great tip LTspice......
Cheers,
Peter
  Reply With Quote
Old 7th April 2020, 08:01 PM   #136
Bob Cordell is offline Bob Cordell  United States
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by PSchut View Post
Hi Bob,
I thought I need to simulate this so after 15 years I entered spice again. I used to use MicroSim PSpice but that doesn't run anymore and found on your website the link to LTspice.
Installed, got it running, a bit of get used to, I still don't like the FFT as it shows -330dB of S/N on a generator.
But I got a diff pair up and running and I am now comparing MAT02 to LSK389C. Also the effect of cascode in this stage looks realy reliable.
The funny part is that they both the MAT2 and LSK389 are part of the fresh install (dated only a week ago) but the LSK389C is broken, the LSK389A and B work.
And yes I get a nice -82 dB of S/N with 3 pieces parallel.

Do you have an idea what could be wrong with the "C" version?

Thanks, and great tip LTspice......
Cheers,
Peter
It's probably a formatting issue. The LSK389C model should look very much like the A and B versions, with just a few numbers different to account for the higher Idss of the C version.

Cheers,
Bob
  Reply With Quote

Reply


Bob Cordell's VinylTrakHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Bob Cordell's Power amplifier book ihan Solid State 10070 Yesterday 08:53 PM
Another realization of Bob Cordell's THD Analyzer giulianodes Equipment & Tools 213 23rd January 2020 01:30 AM
bob don't like digital tomtt Digital Source 1 30th August 2006 01:40 PM


New To Site? Need Help?

All times are GMT. The time now is 07:48 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki