Mpp - Page 246 - diyAudio
Go Back   Home > Forums > Source & Line > Analogue Source

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 3rd August 2011, 10:29 AM   #2451
diyAudio Member
 
Benedetto's Avatar
 
Join Date: Nov 2007
Location: Knifetown-Germoney
Quote:
Originally Posted by hesener View Post

hope you like it!
This is really a cute one.
What will be the cost if anything is said and done ?
  Reply With Quote
Old 3rd August 2011, 11:30 AM   #2452
hesener is offline hesener  Germany
diyAudio Member
 
Join Date: Mar 2005
Location: Munich, Germany
good question..... how much for the hours of work? the material is not that expensive. the kits for the phono PCBs over at pilghamaudio.com were 270 euro, the shunt regs, power supply and case probably another 200 euro or so, all in all maybe 500 euro material. Not that expensive really, comparing it to much more expensive commercial units.... except for the hours, of course. And, I can'T claim that my build quality is as beautiful as the commercials!
  Reply With Quote
Old 3rd August 2011, 11:36 AM   #2453
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
Hesener, this is a much more sophisticated build then i expected. You did everything you could to make it sound good. I am very happy when i see it. I have two questions :
1 : could you please measure the voltage over the gain setting resistors between the input Fets ? It is the low value one, around 2 Ohm.
2 : could you draw a schematic of the modified Salas ? I have sometime trouble when i use it so i can learn here something.
P.S. you can cool the opamps of cause. There are some cooling fins that fit. I send a photo. I am of cause very proud that you prefer it over the Pear that has a good reputation. I find single ended Fet stages just a little high on distortion, even when they are cascoded. Yes, it must be the K2 that adds that kind of warmth. It can mask detail though.
  Reply With Quote
Old 3rd August 2011, 01:02 PM   #2454
hesener is offline hesener  Germany
diyAudio Member
 
Join Date: Mar 2005
Location: Munich, Germany
Thanks for your nice comments, and of course for the nice design that enabled this in the first place!

Will measure the voltage tonight when I get home. This should reveal the current in the input stage, right?

On the shunt regulators, attached is the schematic for the positive regulator. The negative regulator is the same, except for the reversed polarity of the MOSFETs, LEDs, diode and bipolar; otherwise identical.

What kinds of issues do you see with this regulator? Just to point out three things I implemented here:

1) the current source is a degenerated MOSFET, that allows for a nice point for measuring the current, has excellent noise behaviour (much better than LM317 - I tried), and lower voltage drop (the LM317 needs around 3V to start operating properly, this one works at 1V or less voltage differential - 2..3V is better though)
2) The diode in the emitter of the PNP increases the drain voltage of the JFET used as constant current source, making the reference voltage more stable
3) R6=3k3 is on the high side, the current starving for the PNP gives a lowish gain, which combined with the Q3 gain is high enough but still stable.

One issue that might happen is oscillation around Q3, but that can be fixed with R8 (not needed in my design) or a small capacitor from gate to drain (100pF or so).

The temperature stability is not great, as I pointed out in the original "simplistic" thread, due to VBE of the PNP decreasing with temperature, and the MOSFET's positive tempco (at least at these lowish currents). It is not helped with the additional diode.

But the drift is not large, and it could be fixed, but then I thought so what. Adjusting the output voltage after thermal settling is good enough. Of course, 10-turn pots for both the current and the voltage are advisable, especially for the current as the adjustment is quite sensitive.

I did measure the output impedance of the shunt regulators with a electronic load that can sink a DC and AC current, and the output impedance (= noise voltage / AC current) was 4mOhm flat from 20Hz to 22kHz. Low and flat, thats what we want right? (in this case at least ;-)
Attached Images
File Type: jpg shuntreg.jpg (197.7 KB, 246 views)
  Reply With Quote
Old 3rd August 2011, 01:34 PM   #2455
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
Ok, that is a version of the Salas i do not know. I had trouble with oscillation in one case. I think i used the Mesmerize. Maybe i just loaded it to heavy with Elcaps but i also tried decoupling with resistors and it did not help. On another stage it worked well.
Yes, the measurement on the gain resistor gives the idle current in the input stage. It does not matter much. Non matched Fets have no effect on the noise performance ( Scott Wurzer pointed that out ) and a unbalance in Gm does not effect distortion much ether because of the feedback structure. I am just curious. The amount of current that flows totally of cause gives a clue of how low the voltage noise is.
  Reply With Quote
Old 3rd August 2011, 02:11 PM   #2456
hesener is offline hesener  Germany
diyAudio Member
 
Join Date: Mar 2005
Location: Munich, Germany
Interesting point... loading this regulator with large capacitors causes a phase shift at higher frequencies, hence oscillation.... Did you try and put resistors in series with the cap (to increase ESR)? It's counter-intuitive but works well. Audio Research used a similar trick, albeit for other reasons, in their tube line stage power supply (forgot the model, I think it was the LS25).

In my phono stage the 1.5uF at the output is all I use! It feels strange to NOT have big caps there, and when I unplug it goes off immediately..... never had that before.... but the listening results and also the measurements indicate otherwise.

I remember I did measure the voltages across the 2x47ohm resistors in the current sources, both sitting at about 400mV, so 17mA. But I will take another look.
  Reply With Quote
Old 3rd August 2011, 03:17 PM   #2457
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
Yes, i used resistors at the output. In the meantime i got quite good to making my own PSU´s so there was no big motivation to look into this more carefully.
17mA sounds about right.
  Reply With Quote
Old 3rd August 2011, 03:26 PM   #2458
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
hesener:

Hey, no remote sensing? Nice builds.
  Reply With Quote
Old 3rd August 2011, 03:31 PM   #2459
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
That transfers into a noise resistance of ca.15 Ohm including the feedback resistor. That is 0.5nV/qHz, right on target. That gives a signal-noise of ca.-77dB unweighted. So we have more then -80dB weighted. The only stages i know that are better have 0.3nV/qHz but they are single ended and do not have the advantage of balanced. The other issue is that the Johnson noise of the cartridge itself adds to the noise. Say it is 10 Ohm. Then in one case we have 25 Ohm
( FPS ) and in the other we have 15 Ohm ( physical limit ). That reduces the theoretical advantage of the single ended ( with the same amount of the same input devices with the same idle current ) to less then 2dB. I can life with that and the single ended will most likely pick up more hum and that is annoying too.
  Reply With Quote
Old 3rd August 2011, 03:33 PM   #2460
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Quote:
Originally Posted by hesener View Post

But the drift is not large, and it could be fixed, but then I thought so what. Adjusting the output voltage after thermal settling is good enough. Of course, 10-turn pots for both the current and the voltage are advisable, especially for the current as the adjustment is quite sensitive.

I did measure the output impedance of the shunt regulators with a electronic load that can sink a DC and AC current, and the output impedance (= noise voltage / AC current) was 4mOhm flat from 20Hz to 22kHz. Low and flat, thats what we want right? (in this case at least ;-)
Use LEDS instead of Norton Vref if better drift is desired, but it can be inconsequential. You skipped capacitor filtering the Voltage reference resistor and trimmer on purpose? That could be an important aspect. Your build works well for Zo in its simple form, congrats.
  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



New To Site? Need Help?

All times are GMT. The time now is 07:24 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2