MPP

Hi Frans

What is the purpose of the smaller case to the left ? The one with dedicated psu ?

At the back there is an copper box with ADC (left/back) and metal box with mains connector and filter (right/back). Front left meta; box PSU for the ADC next big copper box RIAA preamp next metal box PSU for RIAA preamp.

In the RIAA preamp there are 3 sections, first whider section is secondary PSU, middle small sections tertiary PSU and large sections RIAA preamp.

The primary PSU for the RIAA preamp in the big metal box (right). The copper tube is used to transport mains power to the ADC PSU.

Balanced RIAA preamp | Flickr - Photo Sharing!
Balanced RIAA preamp | Flickr - Photo Sharing!
 
The test for the Rega turntable is quite interesting.
 

Attachments

  • Vinyl Text.jpg
    Vinyl Text.jpg
    121.8 KB · Views: 327
Have been looking at the input impedance, but not really sure how to set up a good measurement for this, the generator input impedance may play a role in the equation...
but according to this the impedance is app 2Kohm,,,to 100 KHz then slowly decreasing.. (input voltage is 1 V and The current is monitored)

Also I have a little problem with the internal shunt, as there's no absolute reference it varies with the input-currents from the floating JFET current sources, could be a LED string reference combined with the JFET CCS's is a good solution, here I have degeneration problems as the CCS must pass app 15mA to maintain a 10 V cascode spread.
 

Attachments

  • PAradise-Lost input impedance.JPG
    PAradise-Lost input impedance.JPG
    153.3 KB · Views: 313
MiiB,
if you could provide me with the LT-Spice file I would look it up. Otherwise, it's though for me to see why we'd need 15mA for a Cascode reference? Same for the floating voltage issue.

Rüdiger

EDIT: yes, the voltage varies a bit when the CS fet warms. But since it's independent from both PSU noise and signal interference I doubt it matters. We should run the fet at zero tempco. I did not bother with in my phono pre, and the offset is stable around 0mV

Rüdiger
 
Last edited:
Rudiger,, I use 10 of those mA for the circuit, 2,5 mA for each input transistor, the rest is for the shunt to work....Attached here is a version where all the deratings are OK, the maximun is 100 mW/device

A JFET Current push, will dissipate app 350 mW, and that just too much for the poor device.. and for tha JFET CCS in the mirror for the buffer it was 600 mW, so these are now changed to simple Resistance elements, I do feel that it no a degradation as we already have a very-good shunt regulated supply..

I have changed the internal mirror shunt to a CCS'ed LED refrence, at app 12 V
I have inserted a cascode in the buffer current souce, to reduce power dissipation in the Lower device....
 

Attachments

  • Lost paradise.JPG
    Lost paradise.JPG
    184.6 KB · Views: 296
Last edited: