Paradise Builders

Right between ground and the collectors of the first power transistors you encounter in the regulator's current source leg, i.e. the two outer transistors on the heatsink.

STUPID ME.

Not the collectors, but of course the emitters of Q103 and Q203!!!

I used 470nF.

So that's the D45H11 and D44H11 middle pins to gnd?

See above. Apologies.
 
STUPID ME.

Not the collectors, but of course the emitters of Q103 and Q203!!!

I used 470nF.



See above. Apologies.
Ah...not Q102/104/202/204 then...the usual culprits under the heatsink.

But still the outer power transistors on the heatsink...pin 3...I'll have to try this.

This mod is not in the build guide?

Sent from my SM-G900H using Tapatalk
 
Werner, you said something about 22 uF - is microF or nanoF?. If is micro F, I suppose you mean polarized. And where exactly ("at the supply input connector")?

At the block terminal on the PCB, or on the connector on the case?

But is it possible that the oscillations occur in the Pre regulator section?

Emil
 
26057002053_6b800fbc37_c.jpg


Picture from Werner's modded unit.

22uf input caps, 470n film on the shunt.

When I fitted mine i just used the exposed GND under the bolt holes, easier than scraping the resist.
 
keywords: shunt regulator oscillation problem solution


Last month I bought a lightly-used Paradise/Calvin. It was built end of 2015.

Upon arrival it appeared that all power supply rails were oscillating. There were traces of 5-6MHz all over the place, plus a larger-amplitude triangular component at 400kHz: 20mV on the regulator outputs, and 2V peak to peak on the regulator inputs (!!!).

Ten minutes with a circuit simulator revealed that this is caused in the shunt reg's current sources, due to trace inductance at their input. With the raw supplies in separate boxes and some 60 cm of cabling between them and the actual amplifier boxes you'll bet there is trace inductance. (Incidentally, the simulation both showed the 5-6MHz component and the 400kHz one.)

The simulation suggested a simple cure: just a little bit of capacitance between current source input and ground solves this rigorously. I added 470nF stacked film MKTs right at the transistors' collectors, and 22uF at the supply input connector.

This solved the problem indeed.

Interesting.... this is a new one, has not shown up at least in my builds so far, and my PSU is 1.5m away..... thanks for the investigation and insight!

And new to me, up until now all oscillating PSU have been repaired without looking at the input of the PSU.

In retrospect this has been an oversight, there should have been a 'small' capacitor at the input of the PSU from day one. Also adding a small buffer at the board input is not a bad idea. Well found Werner :) thanks for that.

I'm still not convinced that this is the only reason for the oscillation problems we have seen, at the lease some of the problems have been due to bad (sub standard) components.

One must also take care of the wiring between the pre-regulator and the shunt-regulator. Shorter, thicker and induction-free (if at all possible) paths for the conductors must be carefully selected.

These are the most important/relevant messages regarding PSU and Pre-Reg (that I collected).
  1. Mpp #4375 PSU-pre-regulator.
  2. Mpp #4610 V2 Schemas UPS and Amp.
  3. Mpp #6506 CCS fine-tuning.
  4. Mpp #6462 PSU with my notes.
  5. Mpp #6809 Do not try to feed the CVS of the CCS of the shunt with a regulated power supply.
  6. Mpp #7127 More about transformer selection.
  7. Mpp #7635 Single and dual transformer wiring (also corrected drawing) (See also #7733).
  8. Mpp #7733 Grounding (PSU, RIAA and Player).
  9. Mpp #7810 35V PSU for the CroMagnon.
  10. Mpp #7830 The PSU explained.
  11. Mpp #8386 My resistor recommendation’s.
  12. My Paradise #56 PSU Noise.
  13. Masterpiece #951 That's not a PSRR test (it's a CIR test) :)
  14. PradiseBuilders #13 Schema's and assembly guide.
  15. PradiseBuilders #155 Vdc input voltage minimum simulated.
  16. PradiseBuilders #970 PSU oscillates; posible solutions?
  17. PradiseBuilders #983 Compensating the PSU for oscillations.
  18. PradiseBuilders #1173 Fixing oscillations (up to #1182).
  19. PradiseBuilders #1322 Fuse :).
  20. PradiseBuilders #1387 Updated Paradise R3 assembly guide (Also German).
  21. PradiseBuilders #1403 Capacitors bad vs good.
  22. PradiseBuilders #1593 PSU Power Transistor selection.
  23. PradiseBuilders #1644 PSU Output Impedance.
  24. PradiseBuilders #1680 PSU Power Supply Suppression Ratio(PSSR).
  25. PradiseBuilders #1772 NJF selection for the PSU.
  26. PradiseBuilders #1887 Empirical NJF Rgs determination.
  27. PradiseBuilders #2127 More about the PSU fets (J113's etc).
  28. PradiseBuilders #2969 About the Paradise PSU using Calvin buffers.
  29. PradiseBuilders #3134 More ESR confusion :).
  30. PradiseBuilders #3808 Werner's PSU fix.
  31. PradiseBuilders #3830 PSU oscillating due to 'missing' input buffer/capacitors.
Have a look at these and their following messages.
 
Last edited:
v3.1 if i remember correctly, but v3.1 had a small bug (one resistor wrong placed, so you had to remove on of the pads on bottom side, easily done).
After v3.1 there was also a board made where circuit and power supply was separated on their own boards - but i think otherwise there was no changes made at all.. hesener is the one that has the knowledge on this though, because he made the layouts...

I think maybe v3 doesn't have room for compensation caps on bottom side, which you put if you have oscillation problems.
 
Last edited: