And the next updateSorry for the time this is costing me, but I really want to get to the bottom of this. One of the LED's (L111) of the CVS (positive supply) was removed and both LED's checked, the removed LED had a forward voltage of more then 2Volts (that is suspect) and it was replaced, after this still some oscillation. Q103, Q104 and Q102 where removed and replaced by checked D44H11, BC337-40 and BC337-30, after this the oscillation where the same. A 2n2 compensation capacitor on Q104 was installed, this halved the oscillation's. C101 was removed and checked, it measured 8uF and 2ohm(esr). C101 was not replaced all oscillation's are gone.
This channel is done, later this week I do the negative supply on this PSU if the same treatment gives the same result then we know more, fingers x'ed.
And a final update, all 4 PSU's function as should be, most transistors where replaced, compensation caps where changed from foil types to ceramics. It is very hard to say what (exactly) fixed the problem(s) it did oscillate when it arrived, and it does not now. Four electrolytic capacitors where removed and not replaced, I do have a strong impression that they where part of the problem, I can not reproduce the problem due to cutting the capacitors (removal by destruction?
) out. The removed capacitors are C101/201 this should have very little impact on the performance of the PSU.
And a final update, all 4 PSU's function as should be, most transistors where replaced, compensation caps where changed from foil types to ceramics. It is very hard to say what (exactly) fixed the problem(s) it did oscillate when it arrived, and it does not now. Four electrolytic capacitors where removed and not replaced, I do have a strong impression that they where part of the problem, I can not reproduce the problem due to cutting the capacitors (removal by destruction?
P.s. Some of the LED's where found defective and replaced. My replacement types are almost 35 year old NOS devices, they produce almost no light at the applied current but the voltage levels are the same as for modern one's. The broken LED's may have set up the PSU in a weird/destructive (non expected) mode, this also will have been a factor in the problem. The LED's can be broken by having a CVS that fails (1st power stage) and then shorting the shunt for constant current testing.
Note: It is important to test every LED before mounting!
Last edited:
Frans, not sure if you are still trying to fix an oscillating PSU, but if you still do I have an idea and you might want to give it a try. I recently read in another thread that a user removed oscillations from a circuit by grounding a heat sink used by multiple devices. I have oscillation problems in my "pumpkin" pre-amp since many months up to the point where I can actually hear the chirping sound in my speakers once in a while. Compensation didn't really change the picture. Today I played a bit with a ground wire and the oscillations seem to disappear when I ground the heat sink. Haven't confirmed with the scope yet though.
Regards
Sven
Regards
Sven
Thanks, that's a good idea. And no, the PSU was fixed (as described a few messages back). This was the second one that I had that needed fixing, and Hesener has also done a few, in the 600 (or so) send out (I do not know how many where actually build (but more then a few
)) there have one been a few that oscillated, and the reasons are not always clear. In most cases compensating as recommended will help, a few suffer from bad components, and then there are the ones like this last one, still this one seems to have hade some component failures, as described I did swap most of then, and left out a few capacitors, it al functions (now) as expected. But I can not tell you what exactly fixed it The mystery goes on ...
Frans, what a fascinating crime story... that was a tough one... thanks for documenting your findings! it appears that as the components become more perfect, unexpected stuff happens...
@ abraxas : the heatsink on the paradise is grounded, if metal screws have been used. what most people did anyway.
@ abraxas : the heatsink on the paradise is grounded, if metal screws have been used. what most people did anyway.
I now had the time to become familiar enough with the Paradise using a Vishay S102K metal foil resistor as a load resistor. So yesterday I finally switched over to a non magnetic Vishay-Dale CMF-55-143, same value.
I didn't really expect to hear a difference, but now it seems I do - the CMF-55-143 sounds leaner, has less volume, overall a bit more contained. I prefer the S102K.
The fact that I hear a difference (or believe to hear a difference) frightens me because of the consequences. The load resistor is not even part of the signal processing, its just a shunt. There are probably dozens, may be hundreds of other resistors in the signal processing later on. Will changing them to S102K yield similar results or is the cartridge load a particularly important resistor?
Regards
Sven
After some help.
I built my paradise last year and have been using it on and off since. I recently powered it up to find the output of one board has a much reduced output and is noisy.
On checking the voltages, I find pin6 on the OPA134 has around 14v! I have tried reducing R43a/b to 47k and initially this seemed to correct the issue, but within a few hours it has returned and now pin6 is up to 16v.
All the voltages between my two boards check out, except for those around pin6 and those of Q50 and Q51 Initially these were bouncing around at about 1-2v different from from the good board ( I should say that the v's here on the good board are rock solid), but are now up at 15-16v too.
Any ideas as to the culprit or what/how I should be checking.
Many thanks
Val
I built my paradise last year and have been using it on and off since. I recently powered it up to find the output of one board has a much reduced output and is noisy.
On checking the voltages, I find pin6 on the OPA134 has around 14v! I have tried reducing R43a/b to 47k and initially this seemed to correct the issue, but within a few hours it has returned and now pin6 is up to 16v.
All the voltages between my two boards check out, except for those around pin6 and those of Q50 and Q51 Initially these were bouncing around at about 1-2v different from from the good board ( I should say that the v's here on the good board are rock solid), but are now up at 15-16v too.
Any ideas as to the culprit or what/how I should be checking.
Many thanks
Val
Last edited:
Ha Ha. Very nice packaging thank you my friend!
A token of appreciation left Canada this afternoon. It should reach Holland in a week or two. I hope you like, but if you don't, you may always sell it
Regards - Pierre
When I get a day off, I will connect the fixed regulators to the phono pre and test and adjust. Then finish the second pre-reg, and then hook everything up for a listen. I will keep you good folks posted on my (slow) progress
Know that I am looking forward to that day also, so go and 'Make it so'
