I'm still waiting for the parts but I had some probing at "good" channel and I can see this on 18V rails:
its 20MHz @12mV with main amplifier disconnected and it changes when I put my hand close to the probe lead , I have Werners caps fitted do I need to install other 2 ceramic caps
its 20MHz @12mV with main amplifier disconnected and it changes when I put my hand close to the probe lead , I have Werners caps fitted do I need to install other 2 ceramic caps
It seems I made mistake when checking power rails , somehow I uncalibrated time constant also got not greatest ground connection .
Now with amplifier and Calvin buffer connected , RIAA in and 10k on the input it shows around 3mV @ roughly 2.5MHz , same on the output but difficult to capture due to output voltage variation
Both rails 17.95 V ,
Vcasch= 14.40 V
Vcascl= -14.37 V
R82 0.568 V and -0.587 V
pin 6 on OPA627 varying between +0.2 and -0.2 V with open lead and goes up to 0.5 V when catches some draft .
all other voltages OK , input zeroed with 10k , Calvins zeroed
output bit wild but small offset up to 7mV with closed lid , with 10R output goes down to around 1mV after couple of minutes ,
Now with amplifier and Calvin buffer connected , RIAA in and 10k on the input it shows around 3mV @ roughly 2.5MHz , same on the output but difficult to capture due to output voltage variation
Both rails 17.95 V ,
Vcasch= 14.40 V
Vcascl= -14.37 V
R82 0.568 V and -0.587 V
pin 6 on OPA627 varying between +0.2 and -0.2 V with open lead and goes up to 0.5 V when catches some draft .
all other voltages OK , input zeroed with 10k , Calvins zeroed
output bit wild but small offset up to 7mV with closed lid , with 10R output goes down to around 1mV after couple of minutes ,
Last edited:
Seems like Werners mod don't cure oscillations entirely , in board I have reduces amplitude to minuscule level but still there , any idea what to do now with over 30 builds you must have had this problem at some point.
I have pulled input stage transistors and they're matched within couple hFe when left in ambient temperature .
Pete this can be very nagging indeed. I was looking for the schematic of this, couldn't find it, could you post, maybe I have some ideas.
But, and please don't laugh, did you connect the scope probe to its ground lead and both to your unit ground, make sure its gone then ... ?
Jan
But, and please don't laugh, did you connect the scope probe to its ground lead and both to your unit ground, make sure its gone then ... ?
Jan
Hi Jan
All diagrams are in hesner's signature , here is the link:
paradise links - Google Docs
It don't include latest Werners mods to cure oscillations , couple of 22uF capacitors at the input of pre-regulator and 470nF caps between collectors of Q103, Q203 and ground.
I'll double check scope lead later ,
Thanks
All diagrams are in hesner's signature , here is the link:
paradise links - Google Docs
It don't include latest Werners mods to cure oscillations , couple of 22uF capacitors at the input of pre-regulator and 470nF caps between collectors of Q103, Q203 and ground.
I'll double check scope lead later ,
Thanks
What run time do you have on your Paradise? They genuinely need 60-80 hours powered up from new for the lytics to fully form. Before this the leakage from the caps will be skewing what you see on the output.
My builds usually go from swinging 150mV when just built to sub 0.1mV after a few days of constant power on, with lid closed.
But as jan says check your probe grounding first.
My builds usually go from swinging 150mV when just built to sub 0.1mV after a few days of constant power on, with lid closed.
But as jan says check your probe grounding first.
Last edited by a moderator:
I must have misunderstood what Jan have said ,
probe ground was connected to the probe and amplifier ground so no missing ground
These boards were working in my friends system for at least 2 years so is no capacitors leakage
I can get output offset down to 1mV but oscillations stay almost at the same level with high or low voltage offset and originate at pre-regulator I guess
probe ground was connected to the probe and amplifier ground so no missing ground
These boards were working in my friends system for at least 2 years so is no capacitors leakage
I can get output offset down to 1mV but oscillations stay almost at the same level with high or low voltage offset and originate at pre-regulator I guess
.
When you have bit of time can you scope output which isn't very easy , I assume you have oscilloscope, or 18V rails , please .
Wow! There's enough parts in there to build a nuclear power plant ;-)
One unsuspected part in such schemes often is the ESR of the output capacitors. You really need something between half an ohm or more as ESR in C105/205 to damp the high gain feedback loop. Most cheap electrolytics automagically include that, but sometimes a very high quality electrolytic, or even a film cap, is used here and the loop is under-damped and oscillates at low level.
Not saying that is the case here but a quick check would be to put 1 ohm in series with this cap and see if that cures it.
Edit: you also test it by temporarily disconnecting C103 and/or C104, which lowers the loop gain and could also stop oscillations, if that was the cause. Especially C104 is dangerous - it un-does the positive effect of the base stopper R107 completely.
Jan
Last edited:
Thank you Jan ,
I think the cap in 105/205 position is a cheap electrolytic but I'll check your suggestion.
Probably I will start will pulling out 103/104 as I have not many parts around here , I'm going to Europe in a week , while there I'll buy some more .
Is getting bit late so I will try to check all this sometime during the week
Thank you
I think the cap in 105/205 position is a cheap electrolytic but I'll check your suggestion.
Probably I will start will pulling out 103/104 as I have not many parts around here , I'm going to Europe in a week , while there I'll buy some more .
Is getting bit late so I will try to check all this sometime during the week
Thank you
Few days in Ireland/Cork then Poland/Krakow and possibly Germany/Berlin(Wildau)
I managed to find different 10uF cap , left it running for an hour and best I can trigger is this :
I'm using 60MHz probe and 100MHz scope , would be beneficial to get better probe -100MHz?
I'm using 60MHz probe and 100MHz scope , would be beneficial to get better probe -100MHz?
This looks more like a measurement artifact to me. Some sweep trigger ringing. You probably see that too with everything shut off.
Jan
Jan