All input stages are classA, so dissipation is constant. The only thermal change (after some settling time) is from ambient temperature, equal for all transistors from input to output from VAS. If you will change temperature of one isolated component, this is unrealistic and useless, here is always thermal dependence of e.g Vbe, you cannot supress this. One must try to do this negligible in whole circuit, and to try to compensate one compenent with other.
Thermal changes and resulting bias changes in output stage (due to variable power dissipation) have to solve Vbe multiplier with proper thermal coupling to ouput transistors.
Thermal changes and resulting bias changes in output stage (due to variable power dissipation) have to solve Vbe multiplier with proper thermal coupling to ouput transistors.
have you seen this inputstage from a well known manufactor?
Look how the VAS stage(Q605 + Q609 positive) is coupled as a "CCS" kind of.
Look how the VAS stage(Q605 + Q609 positive) is coupled as a "CCS" kind of.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
have you seen this inputstage from a well known manufactor?
Look how the VAS stage(Q605 + Q609 positive) is coupled as a "CCS" kind of.
Wilson mirror (Q605+D609+Q609)
Rather post something like the pm11, the manufacturer has been using this topology for over 7 years and has it well worked out.
Why?
Just thought it could be the area where "mistake" has been made. Removing the CCS may be an option but in such a complex schematic I often prefer a better CCS so... At least I'm not suggesting changing the PCB
I tried firing up my Eyesee yesterday without the ops, only nd and pd shorted to the nfb, and I only get the bottom half of the sine wave. Does the ops have to be attached?
Terry what about Kypton-v ,did you have a listening test on this?
Last edited:
No. It was unstable. I feared hurting my ops so I put it aside until you and Pete got it resolved. So now I have Krypton-V, infidel and Eye see all built and none singing. These new designs are a bit tougher to get right.
Is the Kypton-v unstable to you?
Mine is stable so i can't understand why...?
Look Terry you must listen to this Kypton-v.
Try to debug it and do a listening test it is something absolutely necessary
Yes but I haven't pulled the super pair yet. Do you think this one sounds "different"?
Pull the super pair ,do a listening test and report again.I want your opinion about this IPS.
Is any Kypton-v SINGING out there?
Last edited:
I haven't tested yet. These are the changes I made. Please take a look and see if I missed anything.
Thanks, Terry
Attachments
You're right. I forgot the diode clamp. I don't see Csp1 and Csp2 on your board. I routed things a little differently than you did but I hope, if you look at the way I did it, it will follow the schematic. I'll add the diode clamp tomorrow before I test.
Thanks, Terry
