Transistor in VAS overheating
I have breadboarded the CF amp shown in the attached picture. I notice that Q10 gets very hot and it smells like it is melting. I have tried adjusting the idle current but that does not seem to help. I also tried replacing Q10 to no avail. When I tested this amp a couple of days ago I did not have this problem. Could hard testing have burned something out which is making Q10 overheat? Any other ideas? BTW, the rails are +/- 18v.
Edit: Also I measure almost 0.7v accross Re on Q10.
a million of drawbacks
-----i dont belive that amplifier like that will operate properly at this rail voltage
-----Q10 looks too small for this application use somthing like the MJE 340-350 dont remember which one is N or P
-----particular schematic is an sziklai that makes the output stage to have a gain on its own also means that amp like that will oscilate easier than any other efp out stage amp
----- amplifiers like that may not be constructed on veroboard since sziklai circuits will oscilate easy and on a vero board 10 times more
----- also i think tha generally something is wrong with this schematic ....something is wrong calculated in the VI limmiter ..... it seems to me tha someone took bits and parts from some amplifiers and created this
-----there is no miller caps in any of the stages of vas or drivers one idea will be to install 60-100 pf between B-C of vas amplifier and then the same on the B-C of every driver ( some like to put a big cap in the vas area 80-120 pf and use the same in drivers ) others like to put a very small one in the vas 10-30 pf and go for much biger in the drivers ...i have seen sziklai working with 680pf in the driver stage ( !!!! ) in my opinion a sonic killer
----If you ever manage to stabilize this amplifier will require higher rail voltage and better quality of output transistors ....
----if you manage to make it work it will play very nice i think !!!!
happy regards stay in touch
Re: a million of drawbacks
Why wouldn't it work properly at +/-18? I've built plenty of these little amps over the years and they work fine - better than chip amps, in fact. If you could ever get away with breadboarding a CFP, its in a low power circuit like this.
Q10 is just fine with 6mA and 18V. It will get a wee bit hot with 30ma (limited by Q12) which is what's happening. Is the output "stuck" at the positive rail? A fault somewhere could cause this, and if the error amp tries to compensate unsuccessfully, you'll get full conduction in Q10. This is the reason *why* you current limit this stage.
|All times are GMT. The time now is 07:18 PM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio