Hmmm,
Ok my amp is working fine now except when there is no load applied. The waveform shows signs of slight oscillation on the crest of the positive side only. Doubling the LTP current helps almost rid this oscillation but i cant fully clear it up. There are no signs of it on load and all test configurations are good. Clean squares, flat freq responce and no sensitivity to capacitance on the output.
The LTP current set resistor has now been altered to 47R the same as the VAS. This also allows me to reduce Cdom and seems to be the key to making the amp sound clearly natural. I'm a bit concerned about the high current though as the little constant current tranny is now at its limits for power dissipation and runs rarther hot.
Anybody any ideas on how to clean up the positive waveform without further current increases??
Would it help for me to take piccys?? My layout is rarther compact on veroboard, maybe this could be the cause??
Or am i being a bit too fussy as things are fine with a load on the output. Hmmm ,
What do you PPL think??
Regards
Leigh
Ok my amp is working fine now except when there is no load applied. The waveform shows signs of slight oscillation on the crest of the positive side only. Doubling the LTP current helps almost rid this oscillation but i cant fully clear it up. There are no signs of it on load and all test configurations are good. Clean squares, flat freq responce and no sensitivity to capacitance on the output.
The LTP current set resistor has now been altered to 47R the same as the VAS. This also allows me to reduce Cdom and seems to be the key to making the amp sound clearly natural. I'm a bit concerned about the high current though as the little constant current tranny is now at its limits for power dissipation and runs rarther hot.
Anybody any ideas on how to clean up the positive waveform without further current increases??
Would it help for me to take piccys?? My layout is rarther compact on veroboard, maybe this could be the cause??
Or am i being a bit too fussy as things are fine with a load on the output. Hmmm ,
What do you PPL think??
Regards
Leigh
hmm.. if everything else fails maybe there is a 5ohm resistor(or some other low value) to put in somewere.. not sure where but, gaah this is sounding waay to fuzzy but it helped me a year ago or so. maybe on emitter of q10 or collector of q3 or q5 ,q6 collectors.. also for me, quality of the capacitors have played a role.. thinking of c5,c9(schematic from first post)
empirical testing rulez :-D
ooh almost forgot..
Cheers and goodnight,, I'm going to hava a coke now and go to bed :-D (dont like beer very much)
empirical testing rulez :-D
ooh almost forgot..
Cheers and goodnight,, I'm going to hava a coke now and go to bed :-D (dont like beer very much)
ok.. I'll give an example.. if you hava a normal vas transistor and connect in series a cascode transistor(emitter) on the collector of vas and put base on a zener or whatever gives a few volts away from vas emitter , it might oscillate like MAD in real life and be stable as a yogurt in the simulator but adding 5ohm or so between them might make it stable as a bucket of yoghurt in real life too.. hmm..
Ahh,
I see, your talkin base stoppers here oui?
I've got them on the outputs but they are not employed anywere else. After some more investigation with the scope it appears that i can't fully drive up to the positive rail. I'm about 3 or 4 volts short. I think this will be part of the same problem and it all seems to be eminating from the differential stage not being able to properly drive the VAS. More current helps but i loose headroom. I think i may have to finally look at increasing the degen resistors to allow for more voltage swing although i wanted to avoid this as OLG drops so distortion increases.
Waa!! not fair! I want both stability and low THD but i must settle on a compromise!!
Leigh.P
I see, your talkin base stoppers here oui?
I've got them on the outputs but they are not employed anywere else. After some more investigation with the scope it appears that i can't fully drive up to the positive rail. I'm about 3 or 4 volts short. I think this will be part of the same problem and it all seems to be eminating from the differential stage not being able to properly drive the VAS. More current helps but i loose headroom. I think i may have to finally look at increasing the degen resistors to allow for more voltage swing although i wanted to avoid this as OLG drops so distortion increases.
Waa!! not fair! I want both stability and low THD but i must settle on a compromise!!
Leigh.P
Hi all,
Fiddling with this amp again. Thought all was finally o.k until i started to do some serious testing. I have a problem when the amp is run up to its maximum under a low load ( 4R ). A scope picture of the amps output says it all. This is a square wave at 1KHz at around 25VDC ( just as the amp starts to clip its rails ) into a 4R load. Has anybody seen this before???
( referring to the ledge on the positive rising edge of the square )
Fiddling with this amp again. Thought all was finally o.k until i started to do some serious testing. I have a problem when the amp is run up to its maximum under a low load ( 4R ). A scope picture of the amps output says it all. This is a square wave at 1KHz at around 25VDC ( just as the amp starts to clip its rails ) into a 4R load. Has anybody seen this before???
( referring to the ledge on the positive rising edge of the square )
Attachments
Here is the current circuit. P.S before you all pick on my tip drivers it's not them. I've tried swapping them for BD139/140 and the only difference is slightly more noise. The problem seems to be somthing to do with the VAS/LTP const current reg at the bottem. Probing around with a second scope probe reveals a sudden twitch in voltage in these stages when the ledge suddenly appears. All is well at all other levels untill rail clip. Any ideas??
Regards
Leigh
Regards
Leigh
Attachments
Hmmm,
When the amp is run on the test bench it can permenently grossly clip with no detectable change in temperature in the small signal stages and vas. I can see how current can peak quite high if the output is prevented from moving but on a 4R load no destructive type problems seem to occur. The only problems are with the kink when the amp clips on a square. The rails are being dragged down from about 30V to 25V at this power level. Do you think it is a problem with the vas sticking to or snaping up to the rails due to the vas saturating as mentioned by MikeB in the previous post???
Leigh
When the amp is run on the test bench it can permenently grossly clip with no detectable change in temperature in the small signal stages and vas. I can see how current can peak quite high if the output is prevented from moving but on a 4R load no destructive type problems seem to occur. The only problems are with the kink when the amp clips on a square. The rails are being dragged down from about 30V to 25V at this power level. Do you think it is a problem with the vas sticking to or snaping up to the rails due to the vas saturating as mentioned by MikeB in the previous post???
Leigh
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.