Sorry for the late reply everyone, our electronics store didn't have the parts in stock. 🙄
Anyway i did some modifying and made some measurements with my scope. Unfortunately, i don't have any pictures, but oscillations were reduced considerably with C8 and C9 upped to 120pf and 470pf.
The output transistors (2SD718/2SD688) were first replaced with a pair of MJL4281A/MJL4302A. Completely overkill being three times as fast but it was laying around at the time.
I experimented a little further and changed to a pair of 2SA1943/2SC5200, for some reason the oscillations were less. To me is weird because these transistors have comparable specs to the MJW's, perhaps Japanese transistors like each other better? 😕
As a last test i used a pair of 2SB817/2SD1047. These are just a little faster than the original transistors but i didn't get any oscillations. I changed C8 and C9 back to their original value, still no oscillations.
They seem to be just perfect, not too fast but still a little better than the originals.
The sound had noticeably improved, but then again the old ones were worn out anyways. 😉
It's time to measure distortion, i would like to do this through my soundcard, but i'll need to convert the amplified signal.
Will a voltage divider be sufficient to achieve this?
Anyway i did some modifying and made some measurements with my scope. Unfortunately, i don't have any pictures, but oscillations were reduced considerably with C8 and C9 upped to 120pf and 470pf.
The output transistors (2SD718/2SD688) were first replaced with a pair of MJL4281A/MJL4302A. Completely overkill being three times as fast but it was laying around at the time.
I experimented a little further and changed to a pair of 2SA1943/2SC5200, for some reason the oscillations were less. To me is weird because these transistors have comparable specs to the MJW's, perhaps Japanese transistors like each other better? 😕
As a last test i used a pair of 2SB817/2SD1047. These are just a little faster than the original transistors but i didn't get any oscillations. I changed C8 and C9 back to their original value, still no oscillations.
They seem to be just perfect, not too fast but still a little better than the originals.
The sound had noticeably improved, but then again the old ones were worn out anyways. 😉
It's time to measure distortion, i would like to do this through my soundcard, but i'll need to convert the amplified signal.
Will a voltage divider be sufficient to achieve this?
What are the wear-out mechanisms in a transistor?jooch said:
BTW changing the quiescent current can reduce distortion at some power levels and increase it at others, so don't assume the designer got it wrong.
Temperature/load and time will determine wear. Each time a transistor will switch it will wear too, hence the factor time. Every electronic component has a certain lifetime or MTBF.
You're right about the quiescent current, beefing it up will reduce output power, but who listens to 50 watts anyway? 15watts (8ohms) is sufficient for me.
- Status
- Not open for further replies.