The circuit in post 3 is not used, never built it.
Why the amp dilivered less voltage into 2 ohms is because the power transformer is only 500VA and was overloaded, hence the reduced voltage.
However i had something strange happen yesterday, after the above power tests i tried the amp on fullrange, and in no longer than a minute, the speakers let out a loud bang and my apartment went dark.
A quick check of each output transistor revealed a shorted MJW21195 which reads dead short between C and E and 45 ohms between B and C/E in all directions, i have no idea how this happened, i was well within SOA as i had previously checked this by hooking up five transistors in parallel in a simulation program and sweeping the Vce from supply voltage to fully saturated, at any Vce, the current per device was atleast 1.5-2A within SOA which should be more than sufficent to keep the amp alive.
The only thing i can think of is that the failed transistor was faulty from the gitgo and the power testes sent it over the edge, though it does show signs thats its gotten hot at some point as the texture of the silpad insulation pad is embossed into the back of it.
Why the amp dilivered less voltage into 2 ohms is because the power transformer is only 500VA and was overloaded, hence the reduced voltage.
However i had something strange happen yesterday, after the above power tests i tried the amp on fullrange, and in no longer than a minute, the speakers let out a loud bang and my apartment went dark.
A quick check of each output transistor revealed a shorted MJW21195 which reads dead short between C and E and 45 ohms between B and C/E in all directions, i have no idea how this happened, i was well within SOA as i had previously checked this by hooking up five transistors in parallel in a simulation program and sweeping the Vce from supply voltage to fully saturated, at any Vce, the current per device was atleast 1.5-2A within SOA which should be more than sufficent to keep the amp alive.
The only thing i can think of is that the failed transistor was faulty from the gitgo and the power testes sent it over the edge, though it does show signs thats its gotten hot at some point as the texture of the silpad insulation pad is embossed into the back of it.
The actual amp uses a board from an earlier project with the same circuit as this: http://i.imgur.com/9UU4a.png
Except the drivers are MJE1503x rather than MJE3xx and the output transistors are five pairs of MJW21195/96.
Pics: http://i.imgur.com/3dUwW.jpg and http://i.imgur.com/GeMpk.jpg
Except the drivers are MJE1503x rather than MJE3xx and the output transistors are five pairs of MJW21195/96.
Pics: http://i.imgur.com/3dUwW.jpg and http://i.imgur.com/GeMpk.jpg
Maybe a faulty transistor, or the 0.1ohms emitter resistors are to small, to make the current equal between the parallelled output devices.
Sajti
I'll use 0.22 ohms next time, for now, i'll play with N channel power stage using some STE47N50 giant mosfets.
Just testing around.
Now while messing around with a NMOS amp, i also discovered that the driver transistor for the high side drive was also shorted.
Now while messing around with a NMOS amp, i also discovered that the driver transistor for the high side drive was also shorted.
- Status
- Not open for further replies.