Pafi: No, it will improve with load because once the body diodes start conducting the miller stuff changes completely, it stops stealing so much charge from the gate.
Also, note that it's a phase-shift self oscillating design, high output current will only happen when the output goes close to the rails and switching frequency should be expected to drop a lot by that time, resulting in far more time for the gates to charge.
No doubt the design is very precarious, like in most subwoofer car class D amps. We can do more power with less MOSFET.
Also, note that it's a phase-shift self oscillating design, high output current will only happen when the output goes close to the rails and switching frequency should be expected to drop a lot by that time, resulting in far more time for the gates to charge.
No doubt the design is very precarious, like in most subwoofer car class D amps. We can do more power with less MOSFET.
How is this possible?
But it doesn't really matter, because turn on time counts only when the current direction is normal (form drain to source), so body diode doesn't conduct.
If you mean high current is more then 60 A, then yes, but 30...50 A can flow at much lower voltage too. And the task is not only turning FET on at some time, but in a relatively insignificant time.
The miracle is: it works. I would like to know what the efficiency could be!
i replaced the irfp4229 with a pair of irfp264 i had and something must of happend cause i blew a KTA1837 ... maybe i should rebuild the driver board using new transistors ... i found that resistors that should have the same value are quite different so might have to deal with those to ...
it is preatty similar to this SONY XM-D1000P5 or am i wrong ? http://images.elektroda.net/78_1233429744.png
Ps: mgm2000ro measured the efficiency at 73% ...
it is preatty similar to this SONY XM-D1000P5 or am i wrong ? http://images.elektroda.net/78_1233429744.png
Ps: mgm2000ro measured the efficiency at 73% ...
Last edited:
VPPD is about 16V measured from vpp to vppd
Self-oscillating variable frequency closed loop Class D amplifier i think there is a patent on this schematic ...
Self-oscillating variable frequency closed loop Class D amplifier i think there is a patent on this schematic ...
There are probably one thousand patents on the same concept 
Concerning resistors, be aware that in many Chinese amplifiers and inverters they tend to be quite bad quality and resistance can change a lot with aging under high temperature or high voltage conditions. To make it worse, voltage ratings for each size are routinely exceeded on these products.
Concerning resistors, be aware that in many Chinese amplifiers and inverters they tend to be quite bad quality and resistance can change a lot with aging under high temperature or high voltage conditions. To make it worse, voltage ratings for each size are routinely exceeded on these products.
Last edited:
Check for resistors with actual value outside tolerance range first.
I would also check DC current gains of Q110/Q111, Q114/Q116, Q118 and the low side NPN drivers (all the ones that run hot) with multimeter. The pairs should not measure very different.
Last edited:
Same here, but happened in the LTP stage of a class AB amp. It took me forever to find out why the right channel was always a bit lower in volume and more distorted than the left.
And most people say those transistor gain testers on cheap multimeters were never of any use...
Yes, this is the standard solution. But the question was: is it desirable to provide clean square wave in this circuit? I don't know the answer, because there are possible problems.
Have you find the wrong parts yet?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.