Greetings Maestros,
here i am building amp based on iraudamp9 with 3 pairs of output, with the intention of driving IRFB4227 OR IRFP240. old project that i failed to run before but now i seem to know part of the problem, much thanks to Jan Lester for his advice on this amp. it would not be possible without him.
here is the spec:
similar to IRAUDAMP9 notes with some extras, and i play around with its OCSET and CSH
in the hope of not tripping.
on CSH pin: 7k6 to diode and 2k2 to Vs
on OCSET PIN: 3k3 to vref and 7k6 to -supply
on DT: tied to -supply through 2k2
on Vb: bootstrap cap 47u to Vs
on VCC: reserve cap 47u to -supply
currently its driving 3pairs of IRFB4227, it was heavy to drive, so i took external switching supply to supply the mosfets bias, then it stops glitching whenever i try to crank up the freq. seems like the bias current isnt enough as Jan has mentioned. (because my LM regulator is floating in the air without heatsink, i didnt notice it was overheating).
now i could achieve 360khz (maximum trimpot turn: because my trimpot is only 500ohm), but then the
IC gets to almost about 60 degree celcius
while the totem-pole 2n5551/2n5401 gets to almost 70 degree celcius.
again, upon consulting with Jan, it seems that 2n5551/2n5401 is not suitable for the drivers, in term of wattage and that it needs to have low saturation voltages. if anybody could elaborate on whats happening here, i would be glad.
i think IRS2092 has a hard time driving this totem-pole load. (again this is what i feel, if anybody has experience and knowledge on this, feel free to let me know), or perhaps there are other factors causing this irs2092 IC and totem-pole driver to get this hot,
my target is to achieve no more than 50 degree Celsius on both IC and totem-pole driver hopefully.
and hopefully it this could crank up 1k7watt with 90volt 2 ohm without tripping or overheating anything. Amen.
here i am building amp based on iraudamp9 with 3 pairs of output, with the intention of driving IRFB4227 OR IRFP240. old project that i failed to run before but now i seem to know part of the problem, much thanks to Jan Lester for his advice on this amp. it would not be possible without him.
here is the spec:
similar to IRAUDAMP9 notes with some extras, and i play around with its OCSET and CSH
in the hope of not tripping.
on CSH pin: 7k6 to diode and 2k2 to Vs
on OCSET PIN: 3k3 to vref and 7k6 to -supply
on DT: tied to -supply through 2k2
on Vb: bootstrap cap 47u to Vs
on VCC: reserve cap 47u to -supply
currently its driving 3pairs of IRFB4227, it was heavy to drive, so i took external switching supply to supply the mosfets bias, then it stops glitching whenever i try to crank up the freq. seems like the bias current isnt enough as Jan has mentioned. (because my LM regulator is floating in the air without heatsink, i didnt notice it was overheating).
now i could achieve 360khz (maximum trimpot turn: because my trimpot is only 500ohm), but then the
IC gets to almost about 60 degree celcius
while the totem-pole 2n5551/2n5401 gets to almost 70 degree celcius.
again, upon consulting with Jan, it seems that 2n5551/2n5401 is not suitable for the drivers, in term of wattage and that it needs to have low saturation voltages. if anybody could elaborate on whats happening here, i would be glad.
i think IRS2092 has a hard time driving this totem-pole load. (again this is what i feel, if anybody has experience and knowledge on this, feel free to let me know), or perhaps there are other factors causing this irs2092 IC and totem-pole driver to get this hot,
my target is to achieve no more than 50 degree Celsius on both IC and totem-pole driver hopefully.
and hopefully it this could crank up 1k7watt with 90volt 2 ohm without tripping or overheating anything. Amen.
Attachments
I drove 2 pairs of 4227 with BD244/BD243 and they barely got warm so should be good for driving 3 pairs.
The 2092 is easily upset by lots of things.
The VCC has to be rock solid. Decoupling must be close to mosfets and 2092.
The 2092 has a poor gate drive. I found the best I could get it to drive without external gate driver was a pair of 4019 mosfets.
I got my VCC by 3 dropper resistors and a 12 volt regulator.
I also had problem with 2092 and large power supply smoothing capacitors.
On power down I got a siren type wailing ending in a huge thump through the speaker.
I used t106-2 core for output inductor and it barely got warm.
The 2092 is easily upset by lots of things.
The VCC has to be rock solid. Decoupling must be close to mosfets and 2092.
The 2092 has a poor gate drive. I found the best I could get it to drive without external gate driver was a pair of 4019 mosfets.
I got my VCC by 3 dropper resistors and a 12 volt regulator.
I also had problem with 2092 and large power supply smoothing capacitors.
On power down I got a siren type wailing ending in a huge thump through the speaker.
I used t106-2 core for output inductor and it barely got warm.
hei Nigel, just swap the totem-pole with bd139/bd140 based on Jan's Recommendation. the IC runs about degree cooler, bouncing at about 50 degree and the the totem is about the same.
with this TR, i could achive highes osc with the same osc setting.
when i push to 450khz, one of the IC pair get to 110 degree C on idle. what is happening here?
also when you say VCC must be solid, u mean to put 47uf closed to the IC. correct?
and what do you mean by decoupling closed to the mosfet?
like 100n across Vb and Vs pin?
Attachments
or you can use ZXGD drivers! like in my commercial design
Driving the mosfets is hard work for the 2092 so it will get hot at higher frequencies and low gate resistors.
I found I had to decouple the gate driver transistors with 10uf close by.
Yes I put 47uF across VCC to keep it stable neat 2092.
So you going to scan through every project on DIYAUDIO 5/10years back ask for a status update for cloning reasons dont you think designing something yourself is a better use of your time?