Issues with mosfet heat

More
3 months 3 weeks ago #13051 by vladca
Replied by vladca on topic Issues with mosfet heat
Hi, as I promised here are the results of my tests.
It's nothing exact and I didn't take notes but at least something for orientation.

I tried several combinations of MOSFETs and coils. But even after these tests, I do not know
which parameters are important. I'm not at home on switching power supplies.
In any case, here are the results.

Tested MOSFETs and coils:
FDPF51N25, FQP32N20, IRF640, IRF740, IRFP360, SiHP25N40D, STF24N60M6, STP24N60DM2

1. 82uH / 240mOhm / 1.0A (Coilcraft RFC0807B-823KE)
2. 100uH / 220mOhm / 1.2A (Coilcraft DR0810-104L)
3. 100uH / 180mOhm / 1.7A (Ferrocore COIL0810-0.1)
4. 120uH / 250mOhm / 2.0A (Ferrocore COIL0812-0.12)
5. 270uH / 500mOhm / 1.5A (Ferrocore COIL0812-0.27)
6. 330uH / 590mOhm / 920mA (Ferrocore COIL0812-0.33)
7. 470uH / 700mOhm / ??? (unknown)

Used HV caps 2u2/450V (Panasonic NHG) and 3u3/400V (Leaguer REB), diode UF4007.
PCB not built-in, only on the holder, ambient temperature 18 degrees.
MOSFET temperature measured by PT100 contact probe.

I'll start from the end:
Vishay IRF740 ... completely unsatisfactory, high temperature, even above 100 degrees, no matter what coil, 12V source necessary, slow start HV, but the clock is going well (but who knows how long B) )
Vishay SiHP25N40D ... not working !, I couldn't get HV more than 50V (from 12V source) with even one coil, I don't know where the problem is
Onsemi FDPF51N25 ... does not work quite well, heats up quickly, scope showed strange course at GATE (maybe too low Rdson?), HV after calibration only 140-150V, does not work at 9V, manual increase duty cycle required
Onsemi FQP32N20 ... works any-so, scope showed significant overshoot on GATE mainly with coil 5-7 (maybe too low Rdson?), the rest as FDPF51N25 above
STMicro STF24N60M6 ... only works well with coil 3-6, it heats up, but it goes quite well
STMicro STP24N60DM2 ... works well, source 9V and 12V, best coil 330uH, temperature up to 36 degrees (at 9V)
Unknown IRFP360 ... works very well with any coil, HV stable, temperature up to 34 degrees but TO247 case !!!

Infineon IRF640 ... best choice, no problem, HV stable and fast starting, coil 1-4, temperature below 30 degrees, clear winner :)

In addition, I have a problem with MPSA42, the NEONs are not blinking, they are just lit continuously.
Change to 2N5551 and everything goes fine. But he has only 160V Uce :huh:
My MPSA is called CTMPSA42, I think it's CDIL. Hands away from them. I'm waiting for pieces from NTE and DIOTEC. I wonder if they will work better.

That's all friends ... maybe it will help someone :).
The following user(s) said Thank You: Ty_Eeberfest

Please Log in or Create an account to join the conversation.

More
3 months 3 weeks ago #13052 by Ian
Replied by Ian on topic Issues with mosfet heat
I'd go for the 150uH. Often the coils are large tolerance, and I have found that 150uH often hit the sweet spot, as the tend to be -20% rather than +20%.

But 32 degrees is not bad. The other change I have been playing with is to have a sort of level shifter on the MOSFET drive. Often the 5V on the output pf the controller is not enough to fully turn on the MOSFET. I didn't come to a conclusion on that yet though.

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.082 seconds
Go to top
JSN Boot template designed by JoomlaShine.com