Problem Osc. Frequency - IRS2092 and IRFB4227

[luka]

how how will IRS get with that settings? and I also think 45ns will be too small... I don't think you can use that with those fets

 

[SNY]

Change 15V Z Diode to 12V.
Dead time : 45 ns.
Osc freq: max! 400 kHz!

Working fine.. (sorry my english)

 

[nikigada]

Hello, IRFB4620 use transistors with dead-time 45 ns, carrier frequency 500 kHz. Zener diode, 15 volts is replaced with a 12 volt load resistor and 1.5k. Power + - 85 voluntary idling. Output power 500 watts at 4 ohms. There was, however, the following problem D5 and D6 diodes fail - drilled. I remove them or replace them with more powerful. I see that there are schemes in these diodes and circuits which are not represented. Excuse me for bad English, I use translator.

Change 15V Z Diode to 12V.
Dead time : 45 ns.
Osc freq: max! 400 kHz!

mission Impossible
Osc freq: max! 150-200 kHz! - achievable
Regards!

 

[SNY]

Csahnge OCset settings:

IRFB4620 @ 30A (60mOhm) @ 25 C

High Side: Udrain+Uvf (D4) = 1.8V(60mOhm@30A) + 0.6V = 2.4V.
Trip level= 1.2V ---> R18=R19
R18= 10k
R19= 10k

Low Side:
trip level: (Idrain*RDSon) = 1.8V(60mOhm@30A)
R13 = 10k
R12 = 5.6k

Z1=Z102 = 12V

R24=R25=15 Ohm

R8! = 130k @ 85V DC

 

[SNY]

...and place 1uF (or more) PP caps to +/- UT, close to mosfet's.

 

[nigelwright7557]

Do you have the schematics on how 2092 drive 4227 without any buffer?

What's advantage of having no buffer and with buffer?

Just drive into the 4227 gates with 10r resistor in series with 2092 outputs.

Adding a buffer means you can drive more mosfets and get more power.

 

[lithoc]

Just drive into the 4227 gates with 10r resistor in series with 2092 outputs.

Adding a buffer means you can drive more mosfets and get more power.

Isn't 2092 not enough to drive 4227?

By doing that, is that stable enough to drive 4 ohm speaker and better than 4020?

I see no advantage here if using 8ohm load as this can achieve by 4020.

It only make sense with 4 ohm with >80V which means 500W++

If this build to run full bridge, 1kw into 8ohm is possible.

A bit on the dangerous side tho...

 

[lithoc]

I found this IRFB5620PBF to be better than IRFB4620

 

[nigelwright7557]

Isn't 2092 not enough to drive 4227?

By doing that, is that stable enough to drive 4 ohm speaker and better than 4020?

I see no advantage here if using 8ohm load as this can achieve by 4020.

It only make sense with 4 ohm with >80V which means 500W++

If this build to run full bridge, 1kw into 8ohm is possible.

A bit on the dangerous side tho...

With one pair of 4227 I can get 500 watts.

With a buffer and two pairs I can get 1000watts.

If you dont need 500watts then a lesser mosfet will do.

 

[nigelwright7557]

Isn't 2092 not enough to drive 4227?

The 2092 will drive one pair of 4227's providing the deadtime is 105nS and you dont go above 250KHz.

I also found I had to reduce over current senstivity by a factor of 2. I just used 4k7 resistors in pairs on the CSH and OCSET inputs.

An externally hosted image should be here but it was not working when we last tested it.