IRS2092 with IRFB4227 problem!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, one think more.. I saw in AN-1138 document that if I connect NPN transistor bettwen pin 5 and pin 6, base of that transitor is SD pin for external shud down.

Can I conect on example 4N25 opto coupler??? Because my protection is with NE555 with out +12V ref to Ucc-. With that voltage and resistor, when protect is on I can drive led in optocoupler and colector emiter will short pin 5 and 6

What do you think?

Thanks
 
not any bjt will work, it has to be fast, got some good gain, and high peak current... And I don't think you will be able to find any like that... if in Serbia is like here (I know this sounds weird), all you will find is 20y old components, which are just not good for this type of use

yes you can use opto for shutdown, when I had overcurrent, I had my 555 on negative rail just like IR, so I didn't need it... and you can check schematic on my page to see how it was connected
 
Thank you for replay... Nigel tell me that irs can work with one pair 4227. I must explore on which conditional..
But 200KHz maximum I know.. I dont know its to bad 200KHz for d class amp ?


No this amp will be on +-70V and 2 ohms load (full bridge at 4 ohms for 18" subwoofers) :)

I know that is maximum for one pair IRFB4227, but in the future maybe wanna to build stronger and I will need buffer for two pairs.

I have 4KW toroid transformer, one 2x50VAC and one on 2x60VAC, i wanna use it :D

Maybe I'll come back to IR2110, that chip on +-70 and 4227 @ 2 ohms load works good, but i got a feeling that IRS has better sound quality.. :), and better amp..

My protection is very similar with yours, for power supplay use +12V ref to ucc- and on pin 3 of 555 is voltage like that. With that volatge I only can use opto for shorting VSS and CSD (emiter on VSS colector on CSD)
 
Regarding internal drivers of the IRS we have to estimate that the internal resistance is around 10 Ohm. In combination with a 10 Ohm gate drive resistor it will result in a pretty slow 20 Ohms gate drive.
Nevertheless, if you find acceptable results you may stay with it.

If you want to chose the bipolar drivers...
Luka is basically right that not every transistor will work, but I do not think that generally fancy Zetex types are necessary.
As long as you target gate drive resistors of 5-10 ohms, then the good old BC368/BC369 or BC807/BC817 should work acceptable. Hope they are available in your place.
With those it should be possible to reach approx factor 2-3 of driving authority vs. direct drive from the IRS.

If BC368/369, BC807/BC817 are not available you may generally watch out for small signal types with 1A rating, or higher.
All the 100mA-500mA types are not likely to make sense, they would suffer from heavy gain drop at high currents and also have inherent high resistances in the emitter path.
 
105nS set up

R1 = 8.2K
R2 = 3.3K

Is that right? Or other values?
 

Attachments

  • sema 1.png
    sema 1.png
    32 KB · Views: 901
Looking for NPN/PNP high current buffers to drive heavy gates from IR chips?
ZETEX makes many bipolar transistors with excellent pulsed high-current performance.

Conventional small signal bipolars run out of gain at high currents, and can get unacceptably hotter because more losses happen in the transistor and less in the gate resistor, where they should be.
 
Looking for NPN/PNP high current buffers to drive heavy gates from IR chips?
ZETEX makes many bipolar transistors with excellent pulsed high-current performance.

Conventional small signal bipolars run out of gain at high currents, and can get unacceptably hotter because more losses happen in the transistor and less in the gate resistor, where they should be.

No doubt, I also love the Zetex types, but not everybody lives in heaven like we do.
Used BC807 and 817 some years ago and found them surprisingly better than one would expect from their 500mA...800mA ratings.
Of course they cannot drive like this:
IRFB4115 at 41A/500kHz/hard switching, driven by FZT 853 and 953 + 1R2.
http://www.diyaudio.com/forums/class-d/190100-new-mosfets-block-3.html
 
Yes, of course, ground plane is top layer...

Snubers are on board.. 1n and 10R

For shure, I decide 200Khz swich freq. 10r gate resistor and 105nS dead time.. 15x Gain
+-70V per board on 4 ohms load, with bridge adapter (full bridge on 4 ohms)

IRFB4227 output transistors

I dont know for what is cap 150pF to gnd between 47K and 1K in feedback?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.