Hi Baldin,
For such fets, I don't think you would even need external driver to IRS. How fast do you want to switch such fets? Another question would be, how much of additional delay can you get ways with. It seems that 10ns is pretty much minimum here.
Thanks for the reply.
I'm using 4 and 3ohm (occasionally 2ohm) loads.
VCC is derived from the negative rail via a zener diode and a bjt, just like in IRAUDAMP7S.
Perhaps it is for this reason that VCC tends to drop so much (in fact, the bjt / diode zener warm enough but not in a worrying way).
As for the low-volume distortion, I tried everything, but I could not eliminate it.
I thought it was a difficulty of BD139 / BD140 to drive them, but I'm not sure, they are declared with 1.5A of capacity and I think that for a single pair of IRFB4227 they should be sufficient; then, if there are other factors to consider, I have not understood it yet.
Might be worth considering reducing the value of the series resistor "in front" of the zener/BJT combo?
The thing with the BD139/140 is that, being through-hole parts, the minimum leg length is still a good few mm, and at high frequencies (and high rise/fall times), every bit of conductor length (and as such, inductance) matters.
The thing with the BD139/140 is that, being through-hole parts, the minimum leg length is still a good few mm, and at high frequencies (and high rise/fall times), every bit of conductor length (and as such, inductance) matters.
News, AMAZING!
Reading the application notes, I understand how to set the threshold for overcurrent protection, both for the low side mosfet and for the high side mosfet.
I realized with amazement that on the low side (ocset) was already set well (with only one pair of IRFB4227, the maximum current that can be reached is about 45 Ampere, so you have to create a tension of 1.08v on OCSET and it was already ) but on the high side (csh) an anomaly had occurred (I had almost 4 volts) even if the voltage divider R2-R3 was correct;
Incredible but true, it was the blocking diode that was not suitable, in fact I had used a 1n4004 that had a Vf too high and was very slow; I replaced it with a diode wall that is very close to the characteristics required with a Vf of 0.7v.
In one shot I solved the problem of low-volume distortion and even the self-resetting of irs2092.
It remains to create a worthy power circuit, because with what I have now, I can get about 150w rms at 50hz without distortion at 4ohm (28.2v - 5.01A) then you start to see a lot of distortion because the rails are disappointing.
I think I will make a power supply that can produce at least +/- 78V with many Amps.
Reading the application notes, I understand how to set the threshold for overcurrent protection, both for the low side mosfet and for the high side mosfet.
I realized with amazement that on the low side (ocset) was already set well (with only one pair of IRFB4227, the maximum current that can be reached is about 45 Ampere, so you have to create a tension of 1.08v on OCSET and it was already ) but on the high side (csh) an anomaly had occurred (I had almost 4 volts) even if the voltage divider R2-R3 was correct;
Incredible but true, it was the blocking diode that was not suitable, in fact I had used a 1n4004 that had a Vf too high and was very slow; I replaced it with a diode wall that is very close to the characteristics required with a Vf of 0.7v.
In one shot I solved the problem of low-volume distortion and even the self-resetting of irs2092.
It remains to create a worthy power circuit, because with what I have now, I can get about 150w rms at 50hz without distortion at 4ohm (28.2v - 5.01A) then you start to see a lot of distortion because the rails are disappointing.
I think I will make a power supply that can produce at least +/- 78V with many Amps.
Only the DIP-packaged version seems to indeed be discontinued; the IRS2092S (the SOIC version) is still "active and preferred".
Audio' 'Driver' 'ICs - Infineon Technologies
Audio' 'Driver' 'ICs - Infineon Technologies
If you look at the 2092 datasheet it gives a diode to be used in that position.
I have them now and thank you. should I post them here? they have not been 'field tested' yet. -hjp
if you use more than 65v it's stopping ,,thats true
I use 6.8ohms gate resistor (IRFs4227 has 3.8 ohms internal gate resistor)
----total 10ohms
2.set overcurrunt setting to more than 100A
3.63v is good and safe for long time uses,,,,,
use 2ohms speakers (Irfs4227 drive its easly) for 400W
4.output limiting--most important thing if you use 63v power supply you have to set maximum output voltage to 48v peak ....safe and no heating
issue
5.I use only 40V peak (soft clipping)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.