hi,
this is a basic way to use ir2092.
No need to use high side external zener.
Double side pcb , using the other side as shielding would improve your Project and some smd 0.1 ceramic at switches from +B to _B as well.
this is a basic way to use ir2092.
No need to use high side external zener.
Double side pcb , using the other side as shielding would improve your Project and some smd 0.1 ceramic at switches from +B to _B as well.
Good idea but high frequency device can work properly on this PCB? Routes are as short as possible, but I'm only hobbyist...
This inductor will be ok? DTMSS-16/0.022/8.0-V FERYSTER - Inductor: wire | Transfer Multisort Elektronik
This inductor will be ok? DTMSS-16/0.022/8.0-V FERYSTER - Inductor: wire | Transfer Multisort Elektronik
I cant see much decoupling.
The 2092 and the mosfets need decoupling closely.
The layout needs to be as tight as possible.
The 2092 and the mosfets need decoupling closely.
The layout needs to be as tight as possible.
Better now , but could be better if you move the coil and out cap to the place of electrolitics decouplers and place them nearer the MOSFETS.Try to make vs conection thicker .Keep in mind you have to place 0.1uF as close the MOSFETS you can.Place all the circuitry from pin 1 to 6 nearer 2092 to avoid pick emi .
The most of parts arrived, if everything is ok, I will put it all together on this board and turn on...
Done, amp was playing, but when I had disconnected speaker, everything on PCB was hot and in a moment IRS and fet-s burned.
Unfortunately there wasn't C8A and heatsink, I have connected additional parallel resistors to R117A, R118A, R17A, because amp was clipping, when power supply was circa 2x35V, is possible that leads of that resistors was to long?
(img from IRAUDAMP7D: 500Wx1@ 8? / 250Wx2@ 4?, ±45 ~ 60V | DigiKey )
Unfortunately there wasn't C8A and heatsink, I have connected additional parallel resistors to R117A, R118A, R17A, because amp was clipping, when power supply was circa 2x35V, is possible that leads of that resistors was to long?
An externally hosted image should be here but it was not working when we last tested it.
(img from IRAUDAMP7D: 500Wx1@ 8? / 250Wx2@ 4?, ±45 ~ 60V | DigiKey )
Last edited:
The 2092 doesn't like not having a speaker connected.
The output will have lots of high voltage ringing from the output filter choke and will probably fry the output filter capacitor.
The output will have lots of high voltage ringing from the output filter choke and will probably fry the output filter capacitor.
This parts are already installed, but diodes aren't fast - 1N4007.
Sad fact that amplifier powered with no load occurs often, a few years ago I was using one channel in stock PA digital amplifier and everything was OK...
Sad fact that amplifier powered with no load occurs often, a few years ago I was using one channel in stock PA digital amplifier and everything was OK...
I have seen the output sit at +17VDC if there is no speaker connected.
And sometimes it runs OK.
I sold an amplifier and the buyer had tested the output before connecting a speaker and complained about the output sitting at 17VDC. He was worried it would damage his speaker. I had to tell him that the output can do strange things if no speaker is connected.
And sometimes it runs OK.
I sold an amplifier and the buyer had tested the output before connecting a speaker and complained about the output sitting at 17VDC. He was worried it would damage his speaker. I had to tell him that the output can do strange things if no speaker is connected.
You could try a resistor.
I have never used one so don't know what value it would need.
I have seen a similar problem with SMPS where they need a resistor across the output to keep them stable.
That's what I've heard too... This kind of equipment appears to require a certain load to assure stability. can't confirm it, because I have no experience with digital amps.
- Status
- Not open for further replies.