If there's a poll for the most generous company my vote goes for IR. Most companies provide only schematic for reference design, some even has no schematic unless you buy the boards. But IR provide not only the schematic of reference design of IRS2092 (IRAUDAMP7S) but also gerber files!
Working in a company that provide PCB fabrication and assembly service, I can't help making a board with the gerber files when I found it last month. The PCB was done on 10th of September, but it takes a few days to find the parts in the BOM. MOSFETs, transistors, capacitors, connectors and resistors are not problem at all, they are all on hands. It takes a few days to find the similar inductor for out put filter.
When all parts were assemblied, I connected the board to +-50V supply and the output to a 4Ohm speaker. But there's no music all when the power switch is turned on! I checked the boards carefully, for about half an hour. finally I found that the diode was not mounted well. Adding some solder at the pin of the diode, the music comes from the speaker when I turned on the power again! The sound is really excellent!
I remember another member of diyaudio said "I lost interest in TA2022 when I get into IRS2092". Now I understand why he said so
We made 2 different type with the same board: one with IRFI4019H and one with IRFI4020H. Both had excellent performance, the only difference is that the IRFI4020 has higher output power. Soon both version will be listed on our site for sale.
Working in a company that provide PCB fabrication and assembly service, I can't help making a board with the gerber files when I found it last month. The PCB was done on 10th of September, but it takes a few days to find the parts in the BOM. MOSFETs, transistors, capacitors, connectors and resistors are not problem at all, they are all on hands. It takes a few days to find the similar inductor for out put filter.
When all parts were assemblied, I connected the board to +-50V supply and the output to a 4Ohm speaker. But there's no music all when the power switch is turned on! I checked the boards carefully, for about half an hour. finally I found that the diode was not mounted well. Adding some solder at the pin of the diode, the music comes from the speaker when I turned on the power again! The sound is really excellent!
I remember another member of diyaudio said "I lost interest in TA2022 when I get into IRS2092". Now I understand why he said so
We made 2 different type with the same board: one with IRFI4019H and one with IRFI4020H. Both had excellent performance, the only difference is that the IRFI4020 has higher output power. Soon both version will be listed on our site for sale.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
If anything IR are too helpful and there loads of amplifiers like this on ebay and elsewhere and all at low prices.
I got a better response when selling class AB amps.
The 2092 can be a bit touchy about layout and decoupling so I guess that is why they supply gerber files.
My first effort at a pcb just got clicking on the output. The problem was poor decoupling on the mosfets and poor decoupling on the VCC line.
I first tried power inductors and they just fried at 120 degrees ! In teh end I used a t106-2 core with 18SWG enammalled copper wire and that worked great.
I got a better response when selling class AB amps.
The 2092 can be a bit touchy about layout and decoupling so I guess that is why they supply gerber files.
My first effort at a pcb just got clicking on the output. The problem was poor decoupling on the mosfets and poor decoupling on the VCC line.
I first tried power inductors and they just fried at 120 degrees ! In teh end I used a t106-2 core with 18SWG enammalled copper wire and that worked great.
Last edited:
Selecting the right inductor for output filter is very important for IRS2092 or any other class-D. If the wrong type is used the temp will goes high and the sound is going bad.
Since the pwm frequency is up to 800kHz, the selectrion of inductors are largely narrowed. I believe most inductors marked with 22uH/8A in the market are not usable because their magnetic characters is becoming poor when the freq is over 300 or 400KHz. Only materials support over 1MHz can be used for IRS2092. I think that's why your inductors gried
Since the pwm frequency is up to 800kHz, the selectrion of inductors are largely narrowed. I believe most inductors marked with 22uH/8A in the market are not usable because their magnetic characters is becoming poor when the freq is over 300 or 400KHz. Only materials support over 1MHz can be used for IRS2092. I think that's why your inductors gried
Nigelwright,
I THINK that may be what we have been struggling with for the past few months.
Initally, we were having failures of the flyback diodes that clamp the inductive kickback from the speaker load. They were several inches away from the decoupling capacitors and were low enough voltage rating that if someone unplugged the speaker with power applied, it would kill the diodes. I relayed out the board to bring the caps and diodes closer together and got diodes that recovered faster and had higher voltage rating. That SEEMED to solve that problem.
A couple days ago, I read an article about a new Infineon device that drives such FETs "differentially" in order to compensate for false firing due to inductive voltage drops. It started me re-examining our layout and I THINK we may be getting false triggers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.