colleagues,
I'm starting on class D amplifiers and I'm based in IRS2092, but early on I've had a big problem, it worked!!
I mounted the electrical scheme of AN-1138, only changed by the FET IRF540 and my source of 50V symmetrical.
The capacitors are self-osc 1nF and resistor 500R.
I hope someone can help me.
hugs
Fernando
I'm starting on class D amplifiers and I'm based in IRS2092, but early on I've had a big problem, it worked!!
I mounted the electrical scheme of AN-1138, only changed by the FET IRF540 and my source of 50V symmetrical.
The capacitors are self-osc 1nF and resistor 500R.
I hope someone can help me.
hugs
Fernando
question
how much power can i get with your design?
can you post the gerber files please?
My project...
how much power can i get with your design?
can you post the gerber files please?
I rather fancy using an amp like this as a 50 Hz re-generator ( 230 V via output transformer). Anyone tried getting a continuous output from an amp like this?
Another related question. Some class D amps are suitable for 5V. A friend asked me to build some desk top speakers . 500 mA from USB seems too low to be useful . Then I remembered Hypex state 10 to 1 crest factor in audio amps. Maybe the 2 x 2.5 watts is realistic as the USB should tolerate slight abuse. 2.5 watts is almost OK in SE. One chip was not a million miles away from Hypex specs. No output choke used and none needed said the makers.
My plan is to make it bridge and really be naughty. I will have a beefy 5V when practical to use it.
Another related question. Some class D amps are suitable for 5V. A friend asked me to build some desk top speakers . 500 mA from USB seems too low to be useful . Then I remembered Hypex state 10 to 1 crest factor in audio amps. Maybe the 2 x 2.5 watts is realistic as the USB should tolerate slight abuse. 2.5 watts is almost OK in SE. One chip was not a million miles away from Hypex specs. No output choke used and none needed said the makers.
My plan is to make it bridge and really be naughty. I will have a beefy 5V when practical to use it.
Hello to all!!!
I'm new on this forum, but not new in electronics.
I have a problem in this schematic:
[img=http://s6.postimg.org/8n9vzgfy5/sema.jpg]
Which PNP/NPN pairs can go to use in the buffer?? T1, T2, T3 and T4 in schematic)
Thank you!!!
I'm new on this forum, but not new in electronics.
I have a problem in this schematic:
[img=http://s6.postimg.org/8n9vzgfy5/sema.jpg]
Which PNP/NPN pairs can go to use in the buffer?? T1, T2, T3 and T4 in schematic)
Thank you!!!
I had quite a serious problem with my irs2092/irfb4227 amp.
On power down I would get a loud siren type noise through the speaker then just as it turned off a huge thump.
I asked IR about it and they said I needed to add a reset circuit to hold the 2092 in reset when VCC dropped below about 12 volts.
It seemed my problem occurred because I was using large smoothing capacitors giving a very slow turn off.
I came up with a little PIC based circuit with an opto isolator driving the reset pin. The PIC had an a2d to read the VCC level. Worked very well.
Nigel, the amplifier is a commercial design using IRS2092S driving two Irfb4227 directly. I did fix some problems like a broken terminal on one of the high voltage filter caps, a shorted 5.1volt zener in the vaa circuitry, replaced the driver ic and outputs, it worked for a while with the volume limited before resetting. now it resets all the time. looking at Mag's schematic I'm questioning the 10k that goes to pin 14 (HO). Can you shed any light on that part of the schematic?
Thanks.
Thanks.
My first 2092 pcb design had reset problems.
The decoupling was poorly laid out. I needed to decouple 2092 closely and the output mosfets. I shortened pcb tracks as much as possible.
I then had to change the over current resistors to get rid of the reset problem.
My second design used irfb4227's and that reset to start with too. The 2092 just couldn't drive it hard enough so I added dual transistor gate driver and that fixed it.
The 4227 gates were highly capacitive and slowed down the gate pulse and slowed down turn on of 4227, this made 2092 think there was an over current event and it reset.
On my third design I found I needed to decouple the gate driver transistor circuit closely to get full power.
The best way to cope with the 2092 is to use the IRAUDAMP pcb layout.
The decoupling was poorly laid out. I needed to decouple 2092 closely and the output mosfets. I shortened pcb tracks as much as possible.
I then had to change the over current resistors to get rid of the reset problem.
My second design used irfb4227's and that reset to start with too. The 2092 just couldn't drive it hard enough so I added dual transistor gate driver and that fixed it.
The 4227 gates were highly capacitive and slowed down the gate pulse and slowed down turn on of 4227, this made 2092 think there was an over current event and it reset.
On my third design I found I needed to decouple the gate driver transistor circuit closely to get full power.
The best way to cope with the 2092 is to use the IRAUDAMP pcb layout.
Maybe you used fake 2092 , 4227 ?
I made over 100pcs of these amplifiers , 80% with 4227 without any buffers , directly from 2092 via 4.7Ohm. DT set at maximum . Temperature of the 2092 is about 70 deg of Cel. Frequency - 300-330kHz. Many in bridge mode.
No issues.........
click here
Attachments
Last edited:
Dimonis, I have two amplifier modules that are for the Gemini HPS-215 BLU series powered speakers, that uses the IRS2092S driving two IRFB4227,I am suspicious that some of the values used can be wrong, will you be willing to share the schematic of your amplifier, if just the amplifier section?
Thanks.
Thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.