Hello guys, I bought two Class D amps from aliexpress based on the most popular IRS2092
More details here
IRS2092 IRFB23N15 D Class 30A D Amplifier Board / Mono / 350W/8ohm/700W/4ohm UPC1237 Z09 Drop ship-in Amplifier from Consumer Electronics on Aliexpress.com | Alibaba Group
Everything seemed to be alright as I had 2x45vAC which provides far less than the maximum voltage ~63 but the amplifier protection gets on even if the volume is around 10% which is very bad. I shorted the protection and now the amplifier survives until 60-70% volume and again cut off for a sec and start again. What I have found so far is that the power inductor (I think is a power inductor/ between the relay and radiator) is getting very very hot. Is there any possibility that this is the reason to cut off the speakers or is something else.
Unfortunately, these guys do not provide a schematic and it's hard to me to find what may cause the cut as the protection is shorted like this:Screenshot by Lightshot
Any ideas?
More details here
IRS2092 IRFB23N15 D Class 30A D Amplifier Board / Mono / 350W/8ohm/700W/4ohm UPC1237 Z09 Drop ship-in Amplifier from Consumer Electronics on Aliexpress.com | Alibaba Group
Everything seemed to be alright as I had 2x45vAC which provides far less than the maximum voltage ~63 but the amplifier protection gets on even if the volume is around 10% which is very bad. I shorted the protection and now the amplifier survives until 60-70% volume and again cut off for a sec and start again. What I have found so far is that the power inductor (I think is a power inductor/ between the relay and radiator) is getting very very hot. Is there any possibility that this is the reason to cut off the speakers or is something else.
Unfortunately, these guys do not provide a schematic and it's hard to me to find what may cause the cut as the protection is shorted like this:Screenshot by Lightshot
Any ideas?
Hi, blafioras
- in my opinion, Your supply voltage it's too LOW (NOT too High!) Than on playing You are catching Under voltage protection to work ....
- in my opinion, Your supply voltage it's too LOW (NOT too High!) Than on playing You are catching Under voltage protection to work ....
1. the inductor should not get very hot ... something i wrong.
2. It says it should work from 20Vac and up to 50Vac, so I don't see hot conclude the voltage is too low!!
3. Have you measured DC on output?
4. Have you measured output from mosfets with an oscilloscope to make sure it looks right?
5. I would guess the UPC1237 only works on the relay and does not give fault signal to IRS2092 (but this is a pure guess) ... therefore if it shots down it is probably the overcurrent protection
6. do you actually hear the relay cut off, in any of the 2 situations?
Not sure why the developer has opted for a relay on the output .... should not be necessary with IRS2092, and UPC1237 is not really well suited for use with class d
2. It says it should work from 20Vac and up to 50Vac, so I don't see hot conclude the voltage is too low!!
3. Have you measured DC on output?
4. Have you measured output from mosfets with an oscilloscope to make sure it looks right?
5. I would guess the UPC1237 only works on the relay and does not give fault signal to IRS2092 (but this is a pure guess) ... therefore if it shots down it is probably the overcurrent protection
6. do you actually hear the relay cut off, in any of the 2 situations?
Not sure why the developer has opted for a relay on the output .... should not be necessary with IRS2092, and UPC1237 is not really well suited for use with class d
They used a relay because when you turn it on there are no noises, but when you turn off there are noises as if it were a siren (probably because VAA, VSS and VCC are derived from the rails by zener diodes, then shut down, they also go off slowly , still powering the circuit, although the rails are already discharged enough to partially extinguish the output stage).In some designs where VAA, VSS and VCC are derived from power lines separated from the rails, this does not happen and therefore the relay will not be necessary.
In any case, it seems that our friend has an unsuitable output inductor, (switching frequency too high, over 250khz) this turns into heat, so OCP intervenes prematurely.
In my opinion we should replace the inductor with something more appropriate, like a micrometals t157-2.
MarioRestucci
!
== micrometals t157-2 ==
- Thanks for a proof! On my Sure board instructors are not from the betters too ... :-/
== micrometals t157-2 ==
- Thanks for a proof! On my Sure board instructors are not from the betters too ... :-/
Hi MarioRestucci
If you use a undervoltage detection, and puts the IRS2092 into fault mode when detected, you get a smooth start up, ensuring that all the relevant internal circuits are running, and a smooth shut down, with no unwanted sounds.
Looks to be a coil intended for class d on the website .... but looks can of course be decieving
Do some simple measurements before changing anything, and if possible go to a lower ac voltage until you know it is all working properly
If you use a undervoltage detection, and puts the IRS2092 into fault mode when detected, you get a smooth start up, ensuring that all the relevant internal circuits are running, and a smooth shut down, with no unwanted sounds.
Looks to be a coil intended for class d on the website .... but looks can of course be decieving
Do some simple measurements before changing anything, and if possible go to a lower ac voltage until you know it is all working properly
Thanks a lot for the suggestions guys. I will start doing some testings and will let you know what will come's up.
As MarioRestucci suggested, I have also suspected that the inductor is not properly matched (Chinese cheap PCB) as it should not get that hot (cannot be touched for sure is more than 80 Celsius). I got two equal PCBs and both are doing the same. I think that the problem is not related to the power supply as the minimum requirement is "Working voltage AC dual 20V-50V ", I have 2x47AC and if the voltage was so low why when I bypassed the protection it does work well until 70-80% of the volume whenever it just cuts off and starts again repeatedly, doesn't make sense. It's more likely the maximum output is reached and it cut off from the IRS2092 protection.
I haven't taken any measurements on the output still yet. The relay now does not open as the protection do not command it in any way, so the amplifier runs straight away after power up.
As MarioRestucci suggested, I have also suspected that the inductor is not properly matched (Chinese cheap PCB) as it should not get that hot (cannot be touched for sure is more than 80 Celsius). I got two equal PCBs and both are doing the same. I think that the problem is not related to the power supply as the minimum requirement is "Working voltage AC dual 20V-50V ", I have 2x47AC and if the voltage was so low why when I bypassed the protection it does work well until 70-80% of the volume whenever it just cuts off and starts again repeatedly, doesn't make sense. It's more likely the maximum output is reached and it cut off from the IRS2092 protection.
I haven't taken any measurements on the output still yet. The relay now does not open as the protection do not command it in any way, so the amplifier runs straight away after power up.
Last edited:
No, it doesn't get hot in the idle regime, it's getting hot on continuing listening around the maximum possible. The speaker is 4 Ohms http://www.caraudio-image.ru/manuals/Infinity/REF1260W.pdf. I am wondering if the reason for cut off is the speaker protection? Is that possible, and if so where it is built into the amplifier itself or somewhere else.
It is kinda possible because I tried the amp to another speaker JBL studio ones around 30Watts and the cut off is around 10-20% of the maximum volume which does make sense if the cut is coming from a speaker protection.
In any way, I think that the inductor must be replaced but the question is with what kind.
Last edited:
The output filter should have something like a t106-2 core wound with about 10 turns of enamelled copper wire. Mine barely gets warm.
I had problems with my irs2092 amp making siren noises then a massive thump on power down. IR said I had too much smoothing causing amp to turn off too slowly. I fixed it with a reset circuit that monitored VCC, when VCC got low I held the 2092 in reset through an opto coupler and resistor. I used a PIC with a2d but it could be done with a comparator driving an opto coupler. Make sure you put something like 390r in series with output of opto coupler or you will kill the 2092, at least that's what I found !
I had problems with my irs2092 amp making siren noises then a massive thump on power down. IR said I had too much smoothing causing amp to turn off too slowly. I fixed it with a reset circuit that monitored VCC, when VCC got low I held the 2092 in reset through an opto coupler and resistor. I used a PIC with a2d but it could be done with a comparator driving an opto coupler. Make sure you put something like 390r in series with output of opto coupler or you will kill the 2092, at least that's what I found !
I spent ages trying to get around the 2092 resetting on me when played loud.
One problem is the output mosfets not turning on fast enough and the 2092 thinks there is an over current event and resets.
But the OCP resistor settings must be right too.
Also the dead time.
A tip with 2092 is don't probe it while switched on.
Turn it off, let power supply discharge, connect probe then power up.
If you need to move probe go through turn off sequence again before removing probe.
One problem is the output mosfets not turning on fast enough and the 2092 thinks there is an over current event and resets.
But the OCP resistor settings must be right too.
Also the dead time.
A tip with 2092 is don't probe it while switched on.
Turn it off, let power supply discharge, connect probe then power up.
If you need to move probe go through turn off sequence again before removing probe.
Thanks for the suggestions Nigel.
I found what is the part that getting hot it is power inductor as suspected in my case PD17A-223 http://www.powerhen.com/pdf/PD17A.pdf and probably it's getting hot by overcurrent something I have to measure these days.
Is this power inductor good for a replacement http://uk.farnell.com/vishay/ihth1125mzeb220m5a/inductor-22uh-20-28a/dp/1764424 ?
I found what is the part that getting hot it is power inductor as suspected in my case PD17A-223 http://www.powerhen.com/pdf/PD17A.pdf and probably it's getting hot by overcurrent something I have to measure these days.
Is this power inductor good for a replacement http://uk.farnell.com/vishay/ihth1125mzeb220m5a/inductor-22uh-20-28a/dp/1764424 ?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.