Connecting two IRAUDAMP9 Class D Amps in BTL mode

[MKJ]

Hi All working on an intresting project and would like some input.

I am designing a high voltage (1500V), high current (500mA) AC power supply and was playing around with using a high power class D amp to power the primary of a 1:35 transformer. I've used a IRAUDAMP9 ingle channel 1700 W (2 Ω ) Half-bridge Class-D Audio Power Amplifier but I am having an issue with bus pump causing the powersupply im using to shutdown when I get to around half the desired power (~350W). I'm using a modular switchmode PSU from TDK-Lambda thats able to supply +/-75V (1200W total). I've played around with increasing the bus capacitance but have had very little impact. The bus pump is not much of a surprise as the frequency of the signal is only 50Hz.

To solve the problem of the bus pump I have been playing around with the idea of using two IRAUDAMP9 boards in a BTL confinguration to reduce/remove the pumping effect. Im using a op-amp to invert the input signal so that The two are 180degrees out of phase but I'm having a problem where when I connect the inputs to the amps high frequency (25kHz) is present at the outputs. If I only connect one input to one of the boards the output is as expected. 

I've played around with the series resistors (R3/R4) with that only having a detrimental effect.

Does anyone have any ideas what may be causing this or what else I can try to circumvent the bus pump.

Much appreciated

[Pranava_YN]

Hi,

Can you please share the waveform of the output?

One suspicion is that the ground loop might be causing this issue, but before concluding please provide us the waveform showing the amplitude of this high frequency oscillations.

[MKJ]

Hi Pranava,

That would make sense. Unfourtunatly one of the amps died during the testing and dont really want to try another one until I've figured out some sort of a solution. I did find one photo I took of the output of the two amps which should show you what you need to know.

This was a couple weeks ago so struggle to remember what the exact operating conditions where but I believe that the input to both amps at the time of this photo was 0V through the circuit from my original post.

We are currently testing with a single amp but are struggling to deal with the bus pumping becasue we are limited with the amount of bus capacitance we can use with the SMPS.

Thanks

Jack

[Pranava_YN]

Hi,

Can you also specify the bus capacitances at the input and output side? Also let us know the type of capacitors used.

[MKJ]

The only bus capacitance is that which is fitted on the IRAUDAMP9 and the output filter caps of the SMPS (exact value not specified in datasheet). I assume that in a full bridge configuration this amout of input capacitance could be reduced significantly?

I've also noticed that the BOM in the datasheet for the IRAUDAMP9 contains a few now obsolete components, such as the bus caps (ESMG101ELL122MP40S). Is there a list of alternative components that have been used?

Thanks

Jack

[Pranava_YN · Accepted solution]

Hi,

Sorry for the delay. I discussed this with experts internally. We suspect that It might be intermodulation problem, having two different oscillating frequencies (two amps) that are not in sync. Please share the schematic of the complete setup so that we can review.