My latest project 1000 watts class d amplifier.
1000 watts into 4 ohms with +/- 60 volt supply.
It uses basically the irs2092 datasheet circuit but with transistor gate drivers.
The only complication is the addition of a PIC microcontroller to control the shutdown pin.
If the VCC goes below 15 volts the 2092 is shutdown by the PIC.
I found with large power supply capacitors I got huge thumps or siren type noises on power down.
I notified IR of the problem and they said that I had done the right thing adding a simple PIC reset circuit.
1000 watts into 4 ohms with +/- 60 volt supply.
It uses basically the irs2092 datasheet circuit but with transistor gate drivers.
The only complication is the addition of a PIC microcontroller to control the shutdown pin.
If the VCC goes below 15 volts the 2092 is shutdown by the PIC.
I found with large power supply capacitors I got huge thumps or siren type noises on power down.
I notified IR of the problem and they said that I had done the right thing adding a simple PIC reset circuit.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
I cant measure the rails at high output current as I don't have a speaker large enough.
The inductance is around 22uH.
There is quite a bit of carrier left on the output so I might need to increase the inductance or capacitance a little.
It sounds Ok though.
If you have very large capacitance on the 2092 supply rails as you switch off the reset circuit doesn't cut in in time and you get a huge thump or siren type noise through the speaker.
The PIC measures the 2092 VCC voltage and if it drops below 15 volts it holds the 2092 in reset.
I think what happens the "rate of change" that the supply filter caps discharges is too slow for the ADC or what ever I/O you using on the PIC to detect logical "power off state"
You can use the PIC to measure power off immediately using a "loss of AC detector" detecting power down using a tap from the transformer or just don't use a PIC and use a relay with the same method i described.
To test it I am just using a 2 amp transformer.
The rails do sag about 15 volts when driving it hard but that's to be expected with such a poor supply.
It really needs a 100VA transformer not a 200VA.
It is very loud even on the small transformer.
I found the VCC supply was dropping out at full power so I reduced the dropper resistors although this probably wont be needed on a 1000VA transformer.
using a relay does sound much simpler compared to PIC. i'm also using a simple AC detector drawn from the transformer tap to a diode and small filter cap to detect AC loss.
i'm guessing it'll still sag with a large transformer. it's fundamental to all class d.
previously i'm using 500VA toroid and it still sag ~10V during heavy bass scene. adding more rail capacitance, rail decoupling caps and changing the layout doesn't improve much, until i bridge them.
now they hardly sag anymore. during intense bass scene they still go down 1-2V but that's it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.