Hello everyone,
I built a class D amp with the IRS2092s. I've been fighting an issue with the amp cutting out at higher volumes with a speaker load. I can get about 400-450 watts out of it before it cuts out. When it does cut out the output is not yet clipping, and does not look distorted.
Previously, I never tried to use a resistive load to see if it would behave the same. Today I tried it and I'm able to drive the amp into clipping with a 2 ohm resistive load and drawing well over 600 watts at the input and the amp did not cut out.
I'm not sure what to make of this. Any ideas on what I should play with to prevent it from cutting out with a speaker load? Could it be the inductor is saturating? The first stage inductor is only rated 10amps... 😀
I built a class D amp with the IRS2092s. I've been fighting an issue with the amp cutting out at higher volumes with a speaker load. I can get about 400-450 watts out of it before it cuts out. When it does cut out the output is not yet clipping, and does not look distorted.
Previously, I never tried to use a resistive load to see if it would behave the same. Today I tried it and I'm able to drive the amp into clipping with a 2 ohm resistive load and drawing well over 600 watts at the input and the amp did not cut out.
I'm not sure what to make of this. Any ideas on what I should play with to prevent it from cutting out with a speaker load? Could it be the inductor is saturating? The first stage inductor is only rated 10amps... 😀
Attachments
Should have taped a little .47uf across the 2ohms and watched the scope, bet it would have rang like a bell then turn off way before 600w
Cheers George
Does the data sheet say anything about a protection feature for overvoltage/overcurrent?
I mean, the title says: PROTECTED DIGITAL AUDIO AMPLIFIER.
Then on the first page:
Programmable bidirectional over-current
protection with self-reset function
You did check the data sheet, didn't you?
Jan
I mean, the title says: PROTECTED DIGITAL AUDIO AMPLIFIER.
Then on the first page:
Programmable bidirectional over-current
protection with self-reset function
You did check the data sheet, didn't you?
Jan
Jan, it has both low & high side over current sensing. I've adjusted both previously including even disabling them completely to the same effect.
Could it be your power supply collapsing at that high power?
Did you do the test with the 400W just a few seconds or for prolonged time?
How long before it shuts down?
Jan
Did you do the test with the 400W just a few seconds or for prolonged time?
How long before it shuts down?
Jan
My rails are normally +/- 65V to 70V depending on input voltage.
In both cases I am playing a 40hz tone.
With the 2 ohm resistive load they are dropping to around +/- 45-50V, mainly due to my battery bank dropping to around 12V at that power level. I can drive this load into clipping without any issues.
With a speaker connected - I can for example, play it continuously at let's say 400W without any issues, but once I turn it up a little bit more to 405W then it will cut out. When it does cut out, the output waveform is not yet clipping. The power level at which it cuts out also seems to change a bit depending on the frequency as well.
In both cases I am playing a 40hz tone.
With the 2 ohm resistive load they are dropping to around +/- 45-50V, mainly due to my battery bank dropping to around 12V at that power level. I can drive this load into clipping without any issues.
With a speaker connected - I can for example, play it continuously at let's say 400W without any issues, but once I turn it up a little bit more to 405W then it will cut out. When it does cut out, the output waveform is not yet clipping. The power level at which it cuts out also seems to change a bit depending on the frequency as well.
Like I said it's probally unstable into the smallest capacitive loading and turns off.
Tap a little .47uf across the 2ohms and watch the scope, bet it rings like a bell then turns off
Cheers George
Last edited:
A real long-shot but could you be suffering from supply pumping (and hence over-voltage) with the speaker connected? Is there any over-voltage protection built in?
George, I will try that later. But I am aware that an active load causes the issue. What's would I change to prevent it?
Abraxalito, I have a DC protect circuit on the output but the components are not installed. So the only protection that's there currently is what the irs2092s has built in.
Abraxalito, I have a DC protect circuit on the output but the components are not installed. So the only protection that's there currently is what the irs2092s has built in.
Steps in fault finding, it's a what I do to find what is causing it on the bench not in the system, that then give you a direction to go in if it is capacitive loading that turns it off, if it dosen't then look to the other power supply problem as Abraxillito has said eventually you'll nail down the area to work on.
Cheers George
I was able to resolve the problem. Somehow internal VREF on the 2092 was not 5.1V, it was ~3V. Maybe the 2092 got damaged during all the tests I did. Swapped the 2092 to a new one and it no longer cuts out even with a 1 ohm speaker load max output input clipping 🤓
