L15D sometime oscilates

[elic]

Hi everyone,
I have built the L15D class D amplifier (got two of them) and started playing with it. My impressions so far:
1. Even with relatively high power it stays cold as a dead fish
2. I think it's a little attenuated at the high audio frequencies but that could be a wrong impression as I did not actually measured the response.
3. Sound it quite good and clean otherwise.
4. With some high power speakers (which are quite old and poor sounding) the amplifier may not start (the blue led flashes as if trying to recover but fails).
and last:
If I connect the amplifier ground to the mains ground, it starts oscillating bad and one of the resistors (did not check which one exactly) gets over heated.

Any ideas, tips how to proceed?
Thanks,
Eli

[MrSlim]

Originally Posted by elic

Hi everyone,
I have built the L15D class D amplifier (got two of them) and started playing with it. My impressions so far:
1. Even with relatively high power it stays cold as a dead fish
2. I think it's a little attenuated at the high audio frequencies but that could be a wrong impression as I did not actually measured the response.

Are you using it as an "integrated amp with just a volume control on the input: This is a thread discussing what happens when doing that with an ICEpower module, and I suspect the same thing might happen with the L15D, since it has fairly low input impedance as well..

3. Sound it quite good and clean otherwise.

4. With some high power speakers (which are quite old and poor sounding) the amplifier may not start (the blue led flashes as if trying to recover but fails).
and last:
If I connect the amplifier ground to the mains ground, it starts oscillating bad and one of the resistors (did not check which one exactly) gets over heated.

Any ideas, tips how to proceed?
Thanks,
Eli

You might want to post this on LJMLJM's L15D Thread, he might respond more quickly. There was a couple of posts about upping the wattage rating on one of the resistors from 1 W to 2 W (something like that), but I can't put my finger on it at the moment.

[MrSlim]

This is probably unrelated, but worth asking.. Did you assemble the L15D boards yourself, or did you buy them prebuilt? Someone mentioned that some of the prebuilt one's did not have heatsink compound on the MosFets on the heatsink..

[elic]

I built them myself. I also made a few modifications due to the fact I run them on +/- 38V while they are designed for +/- 50V. The changes involved are to decrease values of resistors in series to zeners to retain the same current that would have flown if I used 50V.

[elic]

Oh, and forgot to mention - I intend to use in a box containing four microphone inputs preamp and two guitar inputs all through a mixer (actually opamp based adder) and from there to the L15D.

[MrSlim]

Originally Posted by elic

I built them myself. I also made a few modifications due to the fact I run them on +/- 38V while they are designed for +/- 50V. The changes involved are to decrease values of resistors in series to zeners to retain the same current that would have flown if I used 50V.

Just to confirm, you changed the Zener's also correct?

[elic]

Originally Posted by MrSlim

Just to confirm, you changed the Zener's also correct?

No, just the resistors. I wanted the current to remain the same as it would be if I used 50V, but I did not change the zener values.

[MrSlim]

Ok, to make sure we are talking about the same thing, I'm talking about Z100 and Z101. They are for over voltage(OVP) and under voltage(UVP) protection, meaning that your power supply has to be between the voltages of those two Zeners, or the amp will shut down.(according to IR's reference design document (see page 24 here) I don't think they have anything to do with how much current the amp can put out. Current limiting is handled by the IRS2092 chip itself.

These amps scale in wattage (and current output) depending on the power supply voltage you feed them(as long as the resistors and the zeners are changed). Reducing the power supply means reducing the amount of current output (and wattage). You cannot expect to get the same power (watts) from the amp if you reduce the power supply voltages.

From what I understand the L15D is based on the components for the AMP7S-150(see page 37 for the differential chart) in the Ref design. The OVP and UVP zeners are 68 and 39 Volts for the AMP7S-150, and if your power supply is nominally 38 V, you are likely right on the edge of the UVP protection kicking in, explaining why it doesn't start up sometimes, and possibly the oscillation also.

You need to change the zeners to be the parts used by the AMP7S-100, which are 47V (OVP) and 30V (UVP) respectively for voltage of the zeners. The part numbers for the Ref Design are SMD though, so you'd have to find the equivalent Through hole parts, like 1N5256B-TR (30V) and 1N5261B-TR (47V).

[elic]

(sigh) I asked the guy who sold me the kit for a schematic so he sent me this link:
http://www.nutsaudio.com/photo/iraudamp7s.pdf
I know it does not exactly match the kit I got but it's pretty close.
Anyway, referring to page 21 in this PDF, I changed R117 and R118 which are in series with zeners that provide +/- 5.6V to VAA and VSS.

[MrSlim]

So you only changed R117 and R118? I think you need to change all the components listed on page 37 that are different for the various versions of the circuit.