LM48511: Gain level changes?

David Tiefenbrunn

Part Number: LM48511

Hi,

I am working on getting a design with this power amp. I have 5V power supply. I have the Shutdown pin high, The BSD pin high,The FBS pin low. Fixed frequency mode currently.

The application has 5V power, no need to shutdown, or run off battery, so it is set up to be able to operate in full power mode.

Doing some audio testing, I have discovered 2 things - I need some kind of filter network to get usable readings of the speaker output, since it's PWM square waves, my audio analyzer doesn't like it at all....

The other is this - As I gradually increase the drive signal, the output begins rising, as one would expect. At a certain threshold, the output suddenly [but smoothly] rises up significantly. As if some kind of noise gate turns on. The output level then continues to rise with increases of the drive level, until maximum.

Why does this happen? can it be disable? Or is it some malfunction?

I have to make some kind of filter, then I will be able to make meaningful audio level measurements.
I also have a second PCB, which I will test to see if the odd behavior is the same, or if it is the result of some damage.

Dave

over 7 years ago

0 Luis Fernando Rodriguez S. over 7 years ago

TI__Guru* 84915 points

Hi, Dave,

Is it possible to provide a circuit schematic to have a better approach to this issue?

Usually, we take the measurements on our Class-D amplifiers with the help of our Class-D Filter Boxes: http://www.ti.com/lit/an/sloa107/sloa107.pdf. Is it possible to use a filter like this one?

Please let me know if you have the same results in the second PCB.

Best regards,
Luis Fernando Rodríguez S.

0 David Tiefenbrunn over 7 years ago

Intellectual 785 points

I don't know why I can't get this image to upload the correct orientation, I edited it before uploading. .... Oh well,
This shows a 3rd strange thing. I am driving the amp with a differential source, sine wave, constant amplitude, yet I have this odd in audible level variation.

Using a 1K resistor and 60nF capacitor as a filter to hide the PWM.

0 David Tiefenbrunn over 7 years ago in reply to David Tiefenbrunn

Intellectual 785 points

I am looking at the filter box page... Meanwhile, here is the schematic -

J-668 Rev 00 Amplifier.pdf

... or maybe not? IT's a .pdf, not displaying.

0 David Tiefenbrunn over 7 years ago in reply to David Tiefenbrunn

Intellectual 785 points

Ok, it just showed up as a link. Anyway, I am testing the amp using a bench supply for now.

0 David Tiefenbrunn over 7 years ago in reply to David Tiefenbrunn

Intellectual 785 points

The second board is doing about the same thing. I put it in spread spectrum mode.

0 David Tiefenbrunn over 7 years ago in reply to David Tiefenbrunn

Intellectual 785 points

OK! ...

I built a filter, as shown in the link. Set up the O-scope to measure differentialy, and I get a normal looking sin wave. None of the level shifting stuff like in the image I posted earlier. That earlier image was made with a single ended filter, to one side of the speaker. Still seems weird to me that the peaks are not all the same, but, ok, it works now.

Second part of my problems - the odd level shifting "noise gate" effect I was getting -

Well, now that I have a good scope display, I watch it as I adjust the level, and hear the level jump, and nothing shows on the scope - I connect a different speaker, and problem is gone.

I have been working with audio for decades, never saw a speaker have this kind of malfunction.

0 Luis Fernando Rodriguez S. over 7 years ago in reply to David Tiefenbrunn

TI__Guru* 84915 points

Hi, David,

Thank you for providing your feedback about this issue.

I'm glad to read that you could get a normal looking sine wave. The Class-D filter is always a good option for audio measurements. It helps to leave only the audible portion of the Class-D output.

Regarding the speakers that you used during the "noise gate" effect, are the two speakers in this test from different manufacturers? Or do they have the same properties?

Best regards,
Luis Fernando Rodríguez S.

0 David Tiefenbrunn over 7 years ago in reply to Luis Fernando Rodriguez S.

Intellectual 785 points

Hi Luis,

The speakers are completely different. I also just tried another one I had hanging around.

Also, I connected the speaker to a commercial power amp, and it does the same thing. Also found that applying moderate pressure to the frame of the driver causes all sorts of tone quality changes, and level shifts.

Dave

Audio

Audio forum

LM48511: Gain level changes?