Problems getting TPA2006D1 to work

Hi, Brent,

Sorry to hear of your troubles using our device.

If you look at the outputs, pins 5 or 8, and see what you have with a scope? You should get a signal like shown in Figure 34 of the TPA2006D1 data sheet. You may have to remove your load in order to see the waveform (assuming the problem is due to your speaker).

With no input signal, you should see approximately 50% duty cycle on both outputs (in phase).

-d2

Hi Don,

Both outputs are always at 0V.  I never see them change values from the ground level (other than mV worth of noise).

With the inputs tied to ground, the outputs are also ground.

With a 1 V p-p, 1 kHz sine wave on one of the inputs (and the other grounded), I see about a 800 mV p-p sine wave on the other side of the capacitor.  One item to note that I hadn't grasped before is that the input wave is centered at 500 mV (meaning that the lowest voltage level is 0 V and the highest is 1 V).  At the other end of the capacitor, it is centered around 0 V (meaning that the lowest is -400 mV and the highest is 400 mV).  The power supply for the TPA2006D1 is 0 V and 5 V so I am unsure where the negative bias is coming from.  I have no negative voltages on my board.

At the input pin, the wave is reduced to about a 400 mV p-p sine wave centered around 0 V.  I still see no movement on the output signals.

I have tried this on two different board and with and without the speaker attached (load and no load).  The results are all the same.

Could the negative bias be causing the problem?

Thanks,

Brent

Hi, Brent,

With no input signal applied, what DC voltage do you measure on pin 4? This should measure Vdd/2. 

D2

Hi Don,

It is not Vdd/2.  I am measuring 0 V.

Brent

Hi Don,

Based on what you suggest I should be seeing on the inputs and what I am actually seeing, there is something grossly wrong (as I had suspected in the beginning).  I am still trying to figure out what that something is.  I don't believe I have hooked it up incorrectly as I followed an example directly from the datasheet.  I don't think there are issues in layout because I have measured the signals on the board (mainly +5V and ground).  That leaves me with two other options that I can think of.  First, are the parts bad?  I have looked at parts on two different boards and both behave the same way.  Could it be a bad lot????  Seems unlikely to me but I am running out of other options.  Second, could some power up sequence on the board be blowing the parts?  Again, this seems unlikely since only 5 V and ground are applied to the device and the inputs are decoupled through capacitors.  Any other thoughts of what might be happening?

Thanks,

Brent

Hi, Brent,

Sorry for the delay... 

I am kind of at a loss for what could be wrong. I highly doubt bad parts, but you could always order a few samples from our website and see if those worked.

It sounds like you might be activating the protection circuitry in our part, but I can't see what would be doing this --> the only thing I can't see is -- does SPEAKER+ and SPEAKER- go anywhere?

-d2

Hi Don,

Thanks for getting back to me.  Sorry for the pressure but I am getting a lot of pressure to get this going.

With regards to the SPEAKER+ and SPEAKER- lines, I assume you are referring to my schematic.  No, they don't go anywhere else - just refer to empty pads that I put for a contingency.  The problem is the contingency does not work and I need to get this option working.

I too am at a loss with regards to what could be happening.  I think we need to start thinking outside the box.  What could invoke the protection circuitry.  Is there something in the start up that could cause it to be invoked?  Something unique is going on with my circuit that we need to get to the bottom of.  What other things could I look for that could help you determine what could be going on?

Thanks,

Brent

Hi, Brent,

No worries, we've all been there...

Try shorting the inputs together directly at the chip. You should then see 50/50 duty cycle at the output.

Any capacitance on the output can trip the Short-circuit protection. I don't see any on your schematic, but check that as well. You may even want to lift the pins off the board and scope them like that.

-d2

Brent,

I think you said you already checked it, but make sure the power supply is solid.

And, please don't think I'm insulting your intelligence, but the SHUTDOWN-Z pin is held high, right?

-d2

Hi Don,

No problem with the insulting the intelligence.  At this point, we need to check and double check everything.

With rechecking everything, with the debugger (for the MSP430) attached, I was getting a different behavior than with it removed and I have done all my work before with it attached.  It does not allow the 5V supply to drop all the way to 0V with the supply removed.  It only reaches about 1.2V.  Working through that, I was finally able to get a DC value on the input lines for the amp.  Then, double checking everything, I found that a hidden capacitor in the alternative circuit was inadvertently loaded.  It was not supposed to have been loaded.  It ended up putting 10uF load on the Vo+ pin.  I removed it and am now getting a sound out.  Thanks for pointing me in the right directly of the protection circuitry and capacitance loading.

I am not getting the sound out that I believe I should.  I will keep looking but if you see something blatant, that would help.  I am sending a 1Vp-p sine wave from the DAC of the MSP430.  On the opposite side of the cap, I am seeing 500mV p-p.  On the TPA2006D1 side of the resistor, I am seeing 100mV p-p.  Since the input resistance of the pin is 150kohm and my resistor is 240kohm (2 of them), I believe it should be higher.  Any ideas?

Thanks,

Brent