This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPA6017A2: TPA6017A2- Single Ended Alternative

Part Number: TPA6017A2
Other Parts Discussed in Thread: TPA0212, TPA6138A2

Hello,

I was using a TPA6017A2 in my design until I realized that the harness where the product will plug in had a single ground and thus expects single ended audio. In my infinite naivety I decided to short the negative ends of the IC together into the ground path. Surprisingly, this worked and sounded fine, but the IC started overheating and hitting thermal shutdown.

Anyways, Is there a direct alternative to the TPA6017A2 that supports Single Ended applications? I need 2W per channel and preferably some gain.

I've started looking at the TPA0212, but it has more features than I need. 

  • Hi Theodore,

    Welcome to e2e.
    I assume you need single-ended inputs. When you tried to short negative terminals to GND, was it directly or did you used a capacitor? If you didn't used the capacitors, could you try by adding these and see of the device does not heat?

    Best regards,
    -Ivan Salazar
    Applications Engineer - Low Power Audio & Actuators
  • Ivan,

    Thanks for your response! The audio input is differential and I need the amp to output single ended. When I shorted the negative terminals, I did so to each other. I will try adding a 10uF cap and shorting them to the common ground.

    However, I did try to leave the negative terminals floating and used the common ground (using the EVM for the part) and the audio was noisy and low quality. Any recommendations on resolving that? Perhaps some output coupling capacitors?

    Thanks!
    -Ted
  • Ted,

    Please let us know the results after trying with the inputs caps to GND.
    Regarding your test with inputs floating, were they just disconnected? You mention common ground but what is this about? Were the inputs floating or connected directly to GND? Perhaps if the inputs were just floating they were taking in some nearby noise.

    Best regards,
    -Ivan Salazar
    Applications Engineer - Low Power Audio & Actuators
  • Ivan,

    The 10uF caps to ground the negative terminals resolved the overheating issue and the 10uF output caps resolved the noise issues, but there is nearly no bass being transferred over the amp. I'm comparing it to a TPA6138A2 (this part won't work because of its low output power) and the 6138A2 is performing way better in terms of bass response.

    >Regarding your test with inputs floating, were they just disconnected?

    Yes, they were.

    >You mention common ground but what is this about?

    I connected the ground to the ground terminal of the single ended wire, rather than either of the negative terminals.

  • Ivan,

    Another small note: Attaching either one of the negative terminals to the ground pin of the headphone jack resolves the noise and quality (bass) issues. The problem with this is that it also prevents one channel from ever being "off".

    Thanks,
    -Ted
  • Ted,

    You mention headphone jack: Are you using single-ended output as well? If speaker is connected in single-ended then it would definitively need the output caps. TPA6017A2 outputs are centered at mid-supply, no DC bias should be applied to a speaker.
    TPA6138A2 has better bass response because it does not require output capacitors. This device has an internal charge pump to generate a negative supply rail, so the outputs are GND centered.

    Best regards,
    -Ivan Salazar
    Applications Engineer - Low Power Audio & Actuators
  • Ivan,

    For testing, I am using stock Apple EarPods (single ended jack). The end product is feeding a stock amplifier in a vehicle (using single ended inputs), so perhaps my testing is skewed here. The overheating issue is solved thanks to your suggestion, so I have no qualms using this IC in my application.

    I will test in the vehicle with the negative outputs shorted to ground via the 10uF caps and see if audio quality is acceptable. If it's not, I will add the output capacitors and test again.

    Thanks so much for your assistance so far! I will report back after I've had a chance to test further.

    -Ted
  • Ivan,

    After some testing, I've found an acceptable configuration where the IC does not overheat and audio quality is acceptable.

    I've added 3 10uF ceramic capacitors. One for each negative input and the other to filter noise from GND. I then connected the other side of the caps to the GND output into the single ended system.

    Can you think of any issues which may be caused by this configuration? So far, my testing has yielded no issues.

    Here's the layout, which makes more sense than the schematic. Please look at C19, C20, C21 in the lower left:

    Thanks!

    -Ted

  • Ted,

    I'm a bit confused that you say that C19 and C21 are at "negative input" but these are actually at the output.
    However if this is working fine for your application it should be OK. By placing C19 and C21 it seems that you're balancing the BTL output of the amplifier, if at some point you need to change these components you could try replacing them with resistors.

    Best regards,
    -Ivan Salazar
    Applications Engineer - Low Power Audio & Actuators
  • Ivan,

    My apologies for the confusion -- I misspoke -- I did mean negative output. I will try a 10k ohm resistor in place of C19 and C21 and see if it impacts quality negatively.

    Thanks for all your help and for reviewing my updated design.

    Thanks!

    -Ted

  • Ivan,

    Just FYI, I tried using 10k ohm and 100ohm resistors instead and though they prevented the IC from overheating, they audibly diminished audio quality. I think it's best if this configuration uses the 10uF caps instead.

    Thanks!

    -Ted