• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 751 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPA3110D2 application

andrea penna
Prodigy 150 points
Other Parts Discussed in Thread: TPA3110D2

Hi,
I'm doing tests with a sound card that drives two speakers by BTL 8 ohm 2 watt and I saw doing the math by using an input of  the TPA3110  1 volt peak to peak , 24V power supply, I must have a maximum gain of 27dB, but sometimes I have ,with setting 26db gain,  fault signal go low and the audio is turned off, I have to limit the output with the PBTL? (Not used)
Can I use the same  speakers in parallel  ( 4 ohm  having the same power 2W ) with the 24 V?
Can I connect speakers of the same type with a cable length of 10 meters? I must use a shielded wire with particular size ?

Best regards

  • 0
    TI__Expert 7925 points

    Hi, Andrea. Sorry you're running into trouble with the FAULT signal.

    Can you please tell me what board you are using for your tests on the TPA3110D2? Are you trying on the TPA3110D2 EVM? Is the sound card your own design and you're trying out the TPA3110D2 on it? If it's your own design, can you please post a schematic? I've attached a good reference design to compare to if you've made your own design.

    TPA3110D2 reference design using LC filter (good for long speaker wires): 6866.TPA3110D2PWP-LC_EVM-Schematic.pdf

    Sometimes a FAULT is caused by too much current being pulled from the output of the amplifier. Are you connecting 10 meter cables to your current test setup when you observe FAULT going low?

    What is your power supply like in this test setup? What length of wire is connecting the supply and the amplifier board? We have had some trouble with certain power supplies on our EVM, causing enough ripple to shut down the part.

    Best,
    Matt

  • 0 in reply to Matt Lauer
    Prodigy 150 points

    Hi,

    I had already used TPA3110D2 EVM for demo and I designed a circuit with the audio as in file attached.
    The power supply is a 24V 5A Mean Well DIN rail DR-120-24 7571.dr-120-spec.pdf   and using short wire there are not 'problem, while just stick a wire 10 meters audio disappears with activation of the fault.
    I think it is the induced noise resulting in an increased level which acive protection that sends fault ... I have to use a shielded cable?
    Can I use 24V with 4 ohm speakers 2W, it seems from the datasheet that with greater than 16 V minimum load is 4.8 ohm.
    I used 100uF on PVCC ,  I  must use 220uF?

    Regards

    Andrea Penna

    2313.blue_stereo_panel AMP STEREO.pdf

  • 0 in reply to andrea penna
    TI__Expert 7925 points

    Hi, Andrea. Thanks for following up.

    If you are able to update your board, I suggest replacing the 100 µF capacitor with a 220 µF capacitor and replacing the ferrite bead with a short/wire at the PVCC terminal. Ferrite beads on the supply rail have given us problems in the past, and I don't want them repeating here.

    The recommendation for no less than 4.8 Ω for PVCC > 15 V is there because a smaller load at, say, 24 V can cause a LOT of current to come out of the amplifier as the amp switches. The device will probably trip into overcurrent, because the greatest current that can exit the amp is PVCC/load = 24 V/4 Ω = 6 A, which is above the threshold. Are you able to measure the output current coming from your amplifier? Can you read how much current the power supply is producing when the amplifier goes into FAULT?

    The schematic shows PVCC = 12 V, but the power supply you linked is a dedicated 24 V supply. Can you please explain the discrepancy? Can you use the 12 V model to get the results you want? Does your design require a 4 Ω load, or could you use 6 or 8 Ω?

    Shielded cables help for most situations, but I don't think they are the core cause of the trouble. I think it has more to do with too much current.

    Best Regards,
    Matt

  • 0 in reply to Matt Lauer
    Prodigy 150 points

    Hi, Matt,

    I wrote in my scheme 12V, but my client wanted to use the same 24V is for 8 ohm speakers and also for 4 ohm speakers even if 2 w every speaker.
    I think to convince him of the application design in home automation 12V power supplies.
    However, the main problem, which was in crisis everything, including the touch switches fed from my card, which was long strands of 10 meters with antenna and speaker were induced noise.
    Were activated randomly and thanks to your diagram by putting in place of ferrites, 22uH and 680nF I magically solved the problem.
    The only drawback sometimes the audio breaks up with Plimit off and 26dB gain.
    I put the probe on  + of speaker and I see 6V ( power 12V ) with peak to peak modulation that sometimes reaches 8V with the input  2Vpp maximum.

    I must use Plimit pin with two resistor?

  • 0 in reply to andrea penna
    TI__Expert 7925 points

    Hi, Andrea. I'm glad to hear the LC filter in place of ferrites helped you solve the induced noise issue.

    If you're intending to use PLIMIT to set a maximum output voltage swing on the amplifier, then you will need two resistors connected to the PLIMIT pin (one to GVDD, one to GND) to act as a resistor divider. If, however, you don't want PLIMIT activated, tying it to GVDD as you've done in the schematic should be all you need to do.

    What do you mean when you say the "audio breaks up"?

    Best
    Matt

  • 0 in reply to Matt Lauer
    Prodigy 150 points

    Hi, Matt

    I wanted to say that it blocks the sound and then resumes, perhaps too loud?
    The conditions are those said in the previous message with 26db of gain and 4 ohm load on 12V.

    Regards

    Andrea