• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 30 subscribers
  • Views 570 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.

TPA3118D2 1Hz error cycling

Zachary Lyzen
Prodigy 140 points
Other Parts Discussed in Thread: TPA3118D2, TPA3130D2, LM317

I am building the schematic shown in the datasheet for the tpa3118d2 http://www.ti.com/lit/ds/symlink/tpa3118d2.pdf of the 2.1 sound system, and everything is working on board except the IC for the left and right speakers.  

THe BTL subwoofer IC is working as intended, until I try to incorporate the other part.  When I apply power to the SDZ pin of the IC for the left/right speakers the voltage is pulled low for both chips, causing neither to function.  I've removed this chip various times, checked for shorts, and tried 7 other chips as well as a couple tpa3130d2's in it's place all to no avail.  I am getting the fault ->sdz pin error cycling which gives me a faint 1hz click from the actual IC and from a speaker when hooked up.  

I have built this using the same components/values as shown in the datasheet's schematic, is there possibly a problem with the output filter of the L/R amp causing this fault?

I have tried to power them both up at the same time, or one or the other powered up first.  Oddly enough the subwoofer IC works fine without the L/R IC even soldered on the board... which is weird because the subwoofer IC is in slave mode.

  • 0
    TI__Guru**** 189401 points

    Hi, Zachary,

    It sounds like you are tripping the short-circuit protection. Maybe your filter has some problem? Or maybe your load is too low of an impedance?

    Try removing the caps to ground after the inductors to see if they are saturating or have some other problem. If the problem still exists, then check your load resistance is withing allowed limits - and remember a DMM reading DC impedance may not tell you the entire story...

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 140 points

    Don,

    I have been testing this with both no load connected, and with the intended load which is an 8Ω speaker.

    The schematic on the data sheet uses .68uF caps in parallel with the output I believe. and these come after the inductors.  The capacitors used for the subwoofer filter are 1uF and as I mentioned earlier, that amp is fine, and working.  So I guess my question is what can I do to my filter so that it doesn't trip, I have meticulously checked for shorts to no avail.  This is the filter cap I am using http://www.mouser.com/ProductDetail/TDK/C1608X5R1H684K/?qs=%2fha2pyFadujKzWoApKf6bnQRxZLKXrRygAnC9wKvaeL9GWFOw4FJzye9%2fjSISFG%2f .

    On a slightly different subject I am getting a 60Hz hum on the subwoofer's output.  I know this is because of my unregulated power supply (200VA transformer > full-bridge rectifier > 10mF capacitor)  If I add ~6 lm317's in parallel from the power supply to the amps and drop the voltage the minimum amount will the ripple be smoothed out?  It seems that that could handle 9amps max.  I could even add a bank of capacitors after them too I guess.

  • 0 in reply to Zachary Lyzen
    TI__Guru**** 189401 points

    Hi, Zachary,

    As the load resistance is part of the output filter, you should always operate our ICs with the load attached, otherwise you need additional components to prevent damage to our IC.

    I believe the problem is your inductor, not your capacitor. You need to insure your inductor is not saturating.

    The 60 Hz hum is most likely caused by a ground loop in your system. Our device has good PSRR, so 60 Hz applied at the power supply will not cause hum at the output. Check all your grounds... You only want one point connected to the mains "ground." Keep in mind maybe your source has a different mains "ground" potential. Try using an undocked computer or cell phone for your signal source.

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 140 points

    Regarding the hum, I only have my chassis grounded to earth, and the mains are fused to add safety.  I was using a free-standing ipod to source the audio, which is why the hum was bewildering.  Could've been the board quality/layout though.  Most of the board is ground to also act as a heat sink.

    This is the inductor I am using, it is rated to 6A and I thought that this would be adequate.  What is a good way to tell if it is saturating?  So you're suggesting that I have all the loads connected and then boot it up that way?  Should I do anything to change my filter since my load is 8Ω, not 4Ω, or keep using the one on the data sheet?

  • 0 in reply to Zachary Lyzen
    TI__Guru**** 189401 points

    Hi, Zachary,

    Try removing the caps to ground after the inductors to see if they are saturating or have some other problem. If the problem still exists, then check your load resistance is withing allowed limits - and remember a DMM reading DC impedance may not tell you the entire story...

    Check out the attached presentation one of my colleagues made a few years ago. Review it all, but pay particular attention to page 5.

    -d28357.APA-diff-in-connect-091123.pdf

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 140 points
    It appears the output filter capacitors are the problem.  It seems to be the .68uF caps on the satellite speaker filter.  My capacitors are 50V X5R 0603 .68uF capacitors.  I notice in the evaluation schematic 1206 sized capacitors are used.. besides this I cannot find a difference.  Do you recommend any particular capacitors for this filter?
  • 0 in reply to Zachary Lyzen
    TI__Guru**** 189401 points

    Hi, Zachary,

    That's interesting... How did you determine this?

    The key parameter for this cap is a constant capacitance versus applied voltage. I think sometimes the larger case size have less of a dramatic impact.

    We used to use metal film caps for the output filter as they have much better characteristics in this regard, but, of course, they are more expensive, so most of our customers want to use ceramics instead.

    The changing capacitance usually shifts the cut-off freq of the filter, and impacts THD. I have not personally seen it contribute to a "hum." 

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 140 points

    Hello Don, it's been a while.  I've finally had time to come back to my 2.1 system design.  I had the boards made by advanced circuits, keeping the general layout similar to the EVM but making it a 2.1 system.  

    I am getting a few weird issues upon power up.  

    One side (right) plays normally, though once volume gets from quiet to reasonable it distorts and the amp shuts down (thermal?)

    After a second the left side "thumps" and then is silent

    The subwoofer is silent

    The entire time both IC's are making this unbearable high-pitched whine, even with the speakers disconnected!

    Do any of these symptoms possibly point to what I might have going on?

    Thank you,

    Zack

    ps I followed the EVM for filter layout, and for the snubber at the boostrap/ic output pins.  This snubber wasn't in the 2.1 example,  or on the 3118 datasheet, which also had a different output filter (an extra 1uF cap)