Hi all,
testing our TPA3132D2 filterless implementation resulted in a permanent faulty output .
The outputs are pretty much standard, means 220nF 50V X7R bootstrap-capacitors, no bootstrap-snubbers, no LC or ferrite-filter.
Decoupling per channel: 1nF COG + 100nF X7R + 330uF Low-ESR, AVCC is filtered via an RC (10R + 1uF).
Testsetup:
Idle:
Pin = 20V * 35mA = 0,7W
T(TPA) = 42C
T(amb) = 16.5C
Loadtest:
VCC: 20V
Signal: 2.2Vpp sinewave 30Hz
Gain: 26dB
Pin: 20V*1Arms=20W
1 channel
Load: Monacor SP6/100/8R freeair
Wiring: 2x50cm parallel cable
System temperatures after 10 minutes:
T(TPA) = 65C
T(330u) = 35C
After 20 minutes, the amp failed with no output and tripped FAULTZ
Restarting the test results in a FAULTZ-tripping when reaching ~200mA input-current. (still only on channel loaded)
The bootstrap-voltages measures 6.0V and 6.23V for this channel, while the other is at 6.23V on both bootstraps.
Testing the other channel with the same setup, FAULTZ is triggered between 0.5-1A system-current. (leaving the faulty one open)
Changes made:
- replacing the speaker with 10R resistive load: no change
- replaced bootstrap-capacitors: no change
- adding bootstrap-snubbers 220pF + 10R: no change
- adding ferrites at the output: no change
- adding full LC-Filter 10uH + 680nF: running good, no more FAULTZ tripping even when fully overdriven
Question:
Does the TPA needs the "reactive current" to work correct?
Did a transient due to reflection (caused by "hard-switching") killed the outputstage?
Do we need to limit the current via PLIMIT? (Will this work without reactive current)
As from the datasheet, LC-filter / ferritebeads, bootstrap-snubbers, current-limiting are optional.
------------------------------------------------------------------------------------------------
The TPA313xD2 modulation scheme has little loss in the load without a filter because the pulses are short and
the change in voltage is VCC instead of 2 × VCC. As the output power increases, the pulses widen, making the
ripple current larger. Ripple current could be filtered with an LC filter for increased efficiency, but for most
applications the filter is not needed.
------------------------------------------------------------------------------------------------
Regards,
Christian
