TDA4791LP - no output - SOLVED

anotherandrew · ‎2016-06-12

Posted on June 12, 2016 at 19:41

I've got a TDA7491LP design that I cannot seem to make function. I've followed the ''

Applications Circuit for class D amplifier'' example circuit (Figure 26 on the datasheet). I'm powering the amplifier with a 10V switchmode supply.

Some measurements from the board, all referenced to the circuit common (where SGND and PGND are connected):

MUTE (pin 21) at 3.5V
STANDBY (pin 20) at 3.5V
VDDPW (pin 19) and VDDS (pin 26) are both reading 3.2V.
PVCC is supplied with 10V
SVCC (pin 35) at ~10V
VSS (pin 36) reads ~6.5V
VREF (pin 34) at ~1.8V
SYNCLK output (pin 25) reading ~4V square wave of ~630kHz

I've got the inputs configured in single-ended mode; INNA and INNB are each conected through individual 470nF capacitors to my circuit common. I'm supplying a 1Vp-p 1kHz sine wave on INNA. I should be seeing a PWM waveform at OUTPA/OUTPN but both these pin sets are reading about 200mV; there is no PWM activity at all.

Now I'm a little confused; SVCC is called ''signal power supply'' and VSS is called ''3.3V regulator output referred to power supply'' -- Should I be referring MUTE and STBY to VSS's voltage?

Updated November 7, 2016

I wanted to update the thread and let anyone searching in the future that I have resolved the issue with ST's help (your FAEs are the best!) My issue was actually resolved months ago but I'd forgotten to come back here to update the thread.

I compared my board against the TDA7491LP eval board provided by ST and everything checked out; I then replaced my output filter section with the section from the eval board and as soon as I did this the fault condition went away. I guess my choice of LC filter was very poor and causing the output to appear as a short, which caused the TDA7491LP to protect itself.

I'm currently running the output section the same as Figure 34 in the datasheet; that is, I'm using a small ferrite (BLM31PG601) instead of an output inductor, using 1nF as the output filter capacitor, and using 2.2nF capacitors from each leg of the output to ground.

#tda7491

Nickname1623_O · ‎2016-06-12

Posted on June 13, 2016 at 06:21

hi Andrew ,

according to your measurement , looks like the device had been actived , 1) VSS ,VDDS ,VDDPW and Vref are available , 2) Synclk is right , 2 times of PWM frequency .

the problem was no PWM output from half bridge , please check the Diag pin status with 100K ohm pulled up resistor that should be connected to VDDS .

no you must refer mute and standby level to SGND (or power supply ground ) .

VSS had been generated from SVCC by regulator circuitry with 3.3V voltage drop, and it had been used for gate driver.

Thanks

YT

anotherandrew · ‎2016-06-13

Posted on June 14, 2016 at 00:15

Hmm, ok, so it sounds like it should be working. I'll wire up the DIAG pin (it is currently not connected as I did not need the feedback) as suggested (100k pullup) and see whether it's being pulled low or not.

Thank you for the assistance.

Martin Rowe · ‎2017-04-28

Posted on April 28, 2017 at 15:02

Hi,

I am also are having problems, however I am using the TDA7491HV (which I understand is pin compatible with the LP) and although it is working, the output inductors are getting very hot. Hot means ~85C with an ambient temp of 15C.

Andrew (thread starter) said that he switched to the 'filterless' design in Figure 34 of the TDA7491LP datasheet.

Does anyone know if this is feasible for the TDA7491HV? There is no equivalent to Figure 34 in the TDA7491HV datasheet.

Regards, Martin