0
\$\begingroup\$

I am using an Audio Module. The schematics of the module can be found below. The module uses a audio amplifier (TS4990IST) and STM32F0 MCU.

I found out when powering up the module, the module start heating up. In my module, capacitor C10 is not soldered. Can someone help me to rectify the issue?

enter image description here

EDIT: I have been reading the datasheet of TS4990 and found out it may require Cin capacitor at the input (Page 4). Find the screenshot of page 4 below:

enter image description here

Looks like the manufacturer of these modules has forget to add the Cin Capacitor and may be that's why the module is amplifying DC voltage and hence heating up the module. Am I right?

\$\endgroup\$
8
  • \$\begingroup\$ Check for spurious full scale oscillations and add C10 Often RC snubber is used across output. <10 Ohms 0.1uF. Layout and wiring may have some effect as well as supply noise. \$\endgroup\$ Sep 6, 2018 at 21:00
  • \$\begingroup\$ Could it be related to absence of input feed resistor? \$\endgroup\$
    – abhiarora
    Sep 6, 2018 at 21:01
  • \$\begingroup\$ R7 is feed input that affects gain (x1) thus phase margin or lack of it can cause spurious oscillation and self heating \$\endgroup\$ Sep 6, 2018 at 21:02
  • \$\begingroup\$ Sorry, I mean input feed capacitor? The capacitor should block DC but it is not present in the module. \$\endgroup\$
    – abhiarora
    Sep 6, 2018 at 21:09
  • 2
    \$\begingroup\$ Is your source AC blocked? \$\endgroup\$ Sep 6, 2018 at 21:17

1 Answer 1

1
\$\begingroup\$

The problem maybe Phase Margin is negative at unity gain (See Spec Fig 2) unless loaded with 560pF. See table... 65 deg unity gain phase margin with 8 Ohms 560 pF.

Check DC output across the load. If any DC, add DC blocking to the interface if your source is DC coupled with offset. The 1uF draws power for 20~35 ms after Standby disable. This can be cancelled by RC matching input cap to rise at the same RC time constant.

THis is a transient heat problem with coming out standby in this design on a 8 Ohm load. Starting to heat up is OK. Continuing to heat up is NOT OK. If not continuing, it may be a non-problem.

  • If OK but you want to reduce the trasnsient heat effect, change 1uF to 0.1uF. but note"In the high frequency region, if Cs is lower than 1 µF, it increases THD+N and disturbances on the power supply rail are less filtered."

You have 3 choices;

  • increase gain to 20dB or x10 with Rf/Rin ratio.
  • Add 500pF to each output. which at unity gain does not affect closed-loop BW significantly.
  • some compromise of above. R5=100k gain ~x5 14dB , phase margin >30 deg stable then test Step response for overshoot and add load Cap up to 560 pF

It seems like the OEM designer did not figure out these issues.

\$\endgroup\$
9
  • \$\begingroup\$ Source is analog voltage generated by DAC of STM32F0 controller. Do you think it requires DC blocking caps? \$\endgroup\$
    – abhiarora
    Sep 6, 2018 at 21:36
  • \$\begingroup\$ It depends on your DC noise and speed of V+ rise time, standby disable etc ,, temp rise after 1 minute and application BW down to DC? \$\endgroup\$ Sep 6, 2018 at 21:37
  • \$\begingroup\$ Is it hot 1 minute after startup? \$\endgroup\$ Sep 7, 2018 at 0:56
  • \$\begingroup\$ It keeps on heating. Now the board has stopped working. \$\endgroup\$
    – abhiarora
    Sep 11, 2018 at 16:25
  • \$\begingroup\$ OK the it is oscillating and needs a low Z high f snubber and possibly better heat spreader. \$\endgroup\$ Sep 11, 2018 at 16:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.