1
\$\begingroup\$

I am trying to integrate a dual h-bridge design in to my custom PCB (Arduino Nano with integrated motor driver) using the L9110s IC. However, I am struggling to get it working properly. The h-bridge does not provide enough voltage to drive my motors.

The input to the motor driver is a PWM signal from an Arduino (on the same PCB) which I can confirm works correctly when checking with my scope. The output of the motor driver is faulty though. Instead of outputting the full supply / battery (7.2V LiPo) voltage it sags to 3V and does not "respond" to the variations of the PWM input signal.

Even weirder, when I first install a new L9110s IC in my PCB and measure it without load it correctly supplies the full VBat voltage at the outputs. But after one or two input pulses from the Arduino, the IC seems to get damaged (?) and will only output 3V.

I have replaced the IC a few times now, and every time it will work for 1-2 seconds before it starts outputting the weird 3V signal. The IC did not get hot to the touch and I didn't smell anything funny.

I have double checked the L9110s Datasheet and the provide input voltage for the signals (5V) and the VCC (7.2V) should both be fine. I did my testing without any load so ampere shouldn't be the issue.

I have bought an extra dual motor driver with the same IC and schematic and it works fine with the same Arduino, input signal, battery etc.

I have used the following schematic (updated): enter image description here

My complete PCB is available online here but this is the layouting of the motor driver part: enter image description here

I am at wit's end and don't know what else to replace and measure. I appreciate any help and pointers in the right direction. Thank you.

\$\endgroup\$
  • \$\begingroup\$ Your schematic doesn’t appear to match your layout. \$\endgroup\$ – Andy aka May 26 '18 at 10:57
  • \$\begingroup\$ @Andyaka That is what I was afraid about. Can you point out the mistake you spotted. Thanks. \$\endgroup\$ – Tom May 26 '18 at 11:28
  • \$\begingroup\$ I can’t say it’s a mistake. You have capacitors across the motor terminals. The point I’m making is that you should show your actual schematic. Lack of ground plane strikes me as a big problem. \$\endgroup\$ – Andy aka May 26 '18 at 11:35
  • \$\begingroup\$ I updated to the final schematic that I ended up using. What is the problem with the caps across the motor terminals? Isn't that what the schematic demands? What is the problem with no ground plane? \$\endgroup\$ – Tom May 26 '18 at 13:38
  • \$\begingroup\$ No ground plane means excessive 0 volt bounce with motor current pulses. No localized decoupled caps makes this far worse. Even at the distance you have there can be ground and Vcc bounce that causes input voltages from your cpu to potentially be beyond acceptable limits. How come your circuit is still incomplete re the decoupled caps. This is no matter to be treated lightly. \$\endgroup\$ – Andy aka May 26 '18 at 14:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.