0
\$\begingroup\$

I have recently bought a L9110s H bridge (pictured below), and while testing I accidentally burned it up. I'd like to know why, to avoid making the same mistake and if maybe I should try to get a refund.

Datasheet:
+=========+====================+=======+===========+=======+=========+
| Symbol  |     Parameters     |       |   Range   |       |  Units  |
+=========+====================+=======+===========+=======+=========+
|         |                    |  Min  |  Typical  |  Max  |         |
+---------+--------------------+-------+-----------+-------+---------+
| VCC     | Supply Voltage     | 2.5   | 6         | 12    | V       |
+---------+--------------------+-------+-----------+-------+---------+
| Idd     | Quiescent Current  | -     | 0         | 2     | uA      |
+---------+--------------------+-------+-----------+-------+---------+
| Iin     | Operating current  | 200   | 350       | 500   | uA      |
+---------+--------------------+-------+-----------+-------+---------+
| IC      | Continuous         | 750   | 800       | 850   | mA      |
+---------+--------------------+-------+-----------+-------+---------+
| IMax    | Current peak       | -     | 1500      | 2000  | mA      |
+---------+--------------------+-------+-----------+-------+---------+

My setup:

  • Motors: Generic motors from an old RC car. I can't see any number on them so I don't have much info. They came with a 100nF ceramic capacitor soldered on both ends. I have tried to connect them manually to the 5V and GND terminals of the power supply, and they worked just fine did not show any problems.
  • Power: ATX PC Power Supply (5A/3A current), with VCC connected to the 5V entry.
  • Situation: I connected the VCC and GND to a breadboard with 5V and GND from the power supply. When I connected A-IB to the 5V (to power a motor up), I heard a crackling noise, the two transistors started heating and emitting magic smoke, and they got so hot they actually fell from the chip. I screwed up real bad.

Schematic:

schematic

simulate this circuit – Schematic created using CircuitLab

There are two things I'd like to know:


  1. What was my mistake and
  2. Was my L9110s bridge defectuous? I know it was probably my fault, but I think it's weird that the soldering was so weak the transistors fell out when it overheat.
\$\endgroup\$
6
  • 1
    \$\begingroup\$ Welcome to EE.SE, Gus. There's a schematic button on the editor toolbar. It's very easy to use. Add in a schematic of your connection so we know what your setup was. Just use a box (down at the bottom of the toolbar) for the board and use NODEs (top of the toolbar) for terminals. Use the bulb symbol for the motor. Double-click to change properties of any symbols. \$\endgroup\$
    – Transistor
    Jul 20, 2016 at 19:07
  • \$\begingroup\$ A generic motor can be anything, and it seems like that you do not know the peak current of your motor. Maybe you exceeded the current limit. \$\endgroup\$ Jul 20, 2016 at 19:13
  • 1
    \$\begingroup\$ Where did you have the other inputs (A-IA, B-IA, B-IB) connected? \$\endgroup\$
    – brhans
    Jul 20, 2016 at 19:29
  • \$\begingroup\$ Highly likely you shorted GND and VCC. This could happen if you opened wrong switches in your h-bridge or if probes slipped while doing some measuring. Happens sometimes \$\endgroup\$ Jul 21, 2016 at 4:39
  • \$\begingroup\$ I've added the schematic, thanks for the heads up, transistor. It should answer @brhans question. \$\endgroup\$
    – Gus
    Jul 21, 2016 at 10:47

1 Answer 1

2
\$\begingroup\$

Unconnected inputs on CMOS devices are bad; they cause the internal transistors to be neither on nor off. It's possible that this caused the top and bottom of the H-bridge to be on at the same time, directly shorting through the device. This would explain why both chips failed when you only provided one input to one chip. You need to tie the unused inputs off to GND.

It's also possible that the motor draws more than 750ma - you should connect it up to the power supply via a multimeter in Amps mode to check this.

The solder melts at about 300C. Presumably the chips simply got this hot. I know it's cheap hardware but I very much doubt it was any kind of manufacturing fault.

It might be better to start with a current-limited power supply: e.g. 500ma USB charger. Run it for a few seconds and see if anything heats up unacceptably. It should at least take longer to self-destruct at reduced power.

\$\endgroup\$
8
  • \$\begingroup\$ I was planning to use this bridge with an arduino. When I set the other inputs to LOW, won't they be basically disconnected? Or does arduino's LOW signal works like a grounding? \$\endgroup\$
    – Gus
    Jul 21, 2016 at 11:33
  • \$\begingroup\$ I'm going to check the operating and peak current of the motor with a multimeter and will report back. Another question, can I use an arduino (instead of an usb charger) to power it up? I've been told that it's not safe to connect motors to arduino, so I'm not sure if it's safe to plug the H bridge to the 5V pin. \$\endgroup\$
    – Gus
    Jul 21, 2016 at 11:37
  • \$\begingroup\$ LOW should be grounded. Setting pins to input is similar to disconnection. It's not a good idea to route it through the Arduino because then overcurrent will destroy both devices. You need something with a limit. \$\endgroup\$
    – pjc50
    Jul 21, 2016 at 12:42
  • \$\begingroup\$ I've used a wall charger to power the motor, and after a couple of seconds (not more than 30), the charger stopped working. Now it won't even power LEDs. I haven't measured the current yet because I'm not too good with multimeters and I'm afraid I'll blow it up. How can I reduce this motor's current pull? Will resistors work? \$\endgroup\$
    – Gus
    Jul 21, 2016 at 18:14
  • \$\begingroup\$ You'd need a high-power resistor, or you'll blow that up as well. The best solution is just to get a motor with known max current draw that doesn't exceed the rating of the H-bridge. RC car motors can be in the tens of amps. \$\endgroup\$
    – pjc50
    Jul 21, 2016 at 18:28

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.