Pololu VNH5019 Motor Driver Carrier

[borked]

I’m using this board to drive LED strips with PWM:

VNH5019 Motor Driver Carrier

It’s meant to handle “Output current: 12 A continuous” according to the description but it’s cutting out from thermal protection within ~10 seconds at 12V / 10A / 50% duty cycle. I’ve tried adding some passive cooling but it’s only marginally increased the time before it shuts down. It also requires a power cycle before it will start up again which is really not ideal.

Any insight would be appreciated. From what I can tell it’s a simple case of the part being over-spec’d.

[Shaun21504]

They’re designed for an inductive load (motor). You don’t mention how you measured the 10 amp. If it’s with a multi-meter at 50% duty cycle you could be pulling 20 amp peak through the LEDs. How are you limiting the current to 10 amp?

[borked]

They’re not overheating at 100% duty cycle so I don’t think current is the issue. For the LED strips to be pulling 20A peak or otherwise I’d have to be putting a lot more voltage through and they’d be probably getting fried.

Wouldn’t LEDs be easier for the board to drive than motors?

As far as I can tell the board gets less efficient as it nears 50% duty cycle. I don’t know why that’s the case or why the specs wouldn’t mention that it overheats at far lower current at 50% than the numbers they have in the part description which I guess are for 100%.

[Shaun21504]

The only reason I can think that they would overheat at 50% and not 100% is that there’s a capacitor in parallel with the LEDs.

[Shaun21504]

Do your LEDs have a built in regulator/constant current driver. That could explain the issue as they typically include a capacitor across the input.

[borked]

Nah no other driver, it’s just the Pololu driving the LED strips directly.

[borked]

These are the strips: https://www.amazon.com/gp/product/B078JP8X99?ref=em_1p_1_ti&ref_=pe_1843280_414412250

[Shaun21504]

That counts out the capacitance. They just have a resistor in series with every few LEDs. Even with the long length I doubt it would be a large enough capacitance.

I’m assuming you don’t have access to an oscilloscope to see what’s happening? If not, you could try placing a resistor across the output in place of the LEDs. You’ll need at least 1.2ohm at 12watt for 12volts/10amps or equivalent combination of resistors. That’ll isolate what’s causing the issue.