[SOLVED]How do I calculate the power dissipation for a ULN2003A Darlington array

[SlowBro]

How do I calculate the power dissipation for a ULN2003A Darlington array? Is it Vce(sat) * Ic?

https://www.seeedstudio.com/document/pdf/ULN2003%20Datasheet.pdf

[StillTrying]

Is it Vce(sat) * Ic?

Yes, assuming they're fully switched on, the data sheet gives the typical saturation voltages 100mA to 350mA, 0.9V to 1.3V.
...add a few mW for the resistors and 1st transistor.

[SlowBro]

So for example I want to using multiple channels drive 20 LEDs at 20mA each for 400mA total. Dissipation = 1.6V (max) Vce * 400mA = 640mW. Did I get that right?

[StillTrying]

one of these channels drive 20 LEDs at 20mA each for 400mA total.

I was going to say it might be a bit higher with 400mA though the one channel, 550 to 600mW, but you've already done it.
Don't forget to add the other channels. I can't work out the Max. allowed dissipation from that data sheet.

[SlowBro]

Yeah, all channels combined it will be pushing the Rth j-amb max but the on time will be brief so I feel comfortable about it. Thanks!

[floobydust]

Don't forget the IC has a limit of 500mA total for all channels, and that the SO-16 has about twice the thermal resistance compared to a DIP-16 part, check it won't get too hot...

[StillTrying]

I don't know how many LEDs or channels there are.
These ULNs used to be used to drive the pins in ribbon dot matrix printers, I never came across a blown ULN.

[SlowBro]

Cool I wondered what they were made for. Anyway at the max current of 500mA and 5V the dissipated watts are 1.6V x 500mA = 800mW. The thermal resistance in the surface mount package is 120C/W and junction max is 150 so that leaves a buffer of 54C of ambient. I won't be driving it this hard and even when I do drive it will be intermittent so I'm comfortable. Thanks everyone!

[digsys]

I use FET based arrays for higher switching currents and way less losses. Very similar / same packaging and costs