Hi everyone,
I'm working on an autofocus system for my microscope camera. Everything works fine, I used an ULN2003 as I had a bunch of these laying around. But my unipolar stepper motor draws around 650mA which exceeds the maximum of 500mA for the ULN2003 and turns it into a bbq after a couple minutes. It would be great, if there was a different IC rated for 1A with the same pin layout, so I won't have to rewire my pcb. Does anyone know a suitable ULN/etc. ? 36V btw.
kind regards 
You can use bipolar drivers(4 wires) with any stepper motor that is 4, 5, 6, or 8 wires, whether its unipolar or not. I don't know of anything that has the same pinout as ULN2003 with higher current.
Motor wiring(Same coils, just different wires, some connected internally and some you have to connect yourself to simulate 4 wires. You can always connect wires together to reduce the number of wires to 4):
that would increase the resistance of the coils.
It is common to wire stepper motor coils in series or parallel (hence the "8 wire" variety). With a series connection, double the motor power supply voltage to achieve the same steady state coil current.
yeah i'm running in unipolar mode on 36v now. increasing the supply voltage is no option.
Increasing the voltage would be an option, with a bipolar motor driver.
Please post a link to the stepper product page or data sheet, as it sounds like quite a beast (23 W per coil?). I imagine that you could cook on it.
I'm using a Martin KH42HM2-017. cant find specific data sheets.
but my coil resistance to common point was around 47 Ohms. and it stays cold.
If it stays cold, the code is probably turning off the coils when not actively stepping.
If you don't need the full torque provided by the motor at 36V, reduce the motor power supply voltage, to say 24V, and perhaps the ULN2003 will survive.
I do need the full torque unfortunately. I'm running full steps btw.
Did you consider using a few ULN2003 in parallel to share the load?
In my brushless motor controllers I see a whole array of parallel mosfets, which sparked the thought.
thought about that too. would still need some rewiring and a larger pcb. tried to fit everything in the smallest enclosure I could find. (i hate/love myself for doing that 
). might give it a try, if nobody comes up with a sweet ic swap solution.
Darlington's are bipolar transistors, not efficient=lots of heat.
Jump over that and learn how-to MOSFET where 1A is shallow-end Power. Only caveat is that FET's need to be ON or OFF to run efficiently (over 90%, like 98%) and circuit design has different considerations than BJTs.
MOSFETs got cheap since the 90's, are not as touchy by far but still can be ruined yet I have a bunch with built-in diodes for inductive loads, I spent $10 and got them for < 15 cents each.
Get help and circuits from the EE's here, you should be fine.
I know how to use mosfets. but that's not what I was looking for. there's 1.75A ULNs out there. but with different pinout. so I was hoping for one with the same pinout. that's all I been asking here for.
Good luck then. Maybe call DigiKey and ask? Been there, nice people, I worked in the same industrial park.
How many motor wires?
i'll try that. thx bud
worth a try. thx
So i stacked 2 of them and soldered them together to ensure good connections. it still gets hot after the same time. no difference at all. I even stacked 3 of them. no change. double checked the current draw. ~650mA.
Wow! Is there room for a strip of metal between with tabs sticking out? Or is that unwarranted?
PS, just a thought but with enough strip past the end and some current through the strip, the Thomson Effect could be used to actively cool those chips. I'm not a EE but I do know some physics.
