Darlington array blow out

I am driving a .8 amp stepper motor through a darlington array. The motor is 1.8 degrees per step, and 12 volts. I am constantly blowing out my darlington array ic's, which i suspect is because they are rated for only 500 mA. The trouble is i can not find an array that is rated for more than 500 mA. I am also getting problems with the step motor causing current fluctuations to an lcd which is in the circuit. I suspect that this is due to lack of filter caps. The unit is battery powered (12v gel cell). Anyway, the darlington arrays seem more likely to blow out as the number of steps increases. I am also using rather thin wire to drive the step motor but idk if this would put extra stress on the array.

Any help as to a work around or solution would be awesome, I am avoiding driving each step motor coil with a transistor, or redesigning my circuitboard, but the ic is in a socket so replacing it is no problem.

Thanks!

As you have figured out you need to get darlingtons rated more than your motors require and if it were me I would at least double the rating. You could get mosfets instead. The steppers should also be driven from a seperate power supply to minimize the power fluctuations and caps between the two wouldn't hurt. Good luck with your project.

Thanks for the reply. I am going to look into mosfets because I can not find and darlington arrays with anything other than 500 mA, as wierd as that sounds. I am also gonna chat with a friend who is an EE. I will post what I find out. thanks again

You don't mention which darlington chip that you are using, but I have paralled that outputs on ULN2803 to get more than 500mA per output. Be sure to mind the heat generated, you may need a heat sink.

L293B can do 1A on each of its 4 output pins and needs thorough heat sink construction... http://www.datasheetcatalog.com/datasheets_pdf/L/2/9/3/L293B.shtml http://www.newark.com/jsp/search/browse.jsp?N=500003+1001891&Ntk=gensearch_001&Ntt=categorynumber83704&Ntx= http://www.newark.com/89K0727/semiconductors-integrated-circuits/product.us0?sku=stmicroelectronics-l292&_requestid=5393

a fat condensator and diodes might help, too... http://en.wikipedia.org/wiki/Diode#Over-voltage_protection

-Arne

L293B can

the 293 isn't a darlington; it's an H-bridge driver.

(That IS one possibility - use half of an H-bridge driver. They frequently include all sorts of features including current limiting and over-current protection. Pricey, though.)

Why not make your own Darlington with a power transistor and signal transistor. Then you could make it at what ever power you want.

I am using ULN2004. Ahh yes, I thought about using parallel channels, but forgot to mention that. If that will indeed work it would probably be the easiest solution. The wikipedia entry is also very helpful. It specifically cites the de-energizing of the coils in a step motor. This is something that I forgot about, but had to do back when I built my coilgun (http://www.tyconpowered.com/projects/?content=4), although that was a little higher current :) . From experience the diodes should be hooked up in parellel with the motor coils and ground, but backwards polarity (positive side on the diode goes to ground)?

Yes, they need to be connected cathode to V+. It also helps to use a schottky (fast) diode.

Good Luck

the diodes should be hooked up in parellel with the motor coils and ground,

No put them across the coil with the cathode pointing at the +ve side of the coil. This shorts out the reverse voltage you get as the magnetic field collapses when you remove the power.

I had a breakthough while talking to one of my EE friends. He says he is pretty sure you can stack or "piggyback" the darlington arrays on top of one another in order to form a parrallel circuit. This would save me alot of resoldering. It addition I would also install the schotty diodes.

diodes might help

The ULN200x drivers have built-in diodes that OUGHT to be sufficient.

You can add heat-sinks to DIP chips; that might help (if they're getting hot...) Sometimes all you need is a bunch of copper connecting to the ground pins...