View Full Version : Unipolar stepper motor with ULN2003?

08-07-2010, 11:43 AM

Last time I was designing like this http://www.eleccircuit.com/wp-content/uploads/2008/05/uln2003-control-stepper-motor-by-parallel-port.jpg. I attack this to computer before start computer. After start I write codes and run only 3 steps. I find that IC and motor are going very much hot, then I disconnect all connections to IC. It really was near to be burnt, if I do not disconnect. What was the problem in circuit? I think it may be in its ground or +12V wire. Is it?

Robin Hewitt
08-07-2010, 05:58 PM
So you driving pairs of coils, one forwards, the other backwards or both off? Curious configuration. Where's Irving? He explains these things much better than I do.

The individual darlingtons can sink half an Amp. You could double them up to get 1 Amp but the duty cycle won't be anything to write home about unless you have big coil impedance.

08-07-2010, 08:47 PM
oh dear... there are three problems, but not in the circuit as such...

1/ you are probably using the wrong sort of motor... this type of circuit only works with very small motors rated to work on 12v, not high power steppers of the sort you'd use on a CNC machine

2/ Not reading/understanding datasheet and the limitations of this chip

3/ the basic lack of current limiting in a unipolar design (which I suppose is a circuit problem but is related to problem 1)

what motor are you using? what is the DC resistance of each coil? and how are you trying to drive it (one phase at a time, or overlapping phases?)

17-07-2010, 10:53 AM
I have found the reason not the solution, it is that I am using the computer power supply. The +12V point has 6 Amp, as written on info sticker of computer power supply. But ULN2003 require only 0.5 Amp, thatís why my whole system was getting so much hot. But how to arrange the correct power supply (12V & 0.5 Amp). Which is cheep & easy source?

17-07-2010, 12:19 PM
Yes, you have found the reason, but the solution isn't a 0.5A power supply, unless you plan only to energise one coil at a time which is not efficient and means only full-step operation.

The correct solution to this is to use current limiting in the feed to each pair of coils. Your current limiter can be a big power resistor or an active limiter using a voltage regulator device. I've shown you how to do both approaches in the attached diagrams. Do the calculations for yourself. Which you choose to use will depend on whats available to you. Don't think you can get away with just a lower current power supply - at best it wont work reliably, at worst you will still end up frying your devices and motors...

The resistors in the first diagram are 18ohm 10W - they will get HOT, about 80degC surface temp so be careful - but they are designed to get hot assuming you use a proper power resistor. You can mount them on a heatsink to cool them if you desire.

In the second diagram the resistor are 2.4ohm 1W. The second option requires a heatsink for the LM317T devices of about 16degC/W, which can be done with a push-on heatsink or a piece of aluminium plate about 50mm x 50mm x 3mm thick (one each device).

04-08-2010, 08:49 AM
I find some thing written on my motor. That is:
60C 45
5 DEG 13 (ohm) MAS
You can also read more about this motor from http://www.compucanjes.com/products/view/14268.html. what is basic formula to set at any system.

04-08-2010, 09:57 PM
well thats a 24v 13ohm motor... on 12v it will require 1A/phase, which is twice what the ULN2003 is rated for. Basically you need to learn a lot more about how to calculate electrical and thermal parameters correctly else you are going to waste a lot of money on burnt out parts.

24-08-2010, 01:23 PM
Using power transistors like TIP120 is also batter. But would you like to suggest me a good professional design of stepper deriver of medium level? You may suggest more than one.

24-08-2010, 02:46 PM
well I wouldnt start with that motor... but I cannot recommend anything as it depends on what you want to do with the motor, what your budget is, and, given you are in Pakistan, what possibility you have to buy it.

25-08-2010, 10:00 AM
I would like to know some batter & professional driver IC or schematic design to work in CNC machine. For example TIP120 design can handle 5 to 6 amp, it is good to run some big motors but I am not assure whether it is more professional or just for experiment.

25-08-2010, 10:39 AM
OK, as I said before you need to learn about thermal parameters. A TIP120 running at 6A has a saturation voltage of 4v, thats 24W dissipated. Each transistor needs to be mounted on a 2.5degC/W heatsink to stay inside its thermal parameters. Thats a lot of expensive heatsink. So a TIP120 is NOT a good choice for a stepper motor. Better choices are very low 'on resistance' MOSFETs such as IRFZ44 series, these will dissipate 0.2W at 6A and dont need any heatsink, but are more complex to drive and need more understanding of the dynamics.

Unless you have a good understanding of electronics and circuit parametric design, and access to good PCB manufacturing facilities, designing and building a good stepper driver from scratch is not possible, except for very tiny motors that are never going to be of any use for a CNC machine. You would be better spending money on a commercially available driver, it will be cheaper than building it in the long run.

If you insist on building your own, then look at pminmo.com (http://pminmo.com/) where there are several tried and tested designs based on various chips as well as PCBs you can buy.

A final word of advice... design your CNC machine first, then work out what motors you need, then decide on the drivers... then go back and refine your thinking to what you can afford... doing it any other way will end in disappointment...