Maker Pro
Maker Pro

ULN2003 paralleling current

I am seeing on many data sheets and websites that you can parallel the output of the chip for higher current. I need this higher current for a stepper driver I am working on. Here is the wiring that I am using: http://electronics-diy.com/stepper_motors.php how could I parallel the output going to the motor without blowing the chip as they have a 500ma limit.???
 

Harald Kapp

Moderator
Moderator
Connect:
Pins 1-5 as well as 12-16
Pins 2-6 as well as 11-15
etc.

To ease layout and wiring, you may want to swap more pins such that the circuit looks like

driver 1: Pins IN=1+2, OUT=16+15
driver 2: Pins IN=3+4, OUT=14+13

and so on.

Should be good for max. 1A.

Harald
 
Connect:
Pins 1-5 as well as 12-16
Pins 2-6 as well as 11-15
etc.

To ease layout and wiring, you may want to swap more pins such that the circuit looks like

driver 1: Pins IN=1+2, OUT=16+15
driver 2: Pins IN=3+4, OUT=14+13

and so on.

Should be good for max. 1A.

Harald

Sorry, I don't understand (I am a bit of a noob) I have tried stacking 2 chips and soldering together the corresponding pins, it did not work, all the power still went through one chip. is your solution any different. If so, please give me a more noob orientated answer.
 

davenn

Moderator
Sorry, I don't understand (I am a bit of a noob) I have tried stacking 2 chips and soldering together the corresponding pins, it did not work, all the power still went through one chip. is your solution any different. If so, please give me a more noob orientated answer.

yes his answer is different

he's telling you which pins of the chip to connect together to double them up :)
reread his post with that in mind

Dave
 

Harald Kapp

Moderator
Moderator
There are 8 drivers in the chip. Only 4 are used in the schematic you linked in your original post. What I suggest is using the other 4 drivers by paralleling them using the pin-numbers i gave. This will be good for ~ 1A.
If you need more power, you could stack chips on top of each other, (or equivalently connecting the pins.
Thi is not the best soulution, however. If you need more power than oned single chip can deliver, you better look for more powerful drivers.

Harald
 
There are 8 drivers in the chip. Only 4 are used in the schematic you linked in your original post. What I suggest is using the other 4 drivers by paralleling them using the pin-numbers i gave. This will be good for ~ 1A.
If you need more power, you could stack chips on top of each other, (or equivalently connecting the pins.
Thi is not the best soulution, however. If you need more power than oned single chip can deliver, you better look for more powerful drivers.

Harald

I have tried staking the chips (by equivalently connecting the pins) and using a 800ma power supply but within a few minutes, one of the chips was smoking, and the other was cold. This leads me to believe that staking the chips, for some reason, still funnels all the power through one chip.

PS: there is actually only 7 drivers ;)
 
Last edited:
Idea

I have had an idea, what if I only paralleled the output of the chips, and ran both chips on a single lower amperage power supply??? would that double the amperage???
 

Harald Kapp

Moderator
Moderator
Oh, I see, 7. Helps if I could count :)

Paralleling only outputs is not helpful. You need to connect the input signals, too.

If one chip was smoking but the other was cold, it seems that no power was drawn from the second chip. That sounds funny. You didn't forget to connect the power supply (pins 8, 9)? Otherwise it may help to add a small resistor (0.22 Ohm) in series with each output (before paralleling them). Thus any imbalance in the IC's outputs will be belanced.
Also you should tie the unused inputs to GND.

And what do you mean by
I have had an idea, what if I only paralleled the output of the chips, and ran both chips on a single lower amperage power supply??? would that double the amperage???

Could you post a schematic diagram of the circuit as it was modified by you?

Other than that I'm at a loss. Maybe a discrete solution using power transistors would be better http://itp.nyu.edu/physcomp/Tutorials/HighCurrentLoads .

Harald
 
IT IS ALIVE!!!!!!

:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:D:

I got it working!!! and just in time; the science fair is in 3 days!!! no time to enplane how now... but I will update in a few days....
 
Top