I've run into a problem when chaining 595's and using them to switch on and off DC-motors using FETs.
The motors are in parallel, and the first 8 can be switched on and off perfectly using the first shift register.
When switching one of the next 8 motors in the chained 595, the motors seem to go on very shortly but not strong at all.
When I bypass the shift registers and send a pulse to any of the FETs, the motors turn on just fine.
I've placed capcitors from the 5v to the gnd in front of the 595s, a large one and a small one. (20uF and 0.1uF)
Is there a problem with my shift registers?
Hope to get some help
I would put more decoupling capacitors (0.1uF) as close to the chips as you can get them. Those you have are rather a long way off.
Can you get any motor on the second shift register to turn on? I would disconnect the other motors and try just the second one. Check for grounds and also try swapping the chips in the sockets round. Does the fault follow the chip?
Even by disconnecting all the motors but one, the second register doesn't switch anything.
I managed to get everything working yesterday by connecting the shift registers to a seperate power supply instead of the Arduino 5V.
Maybe by placing the caps closer I wouldn't need a seperate power supply?
If I were to place the capacitors closer to the chips, how should I place them? It seems rather tricky to have a capacitor between the 5v and gnd since they are far apart on the chips.
The chips are working perfectly when I test them on LEDs. Today however, I tried duplicating yesterday's setup. Unfortunately I didn't remember or document it, and I seemed to have damaged the second shift register.
Before I damage more registers, here is how I think I had the working setup:
Circuit for the motors powered by an external 5v power supply.
From the same supply I used the 5v and the gnd for the shift registers.
From the circuit of the motors a gnd running to the Arduino.
From the circuit of the motors a gnd running to the Arduino.
I assume that this is the ground of the external supply.
How was the shift register damaged? Some output pins or would the whole thing not shift at all. It is hard to damage shift registers, over (or negative) voltage is about the only way. You appear to have diodes across the motor so that is fine.
