120Views6Replies

Author Options:

Power supplies struggling, should I be limiting the amps going to these DC actuators? Answered

This project started as this instructables project 
https://www.instructables.com/id/Arduino-Controlled-Robotic-Drum

However, instead I'm using a relay board (sainsmart 8 channel) and arduino to switch power from 4 separate wall plug DC power supplies to the individual actuators. Each power supply is connected to 2 channels on the relay, powering 1 actuator each.

These are the car door lock actuators
Working voltage: DC12V
Current consumption: 0.15A-2.22A

I am using 1 x 12v, 4.5A power supply, and 3 x 9v 1.5A power supplies (since this is what was needed for the original instructables project)

However, I'm noticing that the actuators will sometimes not fire when the circuit is connected, and they seem to be limited as to how quickly they can fire. This is more of a problem with the 9v than the 12v power. The 12v, when supplying power to two actuators at the same time, will not work. I don't understand how the instructables project seemed to have no issue powering these actuators with three 9v 1.5 supplies powering 12 of these actuators (with shields that can only output max 1.2A per actuator).

So that got me thinking that the motor shield is doing some sort of power regulating for the actuators. I think a single actuator, unregulated, is gobbling up all the power to the point that the power supply is shutting off, or at least has no more power to supply to anything else.

Q:

- Do I need to build some sort of regulator for each motor to limit it's power consumption? If so, what kind of regulator?

- Should I just increase power supplies to higher Amps until it's not a problem?


Any help is very much appreciated.


Discussions

0
None
-max-

2 years ago

I had a friend who had simalar issues with jittery servos that appears to work fine (but with severely lacking torque) off the small linear 5V arduino regulator (or maybe a small UBEC regulator, I don't remember...), but his servo's acted very chaotically and jittery when powered from his huge 5V 10A power supply brick. with a suspicion of signal integrity instead of a supply fault, it It took awhile to get to the culprit of the problem, which happened to be the rather annoying foldback current limiting "feature" of his, (and basically all) cheap modern switching power supplies.

What is it?

Foldback current limiting is essentially when the power supply detects overcurrent, even for a microsecond, it shuts the output down to like 1/10th of the original voltage until and keeps it near zero until the controller detects that the overload has been removed. this is a pretty common implementation for short circuit protection in modern power supplies. Typical current limiting works by keeping tabs on both voltage and current simultaneously and keeping both within limits. If the load is light, then the regulator will regulate the output in accordance to the voltage, and conversely, if the load suddenly becomes a power hog, such that at the rated voltage of the supply, it draws too much, then the supply begins to regulate current instead, and the output voltage will sag to a point where the load can only draw the maximum current.

how to fix it?

The best solution obviously is to use "better," more robust supplies, ones that have "real" current limiting solutions, or perhaps no current limiting at all. If you are tight on funding or just real cheap like me, then you could also try some of the following things:

* Hack the power supplies (I did exactly this to bypass nearly all of the protection mechanisms in an old XBOX 12V 12A power supply, that way the supply does not shut down when I accidentally short the output of even just connect the output to a capacitor bank. I did this by soldering in some mod wires, one to completely short out the current sense resistor, which proved not good enough, and another to bridge a feedback path from a unmarked microsoft chip (which connected to 1 of 3 optocouplers) to +5Vdc effectively blocking its ability to send an error signal to the converter circuitry on the high voltage side of things.)

* Add large buffer (bypass) capacitors to help minimize and smooth out large current spikes. You may want to use a proper LC network, with large inductors as well for some LC filtering. Watch out for inductive spikes though. implementing *just* giant capacitors, especially if they are near the supply might upset the stability of the regulation, some supplies might not be happy with lots more capacitance on the output. Keep all the extra capacitance connected as close to the load(s) as possible.

* Use better wiring: Hopefully this is obvious to you, but make sure you have a good low impedance connection to the output of the supplies. Do not use thin or unnecessarily long wires. Using higher voltage lower current actuators may also be an option to reduce current and voltage sag across power cables.

0
None
Downunder35m

2 years ago

I only know my actualor struggle to open the door when the battery of the car is going below 11V.
So using an underrated power supply make no real sense to me.
Some big capacitors parallel to it might help to deal with the inrush current but I fail to see why 9V supplies instead of 12V one are used.

0
None
TrrlDownunder35m

Answer 2 years ago

I don't know either, it's just what the original instructables said. I will be replacing the 9v 1.5 with 12v 4.5's. However, even with the 12v supplies, two acutators cannot be triggered at the same time. It will just trigger one and the other will not fire, even though the relay will activate on both.

0
None
steveastroukTrrl

Answer 2 years ago

Make some voltage measurements. If one fires, and you see a significant voltage drop at the other. Check the voltage AT the power supply at the same time.

0
None
steveastrouk

2 years ago

Use heavier cables for the supplies to the motor shields.

Add BIG, say 1000uF electrolytic caps AT each motor shield, and try again.

0
None
Trrlsteveastrouk

Answer 2 years ago

Thank you.

I'm using 18 ga. speaker wire to connect the actuators. I am not using motor shields, just relays connecting the power supply directly to the actuators.

I will try adding caps to each actuator and let you know!