2

I am trying to use only 1 power supply instead of 2 to power Raspberry Pi and motor, but when I turn on the high current (>10A) motor, it will lead to the Raspberry Pi restarting. I guess it is something about the current overload protection? Below are the wiring I have done.

1st wiring using two 12V power supply. Everything works perfect.

enter image description here

2nd wiring using only 1 12V power supply. The Pi would restart once I turn on the motor.

enter image description here

Any suggestions? Any module I can buy or any rewiring I can do to prevent this case? Thanks.

Pak Ho Cheung
  • 123
  • 3
  • 1
    Add a large capacitor and a diode just upstream of the 12 to 5V Converter. Diode is to ensure current only flows from supply to cap. cap is to hold charge for the moment supply gets hit by motor. Of course it wont work if steady state of power supply output while running motor +a few amps is less than 5V + diode drop. – Abel Oct 24 '21 at 10:58
  • 1
    @Abel any model of diode you would suggest? Thanks – Pak Ho Cheung Oct 24 '21 at 11:06
  • 1
    what's the model of 12V to 5V? I'd get a rectifier diode that covers ~2X the max 12V amperage that thing could draw... but pi probably doesn't need that. Just pulling from what ai have now, I'd use SR560. – Abel Oct 24 '21 at 11:38
  • @Abel I’m using XY-3606 – Pak Ho Cheung Oct 24 '21 at 12:50
  • Use a Schottky diode - one with a low forward voltage drop (see answer below). – Seamus Oct 24 '21 at 20:53
  • Simple - whatever power supply you are using is inadequate. This is not a Pi problem. – Milliways Oct 24 '21 at 21:51
  • @Seamus a voltage drop upstream of the 12-5V shouldn't be a big deal. you're dropping it to 5V later anyway. – Abel Oct 25 '21 at 12:29
  • @Seamue and Abel Thanks for all the effort and long answer. I need some times to digest and do some little testing and searches. Will come back later. – Pak Ho Cheung Oct 25 '21 at 19:32
  • @Abel: The diode and cap should be on the RPi-side of the 12-to-5 regulator. Putting it on the 12V side is not the way to do this. – Seamus Oct 27 '21 at 07:12

1 Answers1

1

You asked for suggestions. Perhaps it would be better if you first understood the source of the issue you've described?

Answer:

Your issue is caused by inadequate load regulation in your power supply. Here's a short video that explains what's happening, and shows one way to test a power supply's load regulation performance.

Your issue is exacerbated by two other factors:

  1. RPi's tight input voltage tolerances, and
  2. Inrush currents to your DC motors

You need more information to move toward a resolution in an objective fashion, but unless you have some decent test equipment, you'll have to sort this out through a trial-and-error process. The good news is that you'll learn some things!

The simplest approach is to stick with the two-supply-solution shown in your question. However, the VNH2SP30 drivers have the capability to reduce the inrush current drawn by the motors; managing the inrush current may be enough to get you to a workable solution.

You should also understand that these power-induced restarts may corrupt your filesystem. Therefore, it's especially important that you maintain a reliable backup.

That concludes my answer - following are some more detailed comments and suggestions that may or may not be useful:

Other Comments & Suggestions:

  1. Comment: Your power supply (XY-3606) apparently has no published specifications. This is unfortunately typical of the electronic junk that is sold in our mainstream retail outlets today. As long as we continue buying this junk, vendors and manufacturers will continue to sell it.

  2. Suggestion: Low cost power supplies for hobbyists are available from more responsive (and responsible) sources. Ask questions before you buy, and insist on documentation and product specifications. Don't buy a pig in a poke.

  3. Comment: Raspberry Pis have fairly stringent tolerances for input voltage. The "official documentation" is fairly sketchy on this point, but lists 4.63V as the "warning level". A more informative discussion of input voltage requirements is in the RPi SE Q&A, "Powering the Pi 4". Much has been written on this subject, most all available for the cost of an Internet search.

  4. Comment: Your project may exacerbate the RPi's tight input voltage requirements. This, due to the fact that dc motors draw transient inrush currents that may be several times the value of their rated steady-state current. In other words, the motor's inrush current may be so large as to exceed your power supply's load regulation specs (which aren't published, so yeah - double trouble).

  5. Suggestion: Hopefully you can now see that you are dealing with several unknowns - any one of which could account for the RPi shutdown behavior you've seen. An oscilloscope would allow you to see the depth, and the timing, of the voltage transient on the output of your power supply. This would inform you as to the energy storage required, and whether a diode-isolated holdup capacitor might be useful.

    We can tell you that the energy storage available from a capacitor is straightforward to calculate (W = (1/2) * C * V2), but a measurement (or some trial-and-error) are required for an accurate estimate of how much energy is needed. For example, a voltage transient that dips to 1 volt for 20 msec requires far more energy (and a larger capacitor) than a transient that dips to only 4 volts for 2 msec.

    The diode is important also as it consumes power (due to its voltage drop) as it conducts current from the supply. In general, Schottky diodes are best for these applications.

Seamus
  • 21,900
  • 3
  • 33
  • 70
  • Thanks for the answer. I was waiting for my pi replacement, so it takes a little long. I finally checked I need to use around 700mA current, so I picked a SR140 diode, which is good for 1A and 40V. Thanks again – Pak Ho Cheung Dec 20 '21 at 19:57