Relay and "Brownouts"

Other Hardware Motors, Mechanics, Power and CNC
1

Hey everyone,

I have an issue i cannot wrap my head around. I've tried searching for this issue and have come close but everything I've been finding is not quite the same issue.

Background:
The project is a glorified RC car made from a "Jazzy Select Elite" powered chair. It is powered with 2 U1 12v AGM batteries in series. Im using two dc motor controllers rated at 30A(way more than they are actually drawing) that is controlled with PWM and direction pins. Input is from an RC Receiver/Transmitter that is passed on to an Arduino Nano which then sends out the necessary motor control. That much is working just fine.

The Arduino and RC Receiver are powered by a 24V to 5V 3A regulator.

The problem:
Both motors have brakes attached that can be disabled mechanically via lever or released via 24V, up to this point i have just had them disabled mechanically. I recently added 2, 5V relay boards, to control the brakes when output to the motor controllers are >0. Testing without the brakes engaged was successful, both relays actuated when expected with everything else working. However, when i engaged the brakes mechanically the Arduino would reset as soon as one of the relays actuated, sometimes after 1-1.5 seconds. The relays are powered from the 5V regulator directly and not through the Arduino's 5v out.

My first assumption was that i was pulling too much current from the 5V regulator or causing a voltage drop that caused a brownout, but if this is the case then why does it not happen when the brakes are disengaged? I also checked the regulator and voltage is steady when this happens.

My second assumption is that the brakes were causing a voltage drop great enough to cause the regulator to go wonky. Again I checked the regulators voltage while this was happening with min/max and it didn't so much as flicker. The 24V side saw about .1-.2V drop for a brief moment but nothing i would think could cause this.

Some more information:
The brakes draw ~.55-.65A (measured) @ 24V each, this is consistent with the windings reading 50ohms.

Relay coils are actuated with a digital HIGH from pins 7 and 8 for their respective brakes. No issues when there is no load on the contactor side. It is also worth noting that these relay modules are junk, two of the five were DOA. and it looks like I'm not the only one with that issue.

Hope i explained everything well enough. Anyone have any ideas on what i should be checking? Or if there is a better way to control the motor brakes?

Relays
Regulator

2

You did ofcourse already use a (1N4004) back-emf diode across the 24volt brake solenoid.
Leo..

3

yikes, no I completely forgot about that Wawa. I'll pick up a few diodes this weekend to add in before testing further, good call.

However my understanding is this will prevent back current after the supply is shut off, would not having the diode affect the Arduino when the brake is still drawing current?

EDIT: I just rechecked my testing and of course you were right. I was only energizing the relay briefly before, this time i kept the relay energized and sure enough, no reset until the relay was de-energized. I'll try to remember to post results after i add the diodes in as well.

Thank you very much for the help.

4

Without the diodes you'll get very heavy arcing on the relays, risking welding them shut, these brake solenoids
are massive inductances (may easily be enough to risk a lethal shock - never mess with large powerful solenoids without back-EMF protection). Also operating solenoids this big without back EMF protection risks damaging
the insulation on the windings and wrecking them, as voltages of several kV can be generated.

5

Just realised I advised a 1N4004.
You might PWM that brake solenoid.
In that case, you should use a schottky diode.
The 1N5819 (common/1Amp) should do.
Leo..

6

I tried a 1N4003, it did help but I'm still getting resets intermittently.
I assume your recommendation of a Schottky diode means the 1N4 is not switching fast enough?

Thanks guys.

7

guamish:
I assume your recommendation of a Schottky diode means the 1N4 is not switching fast enough?

Correct.
A normal diode, like the 1N400x, switches 'on' as fast as a schottky diode, but switches 'off' slower.
Leo..