Pitting Switch Contacts

[BlueSmoke]

Hey all, new here - hoping it's okay to ask kind of a basic question.

I built a pneumatic robot using junk laying around including some electric car door locks to work the valves. It's all working, but I'm sort of whacking the door locks with a bit too much voltage, 20 instead of 12 - cordless drill batteries.

The door locks only run for a fraction of a second, the valves open and shut really fast so I'm hoping it's all quick enough that I don't end up burning them out.

I am however slowly cooking the contacts in the switches. I cracked one of the switches open and there's some pretty nasty sparks flying about when energizing the locks.

So I was wondering if there might be some easy fix to stop or minimize the sparking / pitting of the contacts in the switches?

Thanks.

 

[gophert]

You can add a capacitor or capacitor+ resistor in parallel with the switch OR capacitor + resistor in parallel with the load.

In the following link, your motor would be the rectangle (load) in the schematics.

https://www.illinoiscapacitor.com/pdf/Papers/spark_suppression.pdf

 

[dr pepper]

C = I2/10, interesting rule of thumb for cap selection.

The brush gear in the motors will be taking the hammer, but it'll work for a while.
You could also use a large (physically) choke to slow down the rise in current over time.

 

[gophert]

C = I2/10, interesting rule of thumb for cap selection.

I assume the capacitance for the formula is in microfarads.

 

[BlueSmoke]

Thanks so much for the help.

I'm guessing I'm going to add the resistor & cap to the load, that way I'm thinking I won't have to double things up on both sides of the switch when reversing polarity, going from opening or closing the valve.

Um, the formula on the Spark suppression website, I wasn't sure if they were saying C = I2 / 10 or C = I (squared) / 10

 

[rjenkinsgb]

It's C = (I ^ 2) / 10

I could not make sense of it either, so I found another site referencing the C.C. Bates formula.

 

[BlueSmoke]

Thanks for the update on the formula, not that I understand math in the least, but the values were comming up so small it got me thinking something was wrong.


Dang - should of stayed awake in math class

And I'm assuming as did gophert above that it's in µF

 

[gophert]

I was assuming the little 12v motors used in Honda's for door locks draw about 0.25amp so,
C = 0.25*0.25/10 = 0.0125 uF

I think GM uses bigger clunkier motors so maybe a half-amp.
C = 0.025 uF.

A resistor of 1 to 3.3 ohms should work.

Honestly, a 0.1uF has always worked for me with small motors. Add a little 1 to 3.3 ohm resistor and you're done.

 

[BlueSmoke]

Holly Cow, am I screwing up Ohm's Law here - I'm a software guy with a little bit of electronics knowledge.

These are aftermarket or 'General' purpose Door Lock Motor / Actuators for retrofitting older cars with manual plunger style mechanisms, measuring in at 1.8 Ohms, and the Drill Battery fully charged reads 20.4 Volts.

I = 20.4 / 1.8 = 11.33


I mean I tried grabbing hold of the darn thing just to see how strong it was, and the plunger ripped right out of my fingers.

11 Amps seemed really high to me, but it certainly matches up with all the sparking going on along with that wonderful ozone smell. I tried to measure the current draw, but the circuit is only energized for a blink of an eye, and well my Toys-R-Us meter can't catch such a fast current spike.

 

[rjenkinsgb]

11 Amps seemed really high to me

It is!

The running current of a motor depends on its speed and load; the only relation to DC resistance is that the current will be somewhat lower than that provides for when it is running.

The resistive current is the "stall current" with no rotation.

As an example, I have a little 12V motor here that draws 5A when held stationary (stall current) but less than 100mA when running freely.


With yours, the current at the instant the relay switches off is the figure you need to base the calculations on.

 

[BlueSmoke]

When pushing the switch, things get violent, I mean the valves operate in a wickedly fierce manner. And it all happens so fast, I'd have to say that the motors are almost certainly hitting there stall speed as the plunger reaches either it's full retracted or extended position stopping the motor dead in it's tracks.

When I couldn't get a current reading on my meter, I decided to play around with some fuses, 5, 8, 10 and 15 amps, the only ones I had. Everyone of them blew except the 15 amp, so it would seem that it's somewhere between 10 and 15 amps.

I've got another sort of a Toy piece of equipment, it appears to be a 'Teaching' version of an oscilloscope, and quite honestly I don't really know how to properly use the thing. Is there any chance I could set it up to catch a trace over time of the of the current draw?

 

[gophert]

What brand/model car did that thing come from?

 

[BlueSmoke]

Oh it's not for any specific make or model, it's a generic electric door lock actuator mostly used on older cars to convert manual locks to automatic ones. I had a bunch of them laying around, but you can buy them new on eBay for like 5 bucks. Unfortunately the specs for them don't seem to match with what I'm seeing or measuring.

Could be that the ones I have are a different version, not really sure.

**broken link removed**

 

[rjenkinsgb]

I've got another sort of a Toy piece of equipment, it appears to be a 'Teaching' version of an oscilloscope, and quite honestly I don't really know how to properly use the thing. Is there any chance I could set it up to catch a trace over time of the of the current draw?

Yes, you can.
You would need to connect a low value resistor in series with the motor, 0.1 Ohms or less, at 10A+, then capture the voltage waveform across that resistor.

The 2.22A value in the ebay listing is more reasonable for operating current, but it all depends how they work internally.

 

[BlueSmoke]

Cool, I'll blow the dust off the oscilloscope. Oh and when I called it a 'Teaching' oscilloscope, I really should of said a 'Student' oscilloscope. It's like one required / issued for an EE class, not exactly fancy or full featured.

I agree with the 2.22 amps being a more reasonable current, but going back to the fact that the locks are most certainly experiencing a stall load with every use, I could understand it blowing the large fuses.

The 12v motor example above, 100mA draw peaking at 5A stalled, that's like a 50 times jump in current (if I did that right). So I wouldn't think It's completely out of the question to be seeing a normal draw of 2.22A reaching 10 to 15 amps stalled. Well unless that 2.22A rating on eBay is actually the stalled current - heck I don't know.

The only capacitors I have laying around are electrolytic - and I'm guessing thats probably not gonna work since the current flow gets reversed between the open / close cycles of the valves?

 

[ChrisP58]

When pushing the switch, things get violent, I mean the valves operate in a wickedly fierce manner. And it all happens so fast, I'd have to say that the motors are almost certainly hitting there stall speed as the plunger reaches either it's full retracted or extended position stopping the motor dead in it's tracks.

When I couldn't get a current reading on my meter, I decided to play around with some fuses, 5, 8, 10 and 15 amps, the only ones I had. Everyone of them blew except the 15 amp, so it would seem that it's somewhere between 10 and 15 amps.

I've got another sort of a Toy piece of equipment, it appears to be a 'Teaching' version of an oscilloscope, and quite honestly I don't really know how to properly use the thing. Is there any chance I could set it up to catch a trace over time of the of the current draw?

You need to look at the datasheet for the fuses you're using to see what their time-to-blow values are. Especially if they're blowing very quickly.

For an example, the 15 amp fuse on the attached sheet should carry about 18 amps aftfor at least 1000 seconds before blowing. Or ~25 amps for one second. Or ~55 amps for 0.1 second.
Or, looking at it another way, if the duration of the current pulse is short, then the magnitude of the current needs to be compared to that fuses carry value for that time range.

Still, the better way is to use a oscilloscope. Hopefully it's digital storage one so that you can capture a single event that's otherwise too fast to see.

https://www.littelfuse.com/~/media/automotive/datasheets/fuses/passenger-car-and-commercial-vehicle/blade-fuses/littelfuse_mini_datasheet.pdf

 

[BlueSmoke]

Right now I'm using typical automotive blade type fuses, thought about trying 'slow' blow versions just to see what happens.

Not sure, but the voltage I'm using could also be screwing with things. I know when a car's running the alternator can be pushing 14 volts or more just to keep the battery charged, but I'm putting more then 20 through the fuses and door locks which has me wondering if that's causing more sparks & pitting along with blowing fuses faster then normal.

I've got some Zener Diodes I wanted to add, trying to drop the 20 volts down a bit, but wasn't sure if they could handle the current.

Right now I'm using heavy duty switches with the door lock actuators, but eventually plan on running things from a PC controlled relay board, but not til I can cut down on all the sparking.

Killing a switch - not such a big deal, frying the relay board, a little more troubling.

Yea about the oscilloscope, chances are slim to none it's got any memory functions, but I'll have to see once I remember where I put it.

 

[JonSea]

Here's a picture of a similar lock mechanism I brought at a junk store.

At 12 volts, it draws close to 2 amps.

At 18 volts, my current-limited supply at 5 amps shuts it down.

You can't put a reverse-biased back-end diode directly across the motor as the polarity must reverse to change the direction but using a one across each switch position will help with the large current spike of the collapsing inductive field.

 

[BlueSmoke]

Dug out the oscilloscope, not sure it would be any help, even more basic then I remembered.

You can't put a reverse-biased back-end diode directly across the motor as the polarity must reverse to change the direction but using a one across each switch position will help with the large current spike of the collapsing inductive field.

I always thought of collapsing fields being issues with relays / solenoids, not so much for motors, I assumed that with brushed motors, when the field collapses (creating reverse induction) the brushes have already moved on from the energized coil and set of contacts on the commutator and had moved on to the next set. By this time isn't the electrical connection broken, preventing the induced current from heading back through the communtator & brushes

 

[shortbus=]

Both of the lock motors shown in posts #9 and #18 are GM style actuators.