Maker Pro
Maker Pro

24V Electromechanical Relay-sparks but doesn't switch

You can probably tell from the title and the description below that I have little experience in electronics--thank you for taking the time should you choose to help. : )

I have a bar that raises and lowers electronically. It was made in the 90's and the manufacturer no longer exists. It is powered by a windshield wiper motor. When the button to raise or lower the bar is pressed, there is frequently (roughly 8 out of 10 times) only click and nothing else. The other 2 times it operates normally.

I opened it up to see if I could determine the issue and there are two electromechanical relays on the circuit board

these-
https://www.alliedelec.com/product/omron-electronic-components/g2r-1-dc24/70175791/

One switches when the bar is raised, and the other to lowers it.

It appears that what is happening is that when the relay fails to operate properly all I get is a small spark between the contacts (red arrow). The magnet seems to fail to completely close the EM switch 80% of the time, and I just get a small arc. I did open up the covers of both relays and clean the contacts with a needle file which only seemed to helped a bit.

Board.jpg

Is it likely that replacing the relays will correct that issue? Or am I on the wrong path?

Thanks again for your help and I apologize for my lack of proper terminology. : \

Sam
 

Harald Kapp

Moderator
Moderator
Before replacing the relays I'd check the control voltage to the relay coils. Do you have a voltmeter or multimeter to do so? If the control voltage to the coil fails, the relay may be perfectly o.k..
While you have the cover removed you may also check whether the armature (the moving parts) of the relay moves freely. You can do so manually, make sure no power is applied to the board, though.
 
I did check the relays and they seem fine mechanically (i.e. the do move freely when operated manually.)

I do have a multimeter, yes. I could check that-but I think I'd have to work at getting to the underside of the board while it's powered up. I'll see what I can come up with. : )
 

Harald Kapp

Moderator
Moderator
The armature moves freely but the contacts fail to close 80 % of the time? That doesn't sound like a failure of the relay itself. Check the coil voltage as suggested, even if difficult to reach, and check the solder connectionsof the relay. A bad (cold) solder joint could produce the same effect.

There's also the issue of the unit being from the 90's. After more than 20 year in operation the electrolytic capacitors may have failed. Sometimes visible as a bulging of the top, but sometimes invisible, just dried out. Insufficient capacity may also cause the voltage to drop too, thus preventing the relay from operating. If you have a capacitance meter (or capacitance range on your multimeter), check the electrolytic capacitors (the large cylinders). You'll have to de-solder them from the board to do so. Measuring while soldered into the circuit will not give reliable results. Check the measured value against the rated value. Replace the ones that have less than 80 % of the rated capacity.
If you can't measure them, replace them preemptively. The are not that expensive.
To select and buy new capacitors: capacitance should match, rated voltage may be higher, size doesn't matter, you can always bend the pins to fit the existing holes.
 
Sir SamGIII . . . .

Check out your RED arrow marked up relays data below . . . then do an ohmmeter test with the unit not being plugged into AC power.
The data shows 2 of its 3 contacts being closed when there is no power to the unit.
That would be contact 4 resting against 2 with no power applied.
When the relay coil gets powered up with 24V, contact 4 then swings over to 3 to give power to your units motorized aspect.
If you did find contacts 2 and 4 and confirmed their correct markings we will then concentrate on the contact 4 to 3 functioning aspects
of operationality, when 24V power is sent to ;pins 1 and 5..
Prep yourself a U-loop of insulated hook up wire, long enough to have its ends bridge between contacts 3 and 4.
Go to an end of the wire loop and cut an encircling ring through the plastic insulation about 1/4 to 3/8 inch inwardly.
Rotate the short insulation snippet as you work it to have it travel to the end of the wire . . .therewith . . . facilitating its removal
PLUS, its also tightening / compressing the wire strands.
Power up the unit and do the function that is supposed to activate that specific relay. Listen for the units armature to click, confirming the powering up of its coil and armature movement action.
Now either all will be fine for a cycle OR one of your abortive . . .no run . .or . .stops running conditions will ensue. If encountering that lucky situation jumper 4 and 3 and hold shorted across until the cycle completes and you hear the relay click . . .remove jumper.
Get set for another cycle where you will be metering the voltage across relay coil pins 1 and 5 to see what the voltage is being, when the relay clicks on.
We would be fault finding if that coil activation voltage is not up to 24 V spec.

ELECTRO-MECHANICAL RELAY FAILURE CONDITIONS . . . . .
In the case of near load limit use of a relay . . . . its phosphor bronze leaf springs and their end mounted contacts, can have one or two things occur.
  • 1st Situation . . .
The current passage and heating of the leaves excessively, can cause them to gradually lose their tempering and go ductile / limp.
Should you jimmy up the clear plastic cover, you can then get a mechnical evaluation of a unit.
If doing so, if you were then to manually press in on the armature plate you would see the #4 leafs contact initially touch #3 leafs contact and them having a preloaded spring tensioning pressure between the two units contacts BEFORE the armature plate then fully stops, at its mechanical limit of movement.
  • 2nd Situation . . .
On day one, that relay has full contact utilization of ALL of the surfactal area of both contact faces. With time, an insulative oxide can form to decrease available contact area. That means resistance and resistance in that decreasing conductive area means heat, and heat can speed up the odidized area with incremental heat concentration. At some heat level it thermaly conducts to its supportive leafs and starts upping their heat level and their progressive degradation.
At some weakening point, that " preloaded tensioning " dissappears and then has them in a touching /or maybe/ not touching condition. Thats then being the contacts " arc city " time .

If the 24V relay coil power test reveals no fault, and then If things start working after a start / run 'failure" when you use the jumper wire . . .that sort of confirms it as a relay failure . . . . IF . . . . . no floating / fractured solder joint connections in that power loop are involved.
Should you end up with power definitely being "relayed" to the lifting unit . . . all of my Windy- Shields motors had brushes that wore down and out.
But . . . . . I never kept a vehicle long enough to encounter that.





Ringo . . .Please don't take my . . . . . .

PHOTOGRAPH . . . . .Omron Relay
. . . . . away !

upload_2021-2-12_5-18-39.png



Thaaaaaaassssit . . . . .

73's de Edd . . . . .



At my advanced age, "getting lucky" means my being able to find / remember where I parked my car, in the gigantic Wally Woild parking lot.

.
 
Last edited:
i would suspect there is no continuity due to frosted over contact surfaces from all the arcing, such as seen in this example. The surfaces are a precious metal to ensure low resistance, and using a file is probably too coarse for "cleaning" the surface. There are special contact cleaning tools but it is usually hopeless and replacement is the better option.

XX5wnbZb.jpg
 
Last edited:
Thank you so much gentlemen!

I am working on getting the bottom of the board accessible while it is connected so I can do the suggested tests and when I removed the transformer I found this underneath...

Close.jpg

The board is discolored and looks as though there was quite a bit of heat that seems to be centered around the square part in the center of the pic.

I may be out of my depth here, but I'm going to keep at it--I'm pretty sure I'm capable of de-soldering and replacing some components if I can determine where the problem is... : )

Again, thank you for your guidance and wisdom!!

Sam
 
The 7812 is a 12V voltage regulator and the metal tab with a hole in it is supposed to be mounted against a metal heatsink (such as a piece of aluminum) using a screw thru the hole to hold it tight (to wick away the heat). There is a wire-wound power resistor between the VR and the big capacitor that has gotten hot and caused the paint to fall off. Probably not a problem with either part, you can add a heatsink to help.

What did the contact surfaces look like in your relays--were they dark and pitted such as i showed above, or clean and shiny like the unused contacts in your relays?
 
In the highlighted bits of the pic below,
1) is that a scorch mark on top of the cap?
2) is the resistor open circuit?
Whirlpool.jpg
 

Harald Kapp

Moderator
Moderator
The resistor definitely looks bad (insulation lost) but may be in perfect working order. You'll have to de-solder one side to measure its resistance. It should probably be below 10 kΩ, maybe a few Ω only.
The voltage regulator has this pinou:
upload_2021-2-13_20-14-15.png
When you measure from Output to Ground you should get 12 V.
It looks like there is a copper area on the PCB below the case of the voltage regulator. Such an area is usually meant to serve as a primitive kind of heatsink. This will work only, if the regulator is firmly connected to the copper area, not when it is floating above it as in your picture. This may be improved by drilling a hole in the PCB and securing the case with a screw and nut. Take care, however, that you don'T drill through any underlying copper traces of components ;).
 
OK! Sorry about the slow reply--it took me a bit to get at everything and get it set up so I could test things. : )

This is what it looks like installed (if that's of any interest... : )
Overview.jpg


I replaced both relays--they were inexpensive and I figured I might as well rule them out while I was at it. It seems to have helped slightly, but they still fail to switch probably 50% of the time.

Harald - I did check the voltage regulator and I am getting exactly 12V between the ground and output. I did also secure the regulator down to the board with a screw--there was, in fact, a heatsink already built into the board underneath. : )

Close.jpg


So I checked the control voltage to the new relays (the voltage between pins 1&5 when the relay actually works -- thank you Edd!) and I'm only getting 20.6V on the up relay and 21.5V on the down relay. I'm assuming that's my problem at this point?

I have not yet unsoldered / tested any capacitors or resistors.
I see the one damaged resistor that Alec pointed out.
And I think I see 4 capacitors? The huge black one, the two smaller black ones to the left of it, the blue one and the orange one in the upper RH corner. Is there a logical place to start? Or should I just replace them all? I don't really mind spending some money on it--it's more my de-soldering skills that's driving my desire to start with the most likely culprit. haha!

Also--should I be testing the output of the transformer, or is that unlikely to be an issue?

Thank you again gentlemen! I am so appreciative of your knowledge and willingness to take the time to share it!!

Sam
 
Sir SamGIII . . . . . .

I am seeing your raw ~24VAC coming in as the two red wires with painter tape marking and negative 24VDC coming out as the anodes of the bottom two
P6006 power diodes and the positive 24 vdc as the centrally interconnected common cathodes of the top pair of diodes.
You can see the negative foil crossing the board at the bottom. The + rail is hidden on the other board side .
Surely the LARGE black E-Cap is being the filter for that major motor drive supply. Pass us its capacitance and voltage rating.

Your relays coil pin 1 to 5 readings differences just might be due to a motor drain lifting action on one relay pulling down the whole supply voltage, while the other relays enabling a lowering action, which is like coasting downhill on its power requirement. Unless the motor is now mechanically disconnected from needing to lift or lower.
In either voltage situation, I think that either are being enough to produce a solid THONK mechanical action and we know that the relay contacts are now being pristine.

Weighing heavy on my mind, is that the units DC motor brushes are worn down, and now being at an intermittent stage of tensioning into the motor commutator.
If being my situation I would track down the two wires that go to the lift motor and connect a DVM in 24VDC range across them and then power up and run a cycle.
If it reads real erratically during a cycle, blame poor erratic brush or commutator contact action.

Could we get a PCB shot of the foil side of the board to possibly see inter connections to the hot metal film resistor, it could be that it is a series dropping resistor to initially drop the main 24VDC supply on downwards to 18 or so for input into the 7812 reg.
Also the full BZX # of the BLUE zener diode near the 3 term regulator.

73's de Edd . . . . .

Had gender change today . . . . not on purpose . . . . . it's just being that dang cold outside !
 
Ok, here's the PCB top and bottom...Top.jpg
Bottom.jpg

I did attach leads to the motor to monitor it on a normal cycle--it read 36.5V on the up cycle and 35.5V on the down cycle and it was quite steady as the motor ran. Not sure why it's not 24V? Here's how I had it hooked up.
MotorDVM.jpg

This is the big Cap
LargeCap.jpg

And the blue diode (was the best shot I could get--looks like 85C if that's possible?)
BZX.jpg

And finally the blue Cap & what looks like damaged resistor?
20210221_122401.jpg

Hope that helps!
Thanks Edd!!

Sam
 
Top