Anyone had any experience with relays that don't quite work - try to turn a relay on and it sometimes does and sometimes doesn't, or turns on for a bit and then drops out or oscillates intermittently. Seems the coil is faulty or something. Anyone heard of relays that do this and what might cause it.
Thanks
I've experienced it on a production run. Turned out the manufacturer's cleaning solution was bad. The contacts would stick after a number of consecutive toggles during test. Apparently there was a sticky residue on the internal contacts. Happened often enough to return the entire lot of relays.
Oscillation is bad firmware.
Silly question, but is the sustaining voltage above the drop-out voltage?
Al
Or the supply driving the relay can't sustain the output current at DC. Depending on the coil inductance and DC resistance, you can get weird voltage spikes accross the coil. The oscilloscope is your friend in these cases. Depending on the time scale, you might also have a supply or driver going into thermal overload.
Mark Borgerson
Sounds like the coil voltage isn't high enough...
Best regards, Spehro Pefhany
50 years ago we used lots of IBM relays that had a pick coil and a hold coil. In those days computers were large monsters with many vacuum tubes. But the relays were reliable.
You can make peculiar circuits involving diodes and capacitors and multiple voltage supplies that can effectively supply the extra 'pick' effort. Repetition rate is limited.
Herer is OK for this question. If you are having this problem with all of your relays then you have a number of possible causes. Others have alluded to Voltage or Current availability for the relay to hold in. You can confirm this by test-benching a relay from a small bench power supply that has current and voltage adjustments (the lab variety). See what your relays really want by way of drive.
You usually need a good driver transistor powered from a supply rail that has plenty of current drive behind it. Most relays (and solenoids) want
60% of their rated voltage to hold in but you need to give them their full rated voltage to ensure the pick properly. Limiting the current available from supply rails will also limit the voltage they see. This sort of thing you will see on the bench test.Bad batches like those alluded to by Moon-Shine do not happen that frequently but they can happen. So try a number of relays from different batches if you are able. If the relay has a removable cover, inspect the mot
ion as it closes. If it is a DIL type or fully sealed type then check them out thoroughly as suggested above and compare with their data-sheets. Scopes are very useful for this type of exploration.
Yes -
- Some that had polarized coils and I didn't notice that when I designed the board. They'd make a soft clicking sound (being repelled by the magnet I guess) but not close the circuit.
- Cleaning solvents, our own rather than the component vendors (actually this was with PCB mount audio transducers)
Do you have a protection diode to absorb the back emf when the coil turns off? If not it is possible that your driver components are being partially fried. As other posters have hinted at, my first plan of attack would the to check the power to the coil, both in terms of voltage and current.
Aqueous cleaning non-sealed relays can do the same thing.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required