Buzzing relays

[gumpy]

I've been rewiring my electrical system. I got the batteries, inverter and shore input back together tonight, and was testing it. I plugged in the shore line and one of the relays in my transfer switch was buzzing. I've never heard it buzz before. Checked all the connections, and they're correct. Had proper voltage.

I started messing with it, and found that if I put some pressure on it, it would stop buzzing, but would start up as soon as I let up.

So, I played with it awhile, activating and deactivating it manually. This relay controls a couple others in the transfer switch. After cycling it several times in rapid succession, one of the other relays started chattering (like buzzing on steroids). Note that there was no load on the system, so no current draw through the contacts.

Has anyone ever run into buzzing and chattering 120 volt relays?   

I'm expecting I'll probably have to replace them, but I'm really wondering what's causing it.

[junkman42]

Gumpy, is it possible that You have a rectifier that has gone west causing ac voltage to be applied to the relay coils.  My transfer unit has a small bridge rectifier that failed and caused a hell of a loud buzzing.  I do not know if that is Your problem but it is a possible cause.  Hope this might help.  Regards,John

[gumpy]

They are 120vac coils on the relays.  Now, it's possible they do something internal that I'm not aware of, but the coil control is 120vac.

[Dreamscape]

Craig,

I know you said they 120v relays, what is the actual voltage they are getting? Are these the cube type with spades or pins? I know from experience in our industry we have received bad lots at times and the bases do not fit the spades or pins properly. Have you tried switching them around?

Paul

[kyle4501]

Yep, check for low voltage to the coil.

[gumpy]

They are Omron MGN2C-AC120 Power Relays (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=Z267-ND).

Measured 120v on the coil with meter. Connected to 50A shore power. Good voltage.

[Dreamscape]

Craig, After looking at the type you have, I don't know what could be causing it other than age, maybe. Somewhere something does not like the voltage. Sorry wish I could have been more help. Good Luck.

Paul

[kyle4501]

Since the voltage tests good, I'd replace it with a new one. They sometimes do go bad .  .  .  . 

You do have a spare, don't you? 

[Len Silva]

The buzzing or hum at 60 cycles is probably loose laminations and not much you can do except replace it.  The chattering on an ac coil is almost always a low voltage problem.

Len

[gumpy]

So, I found this.....

While relays are usually taught and more easily understood as DC devices, in
practice AC relays are commonly used. These relays typically employ shaded
poles, similar to the shaded pole induction motor. A shaded pole is a coil loop
which is not separately excited by the relay source; it is excited by the flux on
the main relay coil. This coil then produces an opposing current, flux and
voltage (Faraday's and Lenz's law) which holds the contact during zero voltage
intervals on the main coil. Failure of the shaded pole leads to "relay
chatter"— a 120 Hz clicking that occurs every time the main relay voltage
crosses the zero point.

Sounds a lot like my problem.  Maybe I blew out the shaded pole leads on the second relay.

[kyle4501]

Thanks Craig!

[Dreamscape]

Ditto!

Always good to hear WHY!

Paul

[Hartley]

Stick a 4 amp diode in one of the relay coil lines.
that will probably stop the noise. I have done that on industrial stuff
and it solved the problem.

Relay coils get weak over time and use. Most are made with very cheap
lamination sheeves ( iron wafers ). If you could determine that the core
laminations are loose, You could apply a thin shellac to them.

We used a vacuum kettle filled with a shellac mixture, loaded the cores
onto a tray and lowered them into the shallac, closed the lid and pulled
a vacuum for 20 minutes to get all the air out and help the shellac soak
fully into the laminations. ( That's how transformers and relay coils are made!)*
(qualified: Not all are done that way these days. )

Dave....

[gumpy]

Interesting. So a diode on an AC control line might have an affect? 

I would think that would make it worst, as only half the waveform would be controlling the coil, and it would basically be zero voltage for half the period. Maybe I'm not understanding some (many) things about this.

So, what's a 4 amp diode look like?  Any particular part number I should try? I'm not familiar with ratings. I've used 1N4004 diodes on my DC relay coils, and some larger stud diodes on the headlights, but I don't really know the ratings any more on these. I would think the 1N4004 might be about right considering the coils draw something like 320 mA.

[Hartley]

Interesting. So a diode on an AC control line might have an affect? 

I would think that would make it worst, as only half the waveform would be controlling the coil, and it would basically be zero voltage for half the period. Maybe I'm not understanding some (many) things about this.

So, what's a 4 amp diode look like?  Any particular part number I should try? I'm not familiar with ratings. I've used 1N4004 diodes on my DC relay coils, and some larger stud diodes on the headlights, but I don't really know the ratings any more on these. I would think the 1N4004 might be about right considering the coils draw something like 320 mA.

Sorry to be confusing.. Yes a single diode would make it 1/2 wave (30hz) and yes it could make it louder. But the d.c. component will create a stronger attraction on the magnetic part and may be silent. If not the a full wave bridge rectifier could be employed.

The trick with diodes and relay coils is to make sure the PIV rating is very high.
(Peak Inverse Voltage) That is the spike that feeds back when a magnetic field collapses.

The diode must survive that feedback... ie: A Diode with a forward voltage rating of 200 volts might have a PIV rating of 1000 volts. ( They are all different. ) The higher the PIV rating means that it will withstand more voltage surging back towards it.

I don't have my handy dandy diode manual so would have to locate a source and data on which would be the right one to try.  1N4001 series might work, all you would do is fry a cheap diode if it doesn't work.

Well so much for long shots....

Dave...