relay contacts sticking

[readydave8]

2 3/way and 3 4/way switches to the coil on icecube relay, long runs of wire. Relay works a receptacle with lighting transformer plugged in. Often when switch is turned off transformer receptacle stays energized, they can tap relay to make contacts open.

Relay rated for 10 amps. Transformer is 2 circuit, 600 watt, reduces 120v to 12 v ( also has taps for 13, 14, and 15 volts). Loaded approx 135 watts on one circuit, somewhat less on the other (I can get more accurate figures if needed).

Tried changing the ice cube relay and the replacement did the same thing.

Any advice?

 

[nizak]

You mentioned long runs, what is your voltage at the coil?

 

[don_resqcapt19]

Try adding some load in parallel with the relay coil. Take a look at this document

 

[readydave8]

You mentioned long runs, what is your voltage at the coil?

Good question that I won't have an answer until the next time I or others are at jobsite.

I would have thought that voltage drop would have caused problem with coil making connection, not breaking, but see now that if its arcing when making connection could be "welding" contacts and they don't want to open?

 

[gar]

100928-2039 EST

Is your relay something like a P&B KUP-11? 10 A rating would imply silver-cadmium contacts. These are better than pure silver for high currents. Silver is better for low voltage.

You should not have a problem switching a 120 V AC circuit of 1200 W load. Three pole relays in this package are available and you should use this type model with all of the contacts in parallel.

If these are AC to DC to AC to DC voltage changing power supplies, in other words a switching power supply instead of just a plain transformer, then there would be large input capacitors and these could cause very high peak turn on currents. The contacts could weld on turn on.

If it is just a transformer, then the greatest arcing occurs on turn off. But in an AC circuit there is not a rapid build up of a mound on the contacts.

For roughly 40 years I have, on a number of machines, used a P&B KUP DPDT with 10 A contacts in parallel to switch a DC circuit of 1 A at 110 V with a very large inductive load. This is very severe service. The relay switches multiple times every 20 seconds. Typical life is 3 months to about a year. When switching a DC circuit there is a unidirectional transfer of metal from one contact to the other. This results in a conical mound on one contact and a corresponding conical cavity in the mating contact.

For your application I might suggest a GE RR type relay instead of the ice-cube, or a solid-state relay. The GE has an advantage that you can have as many control points as you want with simple wiring. I have upwards of maybe 50 of these relays in my house.

.

 

[hurk27]

Good question that I won't have an answer until the next time I or others are at jobsite.

I would have thought that voltage drop would have caused problem with coil making connection, not breaking, but see now that if its arcing when making connection could be "welding" contacts and they don't want to open?

Nope capacitive coupling from the long runs is holding in the coil, like was said, add a load to the coil such as a resistor, and it will go away, try touching a loaded meter such as a wiggy or volt-con to the coil terminals, and I'll bet it drops out. many ice cub relays have very high impedance coils, and can be held in by very little current.

 

[Jnas480]

I had same problem almost at a power plant.
The low-voltage wire that some one ran about 1,800 feet was in medium voltage and high voltage cable trays and tied in PLC .

Basically the long cable run is like a step-up transformer due to all the power cables around the low-voltage cable.


If the cable is shielded, ground it properly. There are other ways to reduce the stray voltage by resistors. (They must be sized properly)

Like Hurk27 said

Good Luck

 

[dicklaxt]

I'm confused again,what is a 2 circuit transformer?

dick

 

[dicklaxt]

Try adding some load in parallel with the relay coil. Take a look at this document

Nope capacitive coupling from the long runs is holding in the coil, like was said, add a load to the coil such as a resistor, and it will go away, try touching a loaded meter such as a wiggy or volt-con to the coil terminals, and I'll bet it drops out. many ice cub relays have very high impedance coils, and can be held in by very little current.

Agreed capacitive coupling between the long parallel control wires is the problem.A way to combat this is to have the circuit opening switch close to the voltage source and the coil,obviously with 2-3ways and 3-4ways you can't do this.If tapping on the relay works in this scenario then I believe the jarring of the relay is causing the contacts to momentarily open thus acting as a stop button located close to the source and dropping out the relay by breaking the control wire circuit continuity.If the system drops out and stays out then one of the switches is open, if it re energizes then the control circuit to the relay coil is intact(ie: all switches are in the closed state).

The interposing relay in Don's post would support this logic of breaking the circuit close to the source.

Thats my story and I'm sticking to it

 

[don_resqcapt19]

...If tapping on the relay works in this scenario then I believe the jarring of the relay is causing the contacts to momentarily open thus acting as a stop button located close to the source and dropping out the relay by breaking the control wire circuit continuity...

Dick,
I think the tapping of the relay causes it to open because the relay is operating at the very low end of the hold in current. The capacitive coupling is providing enough current to barely keep the relay closed. The physical force of tapping overcomes the magnetic force of the coil and it opens the circuit.

 

[dicklaxt]

That may be true,good thought, in either case the problem is still capacitive coupling.

dick

 

[gar]

100929-0700 EST

I took a P&B relay with a 120 V 60 Hz AC coil of comparable size to a KUP. Pull-in is 80 V, and drop-out is 40 V. In between these voltages it buzzes.

With 120 V and 60 Hz as the source the relay pull-in is at 0.2 mfd and drop-out is at 0.15 mfd.

#12 Romex has a capacitance of less than 20 pfd per foot. 150,000/20 = 7500 ft. A higher source voltage and/or more parallel wires would reduce this distance. Suppose it got down to 2000 ft. Unlikely this application has that much wire length.

I doubt it is capacitive coupling.

Rereading the original post it looks like it is a single 600 W transformer with two circuits in parallel on the secondary. The secondary power load is small, less than 2 times 135 W. We need to know whether the "transformer" is a normal simple magnetic transformer.

.

 

[dicklaxt]

You nay be correct GAR , stranger things have happened,this scenario tho does fit like a glove to long distance motor control using #14 that I have been associated with,we started getting antsy at 1200 feet.

dick

 

[ptonsparky]

Years past pivot installers would try to run control 1/4 mile to roadside. via 12/3 uf. The units would not stop. Capacitive coupling was my guess. Had the same problem with pivot/well @ 900 feet. Adding an extra load at the far end worked but I can see the interposing relay would have been better. I used DC on some and various other scenarios on others to get things to work together. New pivots come with every option but the TP holder.

 

[gar]

100929-0800 EST

To the last two posts. What were the voltages of the adjacent parallel wires, the coil voltage, and coil impedance?

If capacitance in the line is a problem, then a shunt capacitor across the coil will help without much power dissipation.

Another approach is to use a DC relay and a DC source. For very long runs, miles, even on a DC circuit it might be useful to put a capacitor across the coil to reduce momentary operation of the relay.

.

 

[steve66]

My first thought was arcing across the relay contacts caused by inductive kickback from the lighting transformer when the relay opens. It seems like the arcing could cause some carbon on the contacts, which might stick until the relay is tapped.

If long runs and capacitive coupling is the real cause, try putting a resistor across the coil. That should shunt some of the capacitive coupling. If that fixes it, then you have your answer.

 

[hurk27]

100929-0800 EST

To the last two posts. What were the voltages of the adjacent parallel wires, the coil voltage, and coil impedance?

If capacitance in the line is a problem, then a shunt capacitor across the coil will help without much power dissipation.

Another approach is to use a DC relay and a DC source. For very long runs, miles, even on a DC circuit it might be useful to put a capacitor across the coil to reduce momentary operation of the relay.

.

capacitor across the coil? that will delay the let go of the armature, and keep the contacts in for a delay?

we used to just put a resistor across the coil to help bled off the stray voltage, wont a capacitor build up to the stray voltage and in a way help it to hold in the relay?
DC only sometimes help since these relays will still see the stray AC voltage, and most will still operate on AC or DC, generally a 24 volt AC coil will work at 12 volts DC. usually not much difference between an AC and DC relay coils except AC coils are design to not buzz at 60 Hz,

 

[ELA]

... they can tap relay to make contacts open.


Tried changing the ice cube relay and the replacement did the same thing.

Next time can you remove the coil wires and see if it drops out prior to tapping on the relay case?

 

[ptonsparky]

100929-0800 EST

To the last two posts. What were the voltages of the adjacent parallel wires, the coil voltage, and coil impedance?....

Many years ago.

Adjacent voltage would have been 480v. In the same trench but most likely not bundled together. Coil voltage was probably 120v.
Coil impedance...your guess is better then mine.

 

[gar]

100929-0836 EST

An AC relay will work on DC with a sufficiently lower voltage that does not burn out the coil. A DC relay will not work on AC, it will just buzz.

The AC 120 V relay I tested above has a coil resistance of about 2.2 K. A 1.0 mfd shunt capacitor with this coil resistance has a time constant of 2200*10^-6 = 0.0022 seconds. 1 mfd has a capacitive reactance of about 2700 ohms at 60 Hz.

DC pull-in of said relay is about 35 V and drop-out about 4 V.

.