relay sticking

[rhino4578]

Hello Everyone and thanks in advance for any help. I have a problem with some relays sticking on a homemade incubator. I'm not an electrical guy so my knowledge is far from advanced but this is what I have.

I have a 12vdc controller with a 10amp SP NO mechanical relay (built into the controller). I am controlling a 110vac 250 watt heat lamp. The cycle time varies but when it chilly outside it can cycle as often as every 2 to 3 minutes. It sticks in the ON position which I've read may be do to contact arc.

If contact arc is the issue I've read of a couple of fixes but wanted someone input that has knowledge of electrical circuits.

1. add a solid state relay to carry the load and use the controller to trigger it (but if cycle time is the issue wouldn't the mechanical controller still stick?)

2. place a diode between the relay and the bulb to prevent backfeed.

Also is it possible that the bulb surge when it comes on may be drawing more than 10amps.

Lastly my controller requires 12vdc for power but my source outputs 12.33vdc. Could this be causing an issue?

 

[Minder]

If it is due to contact arcing you can use a RC snubber across the contacts.
A SSR would solve the problem also.
Don't place a diode in the AC circuit.
relays are very voltage tolerant and that small OV would not be an issue.
M.

 

[duke37]

You must find out if the relay contacts are welded together or if the supply (12V DC?) is still there. A 12V relay may well stick in with less than 4V.

A solid state relay will solve the problem of welded contacts but will not help if the problem is due to a control transistor leaking.The control transistor can be damaged if there is no diode across the relay coil.

Turn off the power, does the relay drop out - listen for the click.
The turn on surge is not usually the problem since the contacts are cold but could be heated if the contacts bounce.
A diode cannot be used with AC.
The relay should be up to the job.
12.33V is very little above 12V, this should not be a problem with a well designed circuit.

Show us a schematic.

 

[rhino4578]

Thanks for the responses. The relay does not release even when the power is turned off. Tapping it will release it.

I've ordered a new controller (same type) and a solid state relay to try that. The controller with the built i relay is cheap ($8) on Amazon so I'm hoping that putting the load on a better relay and just using the controller to activate it will resolve the problem. I would like to use a better controller (maybe a PID) but I can't find one that gives me the range of control that this cheap one does.

 

[duke37]

I used to look after an old electric fence energiser.. This had a relay to produce the pulse. The relay would stick in and needed its brass screw adjusting so that the pole pieces did not touch. It would wear down as the relay worked once a second. You may find an anti-stick screw in your relay.

I have not used a solid state relay but I think they are controlled by a led. If so, the mechanical relay would not be needed and could be removed, eliminating the unreliability.

 

[Tha fios agaibh]

If I have unused extra contacts on a relay, I like to parallel a second set which allows it to handle more current.

 

[TCSC47]

An emergency repair, if you have access to the relay contacts, would be to spray the contacts with a maintenance spray such as WD40. Such a repair could last 6 months to a year. There are commercial fluids specially concocted to do the job but we all have a can of WD40 somewhere near at hand.

When the contacts are closed, the current is conducted through quantum tunnelling --- yes --Star Trek!!! When they part, the insulating properties resume their insulating state that we see at our scale of the Universe and any arc is quickly quenched.

Many here will know that you can spay WD40 onto volume pots that are producing a scratching noise when operated. I myself kept my electric guitar going for about 5 years before I had to replace the pots as being beyond help. When I supported a teaching electronics lab we were having problems with 25 power supply units giving potentially damaging spurious voltage spikes when the voltage controls were turned. I didn't have time to replace the pots in the units but kept them going for a couple more years by using WD40 every 6 months or so.

I always include one word of warning when mentioning the wonder fluid WD40, however. ---- WD40 is NOT a fine lubricant for mechanisms in case you are tempted. It is fine on garden equipment and such, but for things like clocks, model trains, sewing machines, etc. it is definitely a no-no. WD 40 consists of a wax dissolved and suspended in a thin hydrocarbon solvent, and works as a fine lubricant for a short while. However when the solvent evaporates you are left with your lovely clock gummed up with wax! I speak from experience!

WD40 is truly a wonder product, with a terrific range of things that it can do, but it is most definitely not a lubricant for fine mechanisms -- and they should state that in the adverts and on the can.

 

[Minder]

WD40 is truly a wonder product, with a terrific range of things that it can do, but it is most definitely not a lubricant for fine mechanisms -- and they should state that in the adverts and on the can.

An excellent cutting fluid when machining aluminum!
M.

 

[rhino4578]

I used to look after an old electric fence energiser.. This had a relay to produce the pulse. The relay would stick in and needed its brass screw adjusting so that the pole pieces did not touch. It would wear down as the relay worked once a second. You may find an anti-stick screw in your relay.

I have not used a solid state relay but I think they are controlled by a led. If so, the mechanical relay would not be needed and could be removed, eliminating the unreliability.

I would love to remove the mechanical relay but it's integrated into the circuit board on the controller. I may take the old controller in after I receive my new parts and see if one our maintenance guys at work can remove it. If we mess it up it will be no loss since I'm going to replace it anyway.

 

[duke37]

You could leave the old relay in and connect the SSR with wires tacked onto the coil. Check the current drawn.

 

[Externet]

Unscrew the heat lamp and check if the relay contacts still get consistently stuck and need to tap to disengage when power is removed.
If still sticks, the mechanism of the relay is binding. Means repair, lubricate, add tension to the return spring or replace the relay.
If does not, there is welding happening at the contact points. Try filing/sanding or increasing tension to the return spring.
I doubt a Water Displacement -40 fluid will improve anything unless it lubricates a friction point.

 

[TCSC47]

I doubt a Water Displacement -40 fluid will improve anything unless it lubricates a friction point.

I beg to differ. It is possibly the arcing that causes the welding of the contacts. It is not the water displacement properties of WD40 we are using here but the electrical insulating properties of the WD40 just before the contacts make or break which will stop the arcing. However when the contacts are made, the gap between the contacts is down in the quantum mechanical dimensions of the universe and we get quantum tunnelling actually through the WD40 itself, which conducts the current well. An additional benefit is that oxygen is excluded from the contacts reducing oxidation and further enhancing the durability of the contacts.

 

[hevans1944]

If you can get the relay on the controller board working properly, either by dressing the contacts with a very fine small file or spraying them with WD-40, then you can use those contacts or the voltage applied across the relay coil (as suggest in post #10 by @duke37) to energize the control pins of a solid-state relay (SSR). The SSR input is typically an LED with a series current-limiting resistor, so it just needs a low-voltage DC supply to actuate, typically 3 to 30 V but read the label on the device you use.

And, yes, there is a considerable in-rush current when a heat lamp turns on, often as much as ten times the steady-state current until the tungsten filament reaches its operating temperature.

 

[rhino4578]

Just a quick update.. I ordered a new board and a solid state relay. Unfortunately I was in a hurry and hooked the ac leg to the input (dc) side of the ssr, it seems I fried the ssr so now i'm waiting on a new one. I took the old board in to show one of my maintenance guys and he pointed out the the relay on the board says 5/15 amp and as someone stated above the lamp could easily be over amping the switch. The new ssr I ordered is 40 amp so when it shows up the plan is to use the controller to trigger the ssr and let the ssr deal with the load.

It might also be worth noting that I tried the board without a load and the relay switches fine.

 

[(*steve*)]

I beg to differ. It is possibly the arcing that causes the welding of the contacts. It is not the water displacement properties of WD40 we are using here

I fear the adhesive properties in the longer term.

I have a garage door opener which had sticking relays. The easiest solution was to replace the relays. In my case SSR was not an option (the relays were SPDT).

I'm pretty sure that my sticking relay contacts were caused by the initial surge current of the motor. It sounds like you may have had a similar issue with the lamp.

When you get your new SSR, connect 2 LEDs in inverse parallel with a series resistor (1K). Use this instead of the SSR control input. If both LEDs light, or they light REALLY bright then you need to do some more investigation before connecting the SSR.

 

[tedstruk]

Replace the relay.

 

[Harald Kapp]

Some relays are specified for an inrush current much higher than the mean contact current - some aren't.
An NTC resistor (Negative Temperature Coefficient) can be used to reduce the inrush current while allowing (almost) the full operating current to flow.
When off, the NTC is cold and has a high resistance. At turn on of the relay, this resistance limits the current to a comparatively small value. In the on state of the relay, the current through the NTC will heat it up, thus reducing the resistance and therefore allowing more current to flow. This compensates the PTC (Positive Temperature Coefficient) of the lamp.
TDK/EPCOS, for example, has an application note on the use of NTCs for this purpose.

Of course, replacing the relay is mandatory for future reliable operation.