sticking relay (electric roller door)

[Stephen]

Hello,

I bought a roller door from a well known auction site ;(

The problem is that the box of electronics controlling it has got two relays. I imagine one for up and one for down. The "up" relay is sticking so that the door will not open. I have opened the box of electronics and given the relay a tap and it is working for now but I expect it to stick again soon. How can I prevent it from sticking in the future? Is there anything I can do? I guess I really need to get out the soldering iron and replace it before it fails completely.

Does anyone know the schematic for these boxes of tricks? I presume the relays isolate the electronics from the motor? Is the switch side of the relay connected to the mains on one side and the motor on the other? Is the relay output open circuit when he door is stationary? If so I was wondering whether I could put a switch in parallel with the relay so that I have an emergency override? Is this possible?

My worry is that if the relay output is connected to ground when stationary, I don't want to blow a fuse when I switch the override!

I notice there is a spark when the relays open. I'm not sure whether that's the reverse emf in the coil or just the mains jumping the contacts as they separate. Possibly this is welding the relay stuck. Do you think I should add diodes?

The box is branded Nico IIRC.

Thanks, Stephen.

[RW]

If relay is hard wired then replace with a Maplins equivalent or alternatively hard wire from the terminals a replacement

[Adrian]

Was the controller new (or sold as 'new').? Maybe the dodgy relay is why it was on ebay ?

Probably

If all that your switch is doing is to short out the relay contacts then it shouldn't do any harm.... but do you know how the power to the motor is cut when the door is at 'full open' or 'full shut'. If this is done by the control box then adding a manual control across the relay could be very dangerous...

You're confusing two things here, I think. The reverse emf from the relay coil (assuming that the relay coil is fed from a low DC voltage) will appear on the low-voltage side of things - and should be absorbed by a diode connected across the coil.

If you're seeing sparks then they are coming from the contacts themselves - which are on the mains side of the relay. Don't fit diodes to these contacts - they won't last very long..... There might be some point in fitting a high-voltage capacitor across the relay contacts.... but make sure it's one that's rated for mains use.

Have you mentioned it to the seller - might be a plan before trying to re-engineer it yourself....

Regards Adrian

[Dave Liquorice]

It may never stick again... but see below.

Replace it.

The contacts will be switching the motor current not the coil. Diodes won't be much good on mains you need to fit a mains rated "snubber" circuit across the contacts that are arcing. Something like:

Possibly this is welding the relay stuck.

Quite likely so the simple tap above may not be a cure all.

I detect some rather basic questions, are you sure you are up to delving into this box with live mains?

[geoff]

In message , Stephen writes

Angle grinder

No - I think you should stay away from things you don't understand

There is no such thing as a "sticky" relay, what do you think "sticks" ??

So, you tap it and it works ?

sounds like the coil voltage is borderline, or you have a cracked joint or something

Don't just replace the relay, I doubt it will have any effect

yeah - pour a couple of dozen on top, see what happens

Honestly, I think you should let someone with a clue have a look at it, it doesn't sound like something you should be diving in to, you are likely to do more harm than good

[Harry Bloomfield]

Stephen pretended :

Tapping it might free welded contacts, but not usually make open contacts close. That sounds more likely a poor soldered joint on the PCB.

You could ask the manufacturer if he might supply one, but you may need to convince them that you know what you are doing.

Yes.

24v DC is a supply often used on the electronics/control side of the door. Sometimes the motors are also 24v DC, but more usual is 240v AC mains. The relay will probably be there to isolate one, form the other.

Probably you could, but not for certain without investigating the circuit. If the relay has two connections for the coil, plus two or maybe three for the contacts - and ONLY TWO contacts are used - then probably it will be safe to wire a switch across it. If there are three and all three are used then it definitely is not safe to do - it might indicate some sort of braking on the motor.

The welding takes place as the contacts make. The contacts can spark on both opening and closing, that is perfectly normal.

Reverse EMF and the use of diodes relates to the coil side of the relay, not the switching contacts. If the controller is designed for the door, then the sparking can be disregarded.

If you are certain it is the relay to blame, then fit a replacement. They are not expensive, a pound or two each. Maplins might well have them in stock, or Farnell, or RS. Usually there will be a maker/model details stamped on the cases, type those into a search engine.

[The Natural Philosopher]

Not if switching mains.

Can use a high voltage capacitor - 400v 0.01uf or so in series with a resistor..

If relay easy to replace, replace with better.

If it can be opened, file contacts smooth for another year of life..well it worked in my old XJS anyway ;-)

[meow2222]

Sticking is localised welding of contacts, caused by the contact surface being rough and oxidised, which is normal wear & tear of relays.

take the cover off the relay, use a little piece of medium-fine sandpaper to smooth the contacts.

Dont think theyre standardised.

usually yes

you tell me. Probably, but really you shouldnt need to ask that

cant you figure out how to answer that? We cant see the circuit working.

the latter

yep

no, that only works for dc loads.

NT

[gazz]

maybe, probably, maybe not, who knows,

your emergancy over ride should be the chain dangeling from the door, you usually have to yank it hard in one direction, or wind it one direction or something like that to get it to pop off the holder and onto the manual operate pulley, then wind the door open or closed with it and fix the controller/replace the motor when you get time.

[BigWallop]

Now that you have freed it off, it may not ever stick again. If it had been a small carbon build up over the contacts, the slap you gave it might have been enough to clean it. Daily operation might be enough to keep it clean. But you'd have to tell us what the contacts on the relay look like. Are they blackened and damaged? Looking inside the relay, does it look like it really needs replaced?

What voltage doe the door run on? Have you connected to mains voltage (240Vac)? What voltage does it say on the side of the relay covers or other components? Did you get a transformer with the door, or did you have to supply one yourself?

A separate switch as an override for what? Why don't you remove the electronic controller and install two switches? One for open and one for close.

If this is a mains voltage (240Vac) system, I don't think any part of the control system will deliberately connect to an safety earth bond. That would be really silly.

A relay will arc a little as the contacts make and break. It is why the relay may have stuck in the first place, because the arc can cause a carbon deposit to build on the face of the studs.

The controller most likely contains diode protection against field collapse, so it shouldn't need any more. Definitely not across the mains contacts.

Best to find out who made the controller and search their website for a manual on it. It should give you a better idea of what it can and can't do.

Good luck with it.

[Dave Plowman (News)]

It could be sticking due to a mechanical problem in which case simple replacement should sort it. RS Components etc should do a suitable replacement.

It might be sticking due to the contacts arcing and welding themselves together - in which case the cause of this overload needs to be rectified. There's usually some form of snubber network to prevent arcing when the contacts open on an inductive load like a motor.

[meow2222]

[Stephen]

I was going to use a momentary switch so I could stop it when I could see it was at the desired height, but I will think more about that. I will ask the seller but I expect they buy the boxes from someone else and will not know about the circuitry.

I'm not sure where the spark was coming from; I just know when the relay switches off, it lights up blue for a fraction of a second.

It is a company selling doors, not someone just flogging an old one. They have said they will post a new control box, though I worry will the relay fail on that in time too?

[Stephen]

Thanks for all your help. As a replacement pcb is in the post, I shall wait and fit that. I think the consensus is that the relay could be replaced but with a minority thinking it is a bad joint on the pcb. I'll have a closer look at the old board and relay once I have removed it.

I'm not sure why I started rambling about diodes across the coil; I guess I was typing faster than I could think! I know that the diodes are on the coil side and I know that diodes would not work with ac; I guess I was just in a panic when I typed it. Perhaps a snubber would help but OTOH it's reassuring that you say sparking should be normal and like you said, it was designed fore the job. The question is, was it designed to last.

Thanks again.

[Adrian]

All sorts of safety implications, I'd imagine.... Does the door have some kind of switch / sensor on the bottom edge to stop it from acting like a guillotine ?

No - but it's up to them to make it work. If yours is failing then you are probably not the only person with this problem - get them to sort it out for you....

It'll be from the relay contacts - not the coil. A good design will include a resistor-capacitor snubber to prevent this from happening and prolong the life of the relay contacts. Again - not your responsibility to fix it - and modifying it you'll invalidate whatever warranty there might be and possible cause a dangerous situation...

Quite possibly....

Sorry to sound negative, but, unless you're confident that you know what you are doing, you shouldn't really get involved.... (IMHO!)

Regards Adrian

[The Natural Philosopher]

No. What normally happens is that localised corrosion sets in, and the contacts don't 'make' properly. Its not carbon..its usually silver or copper oxide.

The spark jumps the gap, and welds the things together. Usually leaving a little 'pip' of metal.

Now knocking it may un-weld it, but the pip is there..a nice high resistance point that will weld again.

Filing the silver flat gets you back to square one tho.