Chattering 12v DC Relay When Switching 220v AC?

What is the voltage rating of the contacts? You may have exceeded it.

both relays are rated at 40A, but that is at 12v dc.. The only load going through the relays was 200w at 220ac.

How would this cause the chatter?

Regardless of the cause of the chattering problem, you are creating a problem for yourself by using a relay with contacts rated at 12VDC to switch 220VAC. It is very likely that the relay will have a depressingly short life under these conditions.

It is possible it is inducing a voltage to the coil. If in fact, it is a low voltage DC relay, you will have to find one that is rated for 220VAC, in addition the socket if used will also have to rated at that value. The ratings are sometimes printed on the top of the relay.

I am wondering if the charge controller is causing the problem. If you have a nice fat electrolytic capacitor (say 330 uF or larger) rated for at least 12 VDC you might try putting that in parallel with your relay coil. Make sure you observe the correct polarity. If the charge controller is pumping power into the battery in pulses, it won't bother the battery but the relay might not care for that. A diode to the relay might also be helpful in preventing current from reversing when the charge controller is not providing energy.

The chatter sounds as if your power to the relay is actually dropping out and recovering so fast that you can't observe that with a volt meter. But your relay may also be sensitive to the magnetic field of the AC power flowing through your contacts. In that case, use a different relay whose contacts are not so near the coil of the relay.

A slightly higher DC voltage for the coil may also be helpful but that is not likely to be an option.

Can you draw a diagram of what you've connected? Can you give us a part number for the relay?

Relays used in cars are not guaranteed to be able to handle line voltage AC.

Hi,

I've tried just connectiong the relay coils directly to the battery as I thought the charge controller was causing the problem.. but no change.

I'm just placing the relay / relays (L & N) contacts using N/C contacts between two house RCDs to isolate a circuit in the house. The load on this circuit is no more than 200w @ 220v ac.

I'm confused why this is happening?

You must be sure that the relay does not share coil contact with the AC side.

Some automotive relays share the grounding contact with the switched side.

There is no other reason IMHO that there should be a problem. If the 220 was too much for the contacts, they would be welded closed by now.

If the relay is sharing 220V to the DC side you are lucky that you havent destroyed or damaged anything there - - - yet.

We need the Relay part #.

The relays i've tried are: Nagares SA 02032 RLP/52-12 & 2031 RLP/5-12 I'm sure the coils are not connected to any of the switch terminals but I've not metered them, just gone by the diagram printed on the relay.

A quick search of these P/N tells me that these relays are designed to switch automotive voltages of only up to 30V. You need a better relay or contactor that's rated to switch 220 VAC with a 12V coil.

Do you think the following relay would do the trick? Ebay ref number 190837664664 TQX-12F 30A DPDT Cheers

You have a better chance with this since it is claimed to switch 220 VAC. Review the manufacturer's specifications (not the Ebay claim) about the relay. I would include some modest added margin, too. The manufacturer's data sheet should tell you if 220 V is the absolute maximum or if they tested the design to a higher voltage to assure 220 VAC will work.

Why are there two RCDs? It sounds as though this thing is in the fixed wiring somewhere, in which case it needs to be removed for fear of courting unpopularity with the building's fire insurance company.

What is the interrupt voltage rating of the contacts for this relay? I'll bet that 220 VAC exceeds this. Additionally you may have an isolation problem between the coil and contacts.

Basically it sounds like some attribute of your unspecified relay is unsuitable.

The ebay relay is rated at 30amp 250v ac

"powering the relay from a 18ahr 12v battery" That's half the story. No details about coil driving cirquitry. A relay will NOT disengage untill coil voltage goes about half the rated. This tells us that for some reason in your config, coil voltage is NOT enough (<6V), so leave relay contact specs for now. Automotive relays do need relatively high coil current, that your driving cirquit may not be able to provide. I bet your cirquit involves electronics and they somehow interract with AC, so troubleshoot that. S.M.

The relays are driven from the 12v battery direct for now until the chattering issue is resolved. Then it will be driven from either a 10A or 30A solar charge controller so the relay will drop out when the solar controller switches off. As i've explained.. The relay/s are being connected direct to the 18Ahr 12v battery at the moment and chatter when the switched AC is connected.

As in another answer post, it appears the relay is not insulated to 220 volts between coil and contact assembly. When the 220 is introduced, it leaks through the insulation enough to override the DC signal and cause chatter. Eventually it will short and you will not like that.

There are many relays which will have the properties you require. I did not research your EBAY listing but if it specifies ratings for 220 volts and a 12V dc coil, you are going to be ok.

I think if you tested the actual voltage while energized its low get 12 or a little more to that coil. I wouldn't use a car type in this case

By chance, is the battery/charge controller also powering the 220V load you are switching, like through an inverter? If so, when you switch the load on, the load may be pulling the battery voltage low.

A Great Answer!

I particularly like the concept of examining the contacts of a relay with a loupe. So very much information can be learned from a forensic "failure" analysis. IMHO Real engineer schooling happens not from textbooks but from failed noble efforts carefully studied to see what one has overlooked.

Thanks for the GA.

Just like we used to do to check the mechanical points on an older distributor in a car. Remember the 0.016" feeler gauge?

Yes, we learn more with dirty hands and an inquisitive mind than from academic books!

Good Luck, Old Salt

A lot of good advice here. Most of it also "safe"....

Using a relay designed to switch nominal 12 volts DC for mains AC could be REALLY dangerous. Many have mentioned that already, but you do not appear to see just how serious and dangerous this could be......

Please take notice and act on such posts.....best wishes for a long life!

No way that the observed chattering is a function of the contact rating. (the concerns posted regarding the contact rating is valid however)

Marcus, is one leg of your 12VDC supply referenced to a ground that might be connected to your AC neutral?

You didn't post a connection diagram of your own, a drag to do I know, but can you tell us what you have connected to each pin.

L From 220ac grid primary ac rcd to pin 30, pin 87a to supply / 200w load rcd an the coil to the battery or when the issue is sorted a adequate 12v dc supply from charge controller. The same config for the ac N line.

I suspect the chatting is due to the dc coil inducing the half cycle that is flowing In the opposite direction to the coil current hense 25hz. Automotive relays are very compact with the wiring close to each other. I should try an octal relay obtained from most electrical supplyers.

Hi all, thank you for all the advice and especially Old Salt.

i've just metered to see if I can see any leakage from the AC across the contact to the coil and I've nothing (I've only the L going through the N/C contacts of one relay.

The cable that feeds the coil is about 10mtrs in length. The remote end that I have been connecting to the battery is dissconnected, and so is the relay end, but when I measure between the Grid Supply N and the dissconnected 12v supply cable, I'm getting a reading of 190v ac??? Why, as the cable is completly isolated?

I measured between the coil of the relay that has the L ac going through the N/C contacts and the N line and I'm just getting 1.5v ac... nothing much at all.

What is going on???

I have ordered the 250v Ac 30A contacts rated 12v relay to replace the auto relays as reccomended.

Have you got a charger connected to the battery by any chance?

If you are truly isolated then you shouldn't even see that 1.5VAC.

Photo time if you can't send a sketch....

I've had no charger connected since I 1st had the problem.. just a cable strait from the + and - of the battery to the relay coil.

I basically just want to isolate one mains AC circuit from the supply grid in my villa so I can run this circuit from a solar power system.

I'm running the L and N through two 12v coil relay N/C contacts. When I connect the relays, they chatter at what sounds like 25hz.

I'm sorry, i don't have a scope or scope meter, just a std Fluke meter.

Does anyone know why I'm getting a near mains ac reading between a totally isolated from the relay N and completly unconnected 2 core cable that I'm running from the battery?

Thanks

As far as the near Mains voltage, this is a product of induction, it has virtually no ampacity. I suspect that if you read this voltage at the same point with an analog meter you won't 'see' this voltage. It is essentially a 'ghost'.

BTW with the coil in play you are not totally isolated, inductance can always play a role. Pat...

The 190 VAC reading of "isolated" wires is most likely just leakage currents into a very high input impedance of your meter. In essence what you are doing is testing your insulation. If the leakage current path is purely resistive, then your wire insulation would be about 1.5 million ohms of resistance. I'm assuming that the meter has an input resistance of 10 million ohms. With an AC signal there's also an inductive and capacitive current path that adds to this effect. If you had an oscilloscope, further testing could be done to measure the phase shift in the AC voltage and determine the type of leakage currents you have.

For your sake while "experimenting"-

If you have a DC amp meter, there probably is a scale for it on your Fluke, measure the current that the relay coil is pulling. With this install a fuse holder in series to the coil and use a fast blow fuse about 110-125% of the indicated current.

If you are not pulling any current through the contacts other than your meter current, wire two (2) fuse holders in series between the voltage source and the contacts. Place one (1) fuse holder on each wire leading to the contacts. This way you don't have to worry about which wire is the neutral and which is the hot. These fuses should be fast blowing fuses with a current rating as low as you can obtain (no greater than 1/10 amp if available).

The purpose of these is to prevent any disastrous consequences from a short circuit, personnel contact with the device or other potential problems. Although the fuses might not prevent an electrical shock they certainly will shorten the duration of it. This is why you should use the lowest rated fuses, especially on the 230V contact side.

We want you to be around to read any forthcoming answers.

Good Luck, Old Salt

Sparky-o. Welcome to the insanity.

Mains are inducing voltage. Not uncommon in control circuits mixing vac and vdc in close proximity to each other. You need a properly rated contactor or relay. There is not enough real estate between the coils and the primary contacts in a 30 or 40 amp automotive type relay to isolate them. This is why we don't put DC control lines in the same conduit as AC mains. This component is not going to work properly in your application. Just out of curiosity is the load inductive or resistive? Pat...

I had encountered those chatter problems about 40 years ago and to my recollection I solved or minimized the contact bounce problem by getting the suitable relay that matches the type of load(s) being switched!

The arcing produced in-between the relay contacts opening and closing actions most of the times is caused by the type and nature of the load(s) being switched.. (being purely resistive, inductive or a combination thereof). More inductive will cause produces more contact arcing due to back emf which in turn causes the "contact bounce" as we call it and incorporating snubber ckt only worked and effective to a certain extent. Synchronizing the relay switching action to the line frequency is another approach.

The most economical solution I found was to use a relay with a bigger or wider contact areas. Finding the relay with the widest contact areas that can sustain and dissipate the energy contained in the arcs will minimize the contacts from bouncing back and forth!

Unless you can time the switching action to occur, synchronized with zero crossings of the switched power which will be more expensive approach.. to minimize the contact bounce!

Good luck!

The first thing you need to do is get rid of the relay. You need a relay with contacts RATED for 220 AC volts, with a 12 volt coil. You are very fortunate that you have not had more problems than a chattering so far. While usually, many contact designs, can have both AC and DC ratings which are usually far different, it is possible to design the contacts so that they will work on either but not both. That particular relay will not work on AC. I don't know why and have never taken one apart to see why. I do know that when I had a diode fail in my alternator that a relay like that also failed, yet nothing else in the automotive system failed. The diode failed shorted and resulted in an AC signal being introduced in the DC circuit, but at about the same volts as the battery. It did cause a terrible sound in the radio to alert me that something was wrong, but the only damage it caused was to a relay similiar to that one.

One of the tricks to making a small relay like this work with a robust amperage on DC without welding the contacts or burning them, is to put a diode across the contact in such a way that it will prevent arcing when the contacts open. These contacts usually do not have a polarity, so I would guess that the diode, if there is one, is a back to back zener with a voltage about 18-20 volts. No telling what a 220 volt AC signal would do to that. It is surprising that you haven't let the smoke out of the relay by now. And everyone knows that if you let the smoke out it won't work anymore, don't they? How many of you remember the old "smoke inculcator" that was a spoof in the old GE catalogs that was claimed to "put the smoke back in" to electrical devices.

Are you planning on killing someone with your diode idea!!!

There is AC across that part of the circuit.....that could mean that half of the AC would be blocked when the contact is open, but the other half would be still be there, so it would/should still be considered "LIVE" and it could kill!!

Contacts for DC and AC are different, as someone else pointed out, the AC ratings are usually far, far smaller than the DC.....

You were right about ditching the 12 volt DC relay.

A capacitor can be used across DC contacts to reduce arcing, as in an old fashioned car ignition contact system.

AC needs a "snubber" of some sort to blow out any arc on large contact systems.....or contacts designed to handle the arc (I am not an expert in this area!) on smaller ones.

A diode might be used across a DC relay coil to stop the back emf from destroying other nearby electronics, maybe that what you were thinking off, but it doesn't help the contacts much, except that it might speed up the "drop off" in some cases.....

Yes, i will use a snubber rc network to supress any arcs (hopefully, although loads will be low when switching - 200w), and flypast diode to handle the back emf. There's no way I'll bd fitting diods across the contacts as you don't when switching AC.

or even a G-mov.

What is a G-mov?

It is very similar to a henway.

Bob, you've totally lost me????

http://pml.nist.gov/spd-anthology/files/GEMOV_saga.pdf

I'm surprised no one seems to know.......

Of course everybody knew it is varistor, just a bow tie as represented in typical applications /schematic diagrams! Commonly appear as if two diodes were connected back to back thru a common cathode... GE developed them from metal oxides...

Ahhh..varistor.. Yes i know these, its an option, but what would be best to snub any arcs?

Those are what many are using to snub arcs, the MOV has a specific breakover voltage; you select it to be above the normal carry voltage of the circuit, and when the contacts open and back emf starts, the MOV will conduct to quench the back emf energy instead of it being dissipated into the opening contacts.

Utilization requires the calculation of how much emf will be absorbed, to size the OMV correctly ( joules dissipation )

Many have commented that 220V AC is almost 10 times 30V - not a correct logic as for rating of contacts interrupting DC supply is more difficult (needs longer and faster contact break) as there is no natural Zero in DC as it occurs every 10 m secs in case of AC.

Now what could have caused chattering - can be any or more of following which is causing AC to mix with DC.

1. Short gap between terminals (you can see at the bottom of relay where pins are coming out - their is very small gap between terminals).

2. Failure or increased leakage between insulation between terminals when subjected to 220V AC when these are manufactured for 30V DC. AC voltage is more severe for insulation as compared with DC.

3. Occurrence of tracking at surface of insulation between terminals. This can be either for reason 2 above or any dust/ humidity either inside or outside surface of relay base.

My gut feeling is that the cause is leakage through the insulation as said in previous posts and possibly a voltage induced in the coil, but i don't think enough current would be generated to have any affect? I do understand the different power ratings between ac and dc because of the Zero Voltage situation at the AC switch/cycle phase.

I have ordered some 250V 30a rated contact 12v coil relays that.....I hope will be suitable for my application.

Seems you're taking the most sensible route there mate.

Having to order these must be a drag. Aren't they ex-stock at an electrical hardware shop near you? They're very common pieces of control hardware in most industrial environments....

I have a choice of off the shelf suppliers all within 15 minutes of where I live and every town in Cambodia has someone selling these. A lot quicker than waiting for an order to be delivered. I would have thought these would be easy enough to just go out and buy off the shelf anywhere in Europe....maybe I'm lucky here.

I'm impressed with your decision to have a go with the relay you had on hand.

Looking forward to an update when your alternative relay lobs in and you energise it.

Regards

Its a problem getting things like the relays i've ordered here in this part of spain as my Spanish is no where near good enough and not many Spanish speak English for technical stuff. I could possibly spend hours and cost loads of fuel driving around trying to find what I want, and then everyone closes for Siesta at 2pm for 3hrs! I'm also still waiting for a solar power inverter to arrive as one I have blew up before I started thinking about this project. Once things are working.... If ever.... I'll possibly post the wiring schematic here. I lok forward to hearing more comments and possible soloutions. Thanks

"Its a problem getting things like the relays i've ordered here in this part of spain as my Spanish is no where near good enough and not many Spanish speak English for technical stuff. I could possibly spend hours and cost loads of fuel driving around trying to find what I want, and then everyone closes for Siesta at 2pm for 3hrs!"

So why not try one of the internet-based component suppliers - Element14, Radio Spares, Digikey etc ? Any of them will have more relays than you can shake a stick at, including one suitable for your application, and will get the part to you within 24-48 hours.

Thanks for the head up on suppliers, I'll check them out on line.

I've suddenly realised what the problem is:-

"THIS RELAY IS FEMALE AND CANNOT STOP CHATTERING, PUT A MALE RELAY IN AS QUICKLY AS POSSIBLE!"