Relay Contact Welding

I am having trouble understanding why the contacts on a relay are welding. The relay supplies power to the motor of an electric axis via its amplifier. The contacts on the relay switches 230 VAC but the contacts are always closed. The relay coil is energized as long as all the doors at the station are closed. The only time the relays open is when one or more doors at the station is open or if someone activates the e-stop.

The motor for the electric axis doesn't always run. It does 4 moves every process cycle (down to pick, up to home, down to place, up to home). Every process cycle is 10 seconds, each move takes about 1 second with 1 second dwell. Are these movements enough to cause the contacts to weld? I mean if the in-rush current caused by the start and stop (on & off) of the motor enough to cause the contact welding?

The contacts on the relay is rated at 6 Amp. maximum (resistive load). The line feeding the electric axis amplifier reads 0.4 Amp whe the motor is running. There is a 5 Amp. circuit breaker upstream of the relay and this never trips.

Right now we replace the relays every 4 months because they are welding. The safety PLC trips out because the relay isn't opening.

Any help or suggestions regarding this will be much appreciated.

Just to be 100% sure, there is no relay between the amplifier and motor, the relay is on the input side of the amplifier, correct? I ask because any kind of line break between an amplifier/drive and the motor being driven will need a very high voltage rating and should be avoided if possible.

Even the input side of drives can cause effects on some components, for instance, most VFDs and servo drives can't be use on a GFIC because the rectifier will trip it. It may be that your relay just can't handle these upstream effects and you should try a different brand/model. You would typically use a relay that calls itself a safety contactor for this type of application and they tend to be a bit beefier.

There is no relay between the output of the amplifier and the motor. The relay in question is at the input side of the amplifer.

Normally I would use a contactor to switch any 230 VAC supply but our machine builder used small component safety relays. I suspect because they think it's such a low current application. I don't know if they considered the in-rush current every time the motor starts? The motor starts 4 times every 10 seconds.

I had a simular problem with some relays that were controlling a Motor Operated Valve (MOV). I was using a AB Terminal Block Relay. It was also rated at 5A.

The issue was caused the Inrush Current of the Motor in the MOV. When the contacts would close in the relay there would be a instantaneous rush of current that was so fast I could not measure it with my Fluke nor would it trip the 3A breaker. That inrush would cause the contacts to weld.

To solve the issue I went to a larger relay. One that can handle 15A. The contacts were much larger so they would not weld. Havent had a issue since.

If you have a scope with a current probe, I would take a look at the in-rush like Bullzi mentioned.

I think the inrush to the amplifier to charge its capacitors might be significant. This is not the 'motor start' current. See if that value is available.

bernie_carlton said:

I think the inrush to the amplifier to charge its capacitors might be significant. This is not the 'motor start' current. See if that value is available.

Click to expand...

That's my thought too. If the operation cycle is not opening and closing the RELAY, then the only time the relay sees inrush is when a door opens, then recloses. The inrush it will see is the charging current of the amplifier, but that could be significant. If the motors restart IMMEDIATELY when the door recloses, that could put it over the edge for sure.

A relay with 6A resistive ratings will have very very low inductive ratings. I'm looking at some I have here and at a 6A resistive rating, the rated inductive load breaking current at 240VAC is only 1.5A. Inrush on charging up an SMPS type of amplifier can be as high as 10X the FLA, so if your amplifier is rated for .4A, the inrush could easily be 4A, which is almost 3X the rating of the contacts. That translates to "weld".

so if your amplifier is rated for .4A, the inrush could easily be 4A, which is almost 3X the rating of the contacts. That translates to "weld".

Click to expand...

Yes indeed. You nailed it down tight.

I bet most of us remember that special moment when we realized how inrush and transient currents can screw up the best-laid plans of mice and men.

A problem with the setup noted by the OP is that when the contactor is opened by a safety issue the VFD will still have a full charge & could run the motor for a short period until it gives a low buss voltage fault, I would prefer the contactor after the VFD to immediately cut off the motor.

jraef said:

That's my thought too. If the operation cycle is not opening and closing the RELAY, then the only time the relay sees inrush is when a door opens, then recloses. The inrush it will see is the charging current of the amplifier, but that could be significant. If the motors restart IMMEDIATELY when the door recloses, that could put it over the edge for sure.

A relay with 6A resistive ratings will have very very low inductive ratings. I'm looking at some I have here and at a 6A resistive rating, the rated inductive load breaking current at 240VAC is only 1.5A. Inrush on charging up an SMPS type of amplifier can be as high as 10X the FLA, so if your amplifier is rated for .4A, the inrush could easily be 4A, which is almost 3X the rating of the contacts. That translates to "weld".

Click to expand...

The specification of the electric axis amplifier doesn't show the inrush charging current. All it shows is the input power requirements and the power capacity (1400 VA) which doesn't really help me here.

I read yesterday that you have to de-rate the resistive rating of the contacts by 40% if the load is a motor. If this is true (probably conservative?) then the motor load rating for the contacts is only 3.6 A. If the inrush is 10X the FLA then the contacts can see up to 4 A every time the motor starts. In this case it's a pick and place movement so it's 4 times every operation cycle. I think the frequency in this case is causing the contacts to weld. Does this sound right? The specification for the relay that I currently use doesn't show a inductive load rating.

I found a relay yesterday that is rated at 10 A (resistive). If I have to de-rate it by 40% then the motor load rating should be 6 A. Hopefully this will be OK to use. The relay socket has the same foot print and it's DIN rail mounted so I don't have to modify the cabinet too much to use it. Hopefully this is OK. Now the problem is trying to get the machine builder to replace the relays.

Thank you guys for your comments. Much appreciated!

kku said:

The specification of the electric axis amplifier doesn't show the inrush charging current. All it shows is the input power requirements and the power capacity (1400 VA) which doesn't really help me here.

I read yesterday that you have to de-rate the resistive rating of the contacts by 40% if the load is a motor. If this is true (probably conservative?) then the motor load rating for the contacts is only 3.6 A. If the inrush is 10X the FLA then the contacts can see up to 4 A every time the motor starts. In this case it's a pick and place movement so it's 4 times every operation cycle. I think the frequency in this case is causing the contacts to weld. Does this sound right? The specification for the relay that I currently use doesn't show a inductive load rating.

I found a relay yesterday that is rated at 10 A (resistive). If I have to de-rate it by 40% then the motor load rating should be 6 A. Hopefully this will be OK to use. The relay socket has the same foot print and it's DIN rail mounted so I don't have to modify the cabinet too much to use it. Hopefully this is OK. Now the problem is trying to get the machine builder to replace the relays.

Thank you guys for your comments. Much appreciated!

Click to expand...

I don't know where you read that 40% number, but it might have been specific to something particular. There is no absolute value like that, there is too much that goes into it. On a CONTACTOR, maybe, because there is more of an air gap to help interrupt flow and cool the contacts. But on small control relays, 40% would be wishful thinking, it's more like 25% (those contacts I mentioned work out to be 27%).

Any time you have a power supply that converts AC to DC now, they are going to have capacitors to smooth out the DC bus from the rectifier. When you first energize the line side, those caps must charge up, and will pull current through the rectifier almost instantly at whatever the available current is in the system, even if only for a fraction of a second. Then if there is a transformer on the LINE side of that rectifier (depending on the design), the transformer itself has a magnetic inrush. One way or both, the amount of current drawn in the instant you energize it is hard on contacts. The 10A contacts may help, but I would look at their spec sheets for the inductive ratings, or lacking that, they should at least state the "Make and Break" current ratings for inductive loads.

the contactor should be placed after the driver and should handle the inductive load of the motor

If welding is an issue you might investigate mercury contact relays or solid state contactors.

You could also try these devices. I have not personally used them.
http://www.arcsuppressiontechnologies.com/360-2/

I don't know where you read that 40% number, but it might have been specific to something particular. There is no absolute value like that, there is too much that goes into it. On a CONTACTOR, maybe, because there is more of an air gap to help interrupt flow and cool the contacts. But on small control relays, 40% would be wishful thinking, it's more like 25% (those contacts I mentioned work out to be 27%).

Click to expand...

+1