Relay failures

[jamesjcooper]

Hello all,

I have an application where I have a bunch of Ziplink relay output boards installed, and I am having quite a few problems with relays failing, specifically it appears that the contacts are fusing.

The load is a small motor, which according to the drawings I was provided only pull 400ma's @ 120vac, and are fused at 3 amps, with normal fast blow fuses.

We are having problems where the relays are getting stuck in the 'on' position. I have tried to give them a rattling to see if they are just stuck, but nothing seems to help short of replacing the relay in question.

I am thinking of installing MOV's across all the points, but there are about 200 of them, and we, as the integrator, would be eating the cost. The system we replaced was simply a bunch of push buttons driving 5 amp cube relays. The relays in the ziplink module are stamped at 15 amps at 125vac.

Any ideas?

Thanks,

Jim

[FACTSTech]

Jim,
Please give us more information, which relay ziplink? PLC output driving relay coils? How are the contacts wired, commons series, common to terminal blocks, etc.? What are the measured voltages wrt to source? You maybe having an impedance problem, length of return back to the source.

[jamesjcooper]

I am using a mix of DL-06's and DL-250's with D216TD2P output cards driving ZL-CM16RL24A relay boards. From the relay's I am directly switching 120vac which goes out to these motors, that open and close doors.

[DetroitSound]

Most likely you have two things working against you - inrush and inductive kick. Simply put, (in my Polish hillbilly terms) the contacts are smaller and closer together than in the ice cube relays and with an inductive load that is asking for reliability problems. Does it seem to be the same ones that are failing repeatedly or are they at random?

Adding MOVs is probably a good place to start, and one that may suit your purposes well is at Newark Electronics, a stock number for ones that would probably suit your needs is 89H8968 and they are 23.5 cents each in quantities over 100 pcs.

What Automation Direct means to me in terms of industrial control equipment, Newark is to me in terms of board level electronics, connectors, transistors, diodes, integrated circuits, etc. Like ADC they don't have a minimum order and they have great service too. Check them out at http://www.newark.com

But what might do better than a MOV is a part called an RC snubber network, essentially it is a capacitor in series with a resistor. The CDE brand name for these is Quencharc and those are in a small, blue, epoxy dipped package. There are other brands too. It is designed to absorb the transient inductive kick when the contacts open and help prevent the contacts from welding together. Unfortunately these are considerably more expensive than MOVs (almost $4 each in 100+ quantity) and there is a selection in terms of capacitance and resistor value. Too small of a capacitance / too large of a resistance and they won't offer much protection; too much capacitance / too little resistance and your load may stay energized after the contacts open. Unfortunately I'm not much help at selecting them, I'm not an engineer.

I realize that as the integrator you'll be eating the cost, but consider downtime and dissatisfaction to your client, not to mention your time and trouble, and somebody is paying for the failed parts whether it is you or your client. Look at the big picture and how much is the status-quo really costing?

Hope this helps...

[jamesjcooper]

DetroitSound - Thank you for the insight, and I too know just how wonderful Newark is, I think I funded part of their warehouse over the years.

The relays that are failing are at random, and we have not had any repeats. I was thinking that it was somehow related to the smaller contacts, not thinking about the fact that they are physically closer to each other. I think we are only at a 2% failure rate.

I believe the motors have start capacitors in them, but I am not 100% sure. We are just working with the control system, my responsibility ends at the terminal blocks. I am not TOO concerned about cost, I would rather eat a few bucks and have a happy customer.

[FACTSTech]

Relays only weld due to high current applications. One application is when using 3 wire reversing motors (capacitive start/direction) if you are not delaying when switching them from forward to reverse and R/F, this is similar to a locked rotor, thus switching many amperes. To help solve this problem it is best to have a delay programmed into the PLC, so that the relays cannot be ON at the same time.

[ezfx]

We have had a similar problem where my outputs were staying "ON" and I think we hade the same output card....one thing you can do is check what rev # the output card has, if it has B1 it's faulty and needs to be replaced with B2 rev, AT will take care of that with no charge to you...I have not hade any problems since then.

Gerry

[jamesjcooper]

Wouldn't a locked rotor condition blow the fuse? I have each relay feed protected with a 3 amp fast blow fuse