Contactor Feedback

Good Morning All,

I have a design problem that has been bugging me for a week or so, and would like to get some expert opinions.

It is, how do you generally wire contactor feed backs.

I normally wire mine to a normally open auxillary contact, so that if it dosn't close I can generate an alarm. I also usually generate a 'Fail to Open' alarm as well, however if only one, or maybe two, contacts weld together it is possible for the auxillary to open but still have this faulted condition.

In safety circuits, this problem is overcome by wiring in the normally closed contact to the feedback. This is normally wired to a safety relay so that if a contact welds, the normally closed contact cannot close and the safety circuit cannot reset (this also assumes a second, backup, contactor or a failure warning alarm).

My problem is, should I wire normal control circuit contactor feedbacks to the normally closed auxillary? This would help detect contact welds much better than the normally open auxillary, but may not detect that the contactor has correctly closed.

A third option is to wire a feedback to both a normally open and normally closed auxillary, assuming these are both available. This also has the problem of expanding the I/O count.

I was wondering what are your opinions and experiences with this problem? How would you go about it?

TIA

Doug

Doug if its that important an issue with your machines then you should look at how safety systems are designed and wired. I am not saying you have to be that redundant or use those systems but they may provide more reliable options for failsafe conditions.

AB offers some info on this as does STI:
http://www.sti.com/

Contact weld

Doug_Adam

In most average applications having a set of auxiliary (dry) contacts close to indicate relay closure is sufficient. If you have problems with contacts welding together, that is the problem you should be addressing. Either your load is too great for the size of contacts you are using, you have a poorly aligned set of contacts or something along those lines. Go to a larger relay or try some suppressers and diodes to reduce the arcing. Just as a reminder, inductive kick and arcing can cause a PLC to fault.

Roger

I think his contacts are welding because he using those IEC contactors. They are way underbuilt and the current carrying ratings are way too high for the contact size, they are more like "Lego Mindstorms" equipment than real industrial contactors. He should try an AB Bulletin 500 NEMA Contactor for awhile, and he would never go back.

Just an off-topic opinion that should stir up some action.....

I normally just use a NO aux contact and in most cases this is perfectly adequate. If this is a personnel safety concern and contact welding is a frequent occurrence then you could use a phase monitoring relay or three current switches to detect the problem.

And Vetteboy, there is no doubt NEMA contactors are "better" than IEC. For severe duty that's all I'll use. However, I hate seeing specs that don't allow use of IEC contactors at all. I use a lot of them as well. For light loads or applicatons where the duty is a few starts per week IEC contactors are just fine. Unfortunately a lot of engineers neglect the de-rating charts for loads and starts that apply to IEC contactors, and just look at the max rating in the tables as they are accustommed to do with NEMA contactors.

Motion sensing

Doug,
If you are really very concerned about the contactor closing or not, then you are probably better off to sense if the actual motor or load is actually turning. Contacts from contactors or aux relays are not that dependable for sure, since it really only tells you about the contactor, not the motor. I have used the magnetic sensor and contact feedback to a plc that is shown at this site:
http://www.maxigard.com/zerof.htm

I can vouch for their effectiveness and the cost is minimal as well if you are that concerned about contact welding and the like.

fooloproof way

Doug,
Instead of interfacing to a contactor to detect that the motor is ON or OFF you should use a CT (current transformer) for a feed back. Tom has kind of touched on it with the phase monitoring relay. But those are expensive and for most apps an overkill. Go to CRM Magnetics
they make this very inexpensive current ring with a built in relay.
All you need is one per motor (on 1 phase only). In your logic provide about 1 sec delay before you start monitoring it.
Advantage in doing it this way is the fact that you have a 100%
certainty that the motor is either ON or OFF.
If you are getting the feedback only from the contactor, you may get a false ON condition if someone flips OFF the local disconnect at the motor.

Hi Doug,

I believe the KISS factor is always the best.
We use N/O auxilliary and display on our screens as 'running' unless the machine motion is monitored (e.g. encoder, speed switch) in which case it is part of the running signal.

When the contactor is released our screens will display the status of the auxilliary. If safety is the ultimate desiding factor then the machine motion should be monitored. This is the only 100% failsafe way, in my mind anyway.

Loading problems on the auxilliary should really not be a problem when feeding a PLC so arced contacts should really not be a problem either.

cheers,

shaun

Using a combustion blower as an example; I would typically monitor the starter to indicate that the motor starter had responded to the run command, next I would monitor a up-to-speed relay indicating the motor had reached designated speed, and finally I would monitor a Normal-Air-Flow switch to indicate the correct amount of air was being displaced. All the inputs had to be present before the combustion blower was considered "Running". What I am saying is use more than one indication to confirm "Running", "On", "Moving" Etc.......Depending on how critical it is to confirm the output action. One indication is not as good as two, and so on...

Roger

Thanks for the replys.

To clarify the issue, while it was a safety circuit problem that got me thinking, I don't have an immediate problem than needs to be solved.

It was simply that I started thinking, is there a better way to do what I'm doing?

I currently wire up the contactor feed back inputs to a normally open contact. Over the last year, I've also started using other field inputs such as 0 speed switches and pressure switches as well.

I was wondering if this was the best way, or if their are other, better ways of doing it.

Thanks again,

Doug

Hi Doug
I am another from Ozz. Know where you are coming from. NHP have an inexpensive current monitor and you can adjust the thing down very low. As it is normally used for over current, the output turns on. You then have 2 inputs for your circuit. The device can also be "adjusted" for load level by running multiple turns through the inbuilt CT.
The other option I have used is to monitor a N/O and N/C contact off the auxiliary block on the contactor. You can even use a combination of early make, late make, early break, late break and normal contacts if they suit the application.
By the way, I am at present using Omron safety relays that mount to the PLC rack. They are fully class 4 compliant, are inexpensive and status can be read from the software. The PLC has no control over them - read status only. A bonus is that there are spare "normal" inputs on the card that you can use as well. Using them on an Omron CS1 PLC.
beerchug