PLC Basic troubleshooting guidelines

[rsdoran]

Personal opinion but I am a devout Fluke DVM (because they dont blowup when I forget to change the settings) user in most cases. For many years I carried one of those $5.95 analog meters that most supply house will give you if you buy a good sized order. I had a tendency to forget they were set on continuity and measure voltage with them...doesnt work to well with D'Arsonval movements. I always kept several in my tool kit.

I bought my first DVM from Radio Shack, remember the $59.95 model that opened up like a laptop? That was my first and I learned that even a cheap digital is as efficient if not more efficient than any analog.

I learned the hard way that plc outputs could "LIE". To this day I still keep one of those 5.95 analog meters for things like this. I have attachments for my Fluke 87 to do current but because of old habits I keep an analog type AC clamp (current) meter that will also do voltage and continuity.

I am lucky at my plant, I "confiscated" a golf cart and bolted (after several dumps) a toolbox on the back of it. This allows me to carry a variety of items/tools/prints. Since I deal with encoders etc I also carry a Fluke 97 scope meter.

To each their own, I just never felt comfortable with "wiggys" or those "neon" or "tic" testers, the kind that just lights up if voltage is present. Example:
http://www.fluke.com/products/home.a...GID=3&PID=3394

DISCLAIMER: This is personal preference and in now way is meant to imply that others are or should be compelled to act in the same manner.

[rogerhollingsworth]

I have only had dealings with A-B products but I can tell you that a wiggie or any other test device I have ever heard of WILL NOT blow up the input card. We just discussed this in a different thread covering optical coupling. I suppose in some extreme case and under some strange conditions a single optical coupling circuit could be damaged but I have never experienced it.

Note:
The standard that I have adopted is to use an interposing relay with any solenoid coil and always break both legs. (Pertains mostly to outputs)

I would be happy to share some troubleshooting procedures with you but I would like for you to indicate a topic a little more descriptive than PLC. PLC what?

Roger

[TimeFluxCap]

Just want to add a quick one to this thread in regard to writing up a fault finding procedure.

Q: When you get called to a problem, how often do end up having to change PLC hardware?

A: Very rarely. Almost never.

Q: When called to a problem, is it operation error?

A: Often

Q: When called to a problem, is an apparatus faulty, or is there a wiring breakage/fault?

A: Often

[Ron Beaufort]

Well, looking at post #14, I’d say that Steve D seems to come down squarely in the

I will trust my wiggy any day.

column. And I have no intention of trying to sway his loyalty - I know better than to come between a man and his wiggy. But for those among you who haven’t got your minds quite so firmly made up - you might consider a few ideas. First of all Keith was absolutely correct in post #15 - the reason that the output module can handle a solenoid load - is because the triac device in the AC output module always turns the output off at the “zero crossing point” in the AC power sine wave. So that means that at each and every turnoff event there is little or no built-up energy in the solenoid coil - and therefore no arky-sparky effect back into the module’s output circuit. On the other hand, when a technician uses a wiggy, then the precise instant at which he pulls the probes away from the circuit becomes a highly random event. Yes, he might be lucky for years - with never a problem. Then again, if he just happens to pull the probes away while the AC waveform is at, or near, the maximum - well, then there would be quite a lot of energy stored up in that wiggy’s coil - and we all know that energy is neither created nor destroyed. So where does all of that stored up energy go? Depending on just how the probes break contact, it certainly could go right into the output module’s triac. With that in mind let’s take a look at Allen-Bradley publication SGI-1.1 - April, 1990 which is titled “Safety Guidelines for the Application, Installation and Maintenance of Solid State Control”

http://www.ab.com/manuals/gi/sgi11.pdf

Paragraph “C.2.7 Transient Overvoltage” reads in part: “Solid state devices are especially sensitive to excessive voltage. When the peak voltage rating is exceeded, even for a fraction of a second, permanent damage can occur. The crystalline structure of the device may be irretrievably altered and the device may no longer be able to turn OFF.”

Now I don’t know about you, but that “permanent damage” - “irretrievably altered” - type language makes me a little skittish. Personally, I’d hate to have my boss come to the conclusion that me-and-my-wiggy were the culprits who had been “permanently damaging” and “irretrievably altering” his valuable PLC equipment. There are quite a few knowledgeable people out there who have decided that this wiggy-on-a-PLC issue is something to be avoided. What they say makes sense to me - and so after careful consideration - I believe it. But of course I’m not trying to change anybody else’s mind.

Now Roger brought up a another very good point in post #17.

a wiggie or any other test device I have ever heard of WILL NOT blow up the input card. We just discussed this in a different thread covering optical coupling.

Yes, the optical isolation is definitely a good thing - but still ...

Let’s take a look at Allen-Bradley publication 1746-2.35 - July, 1999 - which is titled “Discrete Input and Output Modules - Product Data” for the popular SLC systems.

http://www.ab.com/manuals/io/1746/1746-td006b-en-p.pdf

Page 50 lists the Environmental Specifications for these common I/O modules - and the specification for “Isolation” is given as “1500 Volts”. A footnote denotes that this specification refers to “Electro-optical isolation between I/O terminals and control logic.”

Now this points out something that we already know: The opto-isolation of a standard input module is NOT intended to protect the MODULE’S INPUTS against higher-than-normal input voltages. Rather, it is intended to protect the PLC’s internal logic circuits from these higher voltages.

And obviously Roger didn’t mean to imply that we could impudently connect a higher-than-specified input voltage to an input module - and then count on the opto-isolation feature to protect the module from “blowing up”. For example: Take an input module rated for 120 volts and connect a 240 volt signal to it. You can kiss that particular input circuit - and all of its fancy opto-isolation - goodbye. Of course Roger knows that - and he said so when he qualified his statement and referred to damage from “a wiggie or any other test device”.

But then what about that published “1500 Volts” rating for “Isolation” - what does it really mean? In simplest terms it means that “optical isolation” is good - but (like just about every other form of protection) it does have its limitations. The book is telling us that once we get up over 1500 volts - then all bets are off. In “extreme cases” - where levels higher than 1500 volts are concerned - then the opto-isolation system just might let a spike pass on through the module - and then right on into the PLC’s logic circuits. That doesn’t sound good does it? What would a high voltage spike DO to the PLC anyway? We don’t know - and we don’t WANT to know. But whatever happens, I can just about guarantee you that it won’t be something good.

Now the question remains - are we SURE that the wiggy is NOT going to give us any more than the specified 1500 volts? And - before you answer - remember that the ignition coil in our car gives us an output greater than 20,000 volts - while using just the 12 volt input of our battery. Now just how much of a spike do you think we might be able to generate with our wiggy - using an input of about 120 volts? Place your bets, gentlemen.

Well, I don’t know about you, but I’m tired of wiggies - at least for now. What follows is something that came to mind while reading about the “leakage current” issues in some of the earlier posts. It’s safety related and comes from the same publication I listed above. Allen-Bradley publication SGI-1.1 - April, 1990 “Safety Guidelines for the Application, Installation and Maintenance of Solid State Control”

http://www.ab.com/manuals/gi/sgi11.pdf

Paragraph “C.2.3 Off-State Current” says in part: “Off-state current is also referred to as leakage current in the literature. A solid state “contact” is a solid block of material which is switched from ON to OFF by a change internally from a conductor to an insulator. Since a perfect insulator does not exist, there is always some leakage current present as long as voltage is applied to the device. The presence of leakage current indicates that OFF does not mean OPEN. The reader is warned that simply turning a solid state device OFF does not remove the possibility of a shock hazard.”

So here’s a scary story -

A technician needs to work on a motor starter - nothing fancy - just a plain-old everyday three-pole contactor with a 120 VAC coil. And incidentally, the coil is controlled by a solid-state triac-type AC output module on a PLC system. So the technician makes durn sure that the 480 VAC, 3-phase line power is disconnected - and locked-out and tagged-out - and checks with his meter to make sure that the juice is really off. Good - so much for the high voltage stuff.

Now the 120 volt AC coil voltage can be easily controlled by the PLC - so the technician just forces the proper output off - and confirms that the coil drops out. Yes, indeed - the LED for the coil’s output is off. And a meter test across the coil terminals indicates that there is less than 1 VAC showing up there. Well, that’s probably just a “false” kind of reading - maybe coupling in from nearby wiring. Anyway, less than 1 volt isn’t going to hurt anything. So now it’s OK to climb in amongst the wiring and fix the contactor - right?

What’s wrong with this picture?

The contactor coil is “loading” the circuit - and “draining off” the leakage current - and so the meter reading looks acceptably low. But a SERIOUS problem shows up once the wiring to the coil has been disconnected. Then there will be NO load on the output circuit - and the voltage will “float” up to its applied voltage level - 120 VAC. Now ... if the technician just happens to brush against that coil feed wire ... this could get very ugly. Especially if the technician happens to be wet with sweat and making good contact with something grounded.

The fact is - to the technician’s skin - that trickle of “leakage” current is going to feel just like a 120 VAC live wire - EVEN THOUGH THE PLC OUTPUT IS OFF! Now some people will say: “Well, that’s going to be a very low current signal - not enough to really hurt you.” Yeah - maybe so. The shock might not hurt you - but it can sure make you hurt yourself.

[Tim]

All I have to say is... pure poetry Ron. Thanks for the information and the real world examples, you don't know how much this helps when you can give an example rather then a bunch of facts. My hats off to yah.

Tim

[LRBrown]

When troubleshooting PLC's I usually approach the problem with the biggest hammer I can find.....

[gauravvohra]

Dear Ron,

unable to find this file(Safety Guidelines for the Application, Installation and Maintenance of Solid State Control) on ab website. Probably they have removed it.

Hope you can help.
thanks,
gaurav

[OZEE]

Probably because the link (and this thread) is nearly 5 YEARS OLD!!! Please check thread dates!!

[Chuck Woodbury]

If you know the output from the PLC, start there and work bacwards.
Much like an electrical circuit, you know the output isn't working so where do you lose the input power.
Eric is correct in saying Isolate, Isolate, Isolate.
I would like to add "Document, Document, Document".

[Mickey]

Originally Posted by gauravvohra

Dear Ron,

unable to find this file(Safety Guidelines for the Application, Installation and Maintenance of Solid State Control) on ab website. Probably they have removed it.

Hope you can help.
thanks,
gaurav

Here:
http://literature.rockwellautomation...n001_-en-p.pdf