Troubleshooting a Micrologix 1200

[Clay B.]

I'm probably being overly careful, as I'm learning here and don't want to do anything that could mess anything up, either PLC or program.

No such thing as overly careful. Just take your time and have paper and something to write on handy.

One thing. Once you upload the program from the PLC. Save it again under another name so you have a backup. This way you will have something to go back to.

 

[mtn71]

I was able to upload the program this morning. I printed it out and am working on it now. There's more to it than I expected: three subroutines with a lot of timers and binary files. It helps that I saved all my notes from before when I traced every single input and output, every wire to every contactor and auxiliary contact. I was pretty good with programming in school so I think I'll be able to figure it out with time.

The dampers are controlled by Belimo damper actuators: http://www.belimo.us/ishop/article/..._10079_10075/subchapter/10084_10079_10075.xml
I traced everything exhaustively, no feedback from the dampers. Funny thing is, the fan outputs go to big contactors, with auxiliarys that control the light on the panel telling you which fan is running, but the actual fan power supply does not go through them. The 480V lines go to VFDs that run the fans. There is a small relay that is controlled by an auxiliary on the fan 2 contactor that is the only thing telling the VFDs to switch to fan 2 instead of fan 1. It seems funny that there is this great big contactor controlling one little light, a small relay, and an input back to the PLC telling it that the contactor is on.

 

[Lancie1]

Funny thing is, the fan outputs go to big contactors, with auxiliarys that control the light on the panel telling you which fan is running, but the actual fan power supply does not go through them. The 480V lines go to VFDs that run the fans. There is a small relay that is controlled by an auxiliary on the fan 2 contactor that is the only thing telling the VFDs to switch to fan 2 instead of fan 1. It seems funny that there is this great big contactor controlling one little light, a small relay, and an input back to the PLC telling it that the contactor is on.

Mtn,

I think you discovered the cause of one of your problems. The PLC is getting erroneous data inputs from the contactors. It thinks it knows when the fans are ON, but in reality the fans are being controlled by the VFD (which could be ON or OFF when the contactor is still ON). I would bet some money that the fans were once run directly from the contactors, but at some time the VFDs were added, and the old PLC program was jury-rigged (but not very well).

The best long-term solution is to send an output from the VFD back to the PLC that shows the VFD is indeed in RUN mode. Most VFDs have at least one relay output which can be used for this function. Otherwise, your PLC is flying blind. That may have worked in good weather, but as you discovered, in bad weather you are going to need to add some radar!

 

[OkiePC]

+1 for what Lancie said.

For proof of operation, we often use a programmable VFD contact and set it up for "At Speed". This way, we know that the VFD has finished accelerating and isn't having any problems (except maybe a broken motor shaft...). If the VFD encounters an excessive load or for whatever reason is unable to attain the speed setting, that relay will open.

"Drive Running" is another common choice, although the drive can be running for quite a long time in some situations when the load is actually stalled.

 

[Clay B.]

If I was a betting man...

First I would bet Lancie is dead on with why that contact is there.

Second I would bet that "drive running" is probably a better function for the drive relay than "at speed". I will bet that the dampers are controlled by fan running and waiting for speed might be to long. You could generate a lot of pressure against the damper if the fan gets up to speed before the damper opens.

 

[OkiePC]

Yes, Clay, you are right. Most of our applications using "At Speed" are conveyors moving loose product. A fan is a different matter entirely.

 

[Lancie1]

Paul and Clay, we are going to have this thing figured out by remote party-line troubleshooting! We can make that system run like a top, and get rid of some of the old trash!

Another wise change would be to add a couple of PLC inputs for each of those Belimo actuators. I have used those before (a clock motor with usually some position contacts), fairly dependable as long as you don't overload the motor.

Belimo Part Nr.: AF120-S US
Auxiliary Switch: 2 x SPDT 7A (2.5A) @ 250 VAC, UL approved, one set at +5°, one adjustable 25° to 85°

Mtn, if you are using the one you posted, then it does have CLOSED and OPEN auxiliary position switch outputs. Run 3 wires back to a Micrologix digital input module, and you will have something to show the damper positions, instead of having to A$$-U-ME that it did open or did close when it was supposed to.

 

[Clay B.]

Yes, Clay, you are right. Most of our applications using "At Speed" are conveyors moving loose product. A fan is a different matter entirely.

I actually use both in our processes. Sometimes we need to know if a conveyor is at speed and sometimes we just want to know we started it.

Just for the record I tried the At Speed on a Fan damper setup once and mangled the damper/diverter. Had to get out a hammer and straighten some linkage so it is something I will never forget. Best part this diverter was part of a solvent exhaust system so hearing metal screech in that environment is real butt puckering experience.

 

[Lancie1]

...hearing metal screech in that environment is real butt puckering experience.

ha-ha! I bet!

Clay, your idea to use Drive Running is most likely the best method. I guess we need to figure out how the contactor auxiliary contact could or did get out-of-sync with the VFD. Apparently it was working or limping along before something changed during the storm. The aux contact couild be stuck ON, or the Fan #1 contactor itself could be stuck ON (even with no "Engergize" signal). That would explain why Fan #2 attempts to start when Fan #1 should start. The PLC thinks Fan #1 is already running due to that erroneous signal for the #1 contact. Most likely an electrican with a screwdriver could go out there and jar it loose (unless it is really welded hard) and have the system running normally in about 5 minutes.

But that is a quick fix for a poor setup. Mtn, if you want it to run well and trouble-free, you need to make some changes.

 

[mtn71]

I would love to tinker with it and make it better, but for a fact the boss won't approve any changes to anything other than to just make it work like it did before.

The feedback that tells the PLC if the fan is running is the air flow sensor. We start it in Hand, wait for it to ramp up to speed, then switch to Auto. If the airflow sensor input is on, it should stay on fan 1. If not it assumes fan 1 is down and switches to fan 2. Now when it ramps up, airflow sensor input indicator is on, yet it still switches. This system was brand new less than four years ago, but there are some funky things about it that make me wonder.

I've been going through the program and have found something odd. I attached scans of the sub that controls the fan and damper outputs. This was from when it was running in auto, on fan 2, before it was up to speed so the air flow sensor is not on yet. I'm looking at lines 10, 11, 14, and 15. I don't yet understand why bits B3:1/4 and B3:1/5 are on. They seem to be controlling the dampers, as well as fan 2 turning on, which turns off fan 1. Never mind- should have waited until I looked closer

I haven't gone through the program to the point where I understand what all the timers and bits do yet. Am I missing something?

Added: I don't think any contactors are stuck. The input indicators show them working right when they turn on and off.

 

[Clay B.]

Well if he has a relativity new drive he should have a form relay on it somewhere. I would tie into that for my drive status. Most drives have a run function that changes this relay whenever you go to the run state. Since it is a dry contact you do not have to worry about voltage (as long as its 120 vac or less)

I bet the contacts are welded. Anyway, if it was me I would get it out of there. Just useless junk waiting to cause problems.

 

[Lancie1]

There is a small relay that is controlled by an auxiliary on the fan 2 contactor that is the only thing telling the VFDs to switch to fan 2 instead of fan 1.

The feedback that tells the PLC if the fan is running is the air flow sensor. We start it in Hand, wait for it to ramp up to speed, then switch to Auto. If the airflow sensor input is on, it should stay on fan 1. If not it assumes fan 1 is down and switches to fan 2.

Mtn, Please read your previous comments above, and then tell us how it really is: Switching to Fan 2 is controlled by a small relay on Fan 1 contactor OR if the air flow sensor is on, it should stay on Fan 1. Your PLC program could be looking at both conditions with the switches in series, in which case both devices have to be working correctly. If either one is stuck or malfunctioning, your controls will not work correctly.

Mtn, after looking at the program rungs, here are some questions:

1. You posted Ladder File 4. What is on Ladder Files 2 and 3? Is that where B3:1/4 and B3:1/5 are used? For many Micrologix, Lad 3 is defined as the "Run On Error" or Fault Routine file, so is where the PLC goes when something goes wrong. Check your Status File bit S:29 to see if a Fault Routine is set.

2. What is Input I:0/0 "EMO" on Rungs 3, 5, and 6? Could this input be "the small relay that is controlled by an auxiliary on the fan 2 contactor" (Electric Motor Operator)? I can see that "EMO" (whatever it is) must be ON before you can run in either HAND or AUTO mode. There are a bunch of timers, one-shots, latches, and stuff that could be greatly reduced. Certainly every instruction needs a Description showing what it does. If we can't figure out what it does, then it probably can be eliminated. Nine timers to run two fans and two dampers? A few too many!

3. Do PLC outputs O:0/13 and O:0/15 control the VFD (send RUN signals to the VFD input terminals)? If not, they should!

4. Or do B3:1/4 and B3:1/5 somehow control the VFDs? If either of these bits gets Latched ON, the the only way to get them off is to turn to Hand mode and wait 5 seconds. (Seems screwy - I don't like latched bits - too much chance for them to get left on during a power blink - wait, you did have a power blink, didn't you?) I can see stuff in the program that can be improved. It chases its tail around through a bunch of rungs that could proably be deleted or simplified.

...the boss won't approve any changes to anything other than to just make it work like it did before.

Ask him if he still drives a horse-and-buggy! "Working like it did before" is about a 1910 Model T version. One reason you have this problem now is that it "worked like it did before".

 

[Clay B.]

I see your puzzlement in with B3:1/4 and B3:1/5. They are triggered by another bit then unlatched by a timer. Only reason to do that would be because you want it to sela in for a certain period then drop out. Kid of a convoluted way of doing things. My guess is this is a repair for something that screwed up before.

B3:1/4 can open the damper without turning the fan on. Same goes for B3:1/5

Can you post the program? This is a rather odd method and I can not see everything from the 2 PDFs.

It looks like the latches are mainly being used as a delay stop. Just not sure about the trigger.

 

[mtn71]

tell us how it really is: Switching to Fan 2 is controlled by a small relay on Fan 1 contactor OR if the air flow sensor is on, it should stay on Fan 1.

The air flow sensor goes to a PLC input. The PLC output controls the contactor which in turn controls the relay. This is the ONLY communication the PLC has with the VFDs.

I actually just fixed it. Nope, actually I just got it working as it was. I think truly fixing it would be overhauling the program and whole setup. I am a beginner, but I'm getting the feeling that this system was designed and programmed by an amateur. All I did was switched to fan 1 in Hand, toggled bit B3:1/5. Output O:0/14 turned off, damper 2 closed. I put it in Auto and everything's working right as before.

Ask him if he still drives a horse-and-buggy! "Working like it did before" is about a 1910 Model T version. One reason you have this problem now is that it "worked like it did before".

You don't understand. I wish I could explain but I can't. All I can say is that we're a small company and things are on the edge right now. Improving this system is the LEAST of his worries right now.

 

[Lancie1]

Yes, I do understand. Been there, done that. You explained one reason that things are on the edge right now, just reversed:

This "Improving this system is the LEAST of his worries right now" always leads to THIS in a small company: "things are on the edge right now." It has always been so. Things never get off the edge because the owner has blinders when it comes to ceratin things: spends money like a drunken sailor on stuff he could do without, but refuses to spend any time or money on stuff that has hurt the business.

All I did was switched to fan 1 in Hand, toggled bit B3:1/5. Output O:0/14 turned off, damper 2 closed. I put it in Auto and everything's working right as before.

This is a very good illustration of "why NOT to use latched relay bits" inside a PLC program. Probably bit B3:1/5 got "orphaned" (left on) during the power blink. After that it was never going to work right unless you somehow unlatched that bit. This is just poor logic, but it is used all too often by beginners and inexperienced programmers.

The air flow sensor goes to a PLC input. The PLC output controls the contactor which in turn controls the relay. This is the ONLY communication the PLC has with the VFDs.

This scheme is nominated for today's Rube Goldberg award!
http://en.wikipedia.org/wiki/Rube_Goldberg_machine

Mtn, We can't rewire it and delete those unneeded contactors, but we here at PLCTalk could help you rewrite the program and clean it up. That would at least be some improvement, and probably would prevent the same problem during the next power blink.