View Full Version : PLC-5/60 power supply problems
May 6th, 2002, 07:47 AM
We are experiencing periodic loss of data with our AB PLC-5/60. This processor (and remote rack with analog/digital I/O) receives its power supply from the city, which is known to be rather unreliable. Currently, there is no UPS to filter the power, and we are wondering if a UPS might prevent this data loss. Indications of data loss are all analog signals (temperature, pressure) go to zero and remain until the processor is turned off/on.
Any help would be appreciated regarding possibilities for this data loss.
May 6th, 2002, 10:26 AM
First and foremost check your rack and backplane. Can the PS,cpu, or cards move out of its slot? We had a PLC5 mounted in a cabinet on the side of an automatic palletizer, the latches that hold the cards in didnt do a good job with the vibration, we suffered numerous problems just like you stated.
If there is a problem with your incoming power I suggest getting a recorder and contacting the power company. Yes it is a good idea to have surge suppression at least but I cant say a UPS will help you. Not much difference with a plc whether you shut off at main panel or if power goes out, on power up it should read the data again if its actually being shut down.
On the other hand we had a PLC5/20 that was AB remanufactured, I spent months either shutting off and restarting to get it to work with analog etc, or just plain reloading the program. That time it was the PLC5 cpu, it finally just quit.
Another thing is look closely at the program, I have seen where 2 things can happen at same time that shouldnt happen at same time and the PLC goes more or less offline....it doesnt know what to do so it stops doing anything in a sense.
May 6th, 2002, 10:37 AM
The analog signals that are going to zero seem to be coming from the same input card. If this is true, and the code is what is causing the processor to get "lost", would this card show a red fault light when it happens? Just curious because it usually seems to happen at night when I am not here, and the operators know enough to turn the processor off and on. I will have them look for the fault light if it might be an indication of a problem with a particular card.
May 6th, 2002, 11:11 AM
You didn't mention the type of input card that you are using. Are the cards 4-20ma or millivolt input cards? If the input card is 4-20ma, and you are using a line conditioner, millivolt to 4-20ma, the problem may be at the line conditioner. Can you see if the raw data is out of range? On a 4-20ma card if the input current falls below 4ma, then your reading a the user interface may say 0, but the card is actually out of range and in error.
May 6th, 2002, 12:29 PM
1. The analog input card in question is a 1771-IFE.
2. The analog input card is in a remote chassis on RIO.
Have your maintenance guy on night shift look at the card when the data goes to zero. If the green RUN light is blinking, what has very likely happened is that the remote rack has lost it's power supply momentarily. If the red FAULT light is on then it's a hardware failure and you should replace the card.
The 1771-IFE communicates with the PLC-5/60 by way of a special PLC-5 instruction called the "block transfer". A block-transfer Write must be sent to the module after it is powered up to configure it. After that, block-transfer Reads must be executed periodically to read the analog input data out of the card.
The most common mistake I see with the 1771-IFE is that the Block-Transfer Write (BTW) is only executed once: when the PLC-5 powers up. If the remote chassis has it's power cycled, the 1771-IFE will lose it's configuration and need a new BTW sent to it.... but the PLC-5 programmer often doesn't allow for that.
Go through the ladder logic and find the BTW being sent to that particular card. I'll bet it only executes on first scan, and not every time the BTR returns a status word with the "Powered Up But Not Configured" bit set to 1. That's the time you should be re-sending your 1771-IFE configuration BTW to the module.
This is just a guess, but it's a mistake I've made and seen made numerous times.
Rockwell Automation / Seattle
May 6th, 2002, 02:04 PM
Ken, I suspect the assumption that the BTW executes only once is true, and that is necessating the turning off/on of the processor, to get the card configured again. But, while this may alleviate the problem, it does not get to the real question...why is this card cycling power in the first place. If it's only momentarily, it may be nothing more than inconvenience and a BTW will fix the problem, but if it happens long enough our alarms and interlocks will start triggering if zero values are registered.
I will start by investigating this BTW command.
May 8th, 2002, 11:29 AM
To Whomever Can Help
Regarding the above messages about the BTR/BTW operations from our processor to the analog input cards, I looked at our program and I believe the assumption is correct in that the BTW is only executing on the initial power-up of the processor.
Before I make an error, I would like to get some advice. My solution would be to look at the BT9:3/ER bit to indicate that the 1771-IFE has had it's power cycled. If this bit is set to true, then I would run the BTW again. Above, it is indicated to run the BTW bit when the bit "powered up but not configured" was set to true. Is this the .ER bit? How else would I know when the power has been cycled?
I appreciate any help, I am more of a DCS person than a PLC person, so I am trying to be as cautious as possible to prevent an error. I see a lot of CAUTION statements in the AB instruction books regarding BTR/BTW operations.
May 8th, 2002, 03:17 PM
The .ER bit isn't the one.
When you let RSLogix build your BTR/BTW instruction, the code that it generates has the proper syntax & address (I don't have the software with me, so I can't just post it here.)
Create a new project. On the Project Window, go to I/O Configuration Add an analog input module to one of the slots. I forget if you have to double-click on a channel to open the slot configuration or not. Double-click on the card. A window will open asking for the BTW control and data, and the BTR control and data. Fill those in with your addresses from the real program.
It will open a dialog box with details about the module configuration. There will also be a button for Create Ladder. Click on that, follow the prompts (where you want it), and the proper ladder will be created.
There is an optional address for a "Manual Configuration" bit (again, I forget what's it's called. This bit is used if you want the program to do the BTW at some other time than on powerup or "loss of configuation" (like perhaps the .ER bit, or on a recovery from a rack fault). I've never had to use it, but you might want it, at least to manually toggle to verify that a BTW will fix the problem.
Hope this helps. If you have problems, I'll post later when I can look at RSLogix5.
May 8th, 2002, 04:15 PM
The Power-Up Bit is not part of the BT control file (like the .ER or .DN bits are). Instead, it's part of the body of the data transferred from the 1771-IFE to the PLC-5 by a Block-Transfer Read instruction. You have to have a successful BTR before you get that data.
If, for example, the destination of your BTR instruction was the 22 integer words addressed N40:0 through N40:21, then word N40:0 is the first word of the BTR destination data file.
It's first bit (N40:0/0) is the Power-Up bit, which is set only when the 1771-IFE has been powered up but has not been sent a BTW instruction to configure it. This is the bit you should use as a condition for the BTW instruction that configures the module.
http://www.ab.com/manuals/io/1771/177165115.pdf is the user manual.
There's some difference between Series A&B and Series C modules of this type so try to find out which one you have. The size of the BTR should be the giveaway; 20 words for Series A and B and 22 words for Series C.