SLC 500 losing program after cycling power

Well, It looks as if I have alot to "look" at when I go in tonight. I will keep all of you informed. Just FYI - the battery light is/ and has not ever come on that I have seen, so i'm thinking since we lost the program with the "old" battery and the new battery then that may not be the problem. To answer Ron's question on how long the line gets shut down for - usually over the weekend, but has been shut down for just a few hours with the same results.
 
Question - I have the program here at home and I'm looking through the I/O Configuration selection under Controller in the COW - if I select the power supply button it shows the rack #, power supply used, estimated loading/margin for rack and a message screen - which states ****OVERLOADED**** Use larger power supply
'Other' card detected- no current load was taken into account for the card(s).
I would be just guessing here if I said this maybe the problem- Have any of you ever looked or seen this type of a message in the I/O config screen?
DB
 
don't be fooled by that ... (MANY people have been) ...

that feature is just a "calculator" ... you have to manually TELL it what model (size) power supply you're actually using by checking the proper "button" ... until you do, it just "assumes" that you have a P1 - and in that case, you WOULD BE "overloaded" ...

put in the model (size) that you actually ARE using and see what you get ... the number that you're looking for (example: 1746-P4) is usually printed on a label right inside the power supply's swing-open door ...

and (anticipating your next question) NO - there is no way to tell it to "remember" the size that you put in ... it always reverts back to the minimum size each time you access the feature ...

trivia: I used to work for an AB distributor - and I'll bet I've had a dozen or so customers call up to order a bigger SLC power supply because their present one was "overloaded" ... this would have been "easy money" - but when I asked them: "How do you KNOW it's overloaded?" - this "calculator" feature was usually the culprit ...
 
Last edited:
Thanks,
Just something I came accross and thought it may be of interest.
Again- I will post my findings in the morning.
I appreciate the help
 
I checked the rack and it was not grounded to tabs. There is a ground now though. There is not any load on the 24vdc from the power supply - the i/o is being fed from a separate PS. I will not be able to try these changes until the weekend when production is down. If all else fails I will replace the PS with a spare that we have.
Just another quick question- with the program offline I verified the project from the tool bar and found numerous errors that were reported i.e. Invalid file type for processor and many unconfigured I/O address'. Could this be an issue when it restarts on power up? However, they are still able to load this program and it runs!
Thanks again for the help guys - I will keep you informed.
 
Try measuring the voltage on the battery just to rule it out. You may have a bad battey connection. Check that the battery is seated tightly. It may be an intermittent lose connection.
 
Are you getting a CPU Fault?
I had a problem recently where I added a new Proportional control Valve with a card that required a 24V supply.
I figured that I could steal voltage from the PLC power supply.
After this, it would intermittently fault the SLC 5/02, and I noticed that it was doing this when I was sending a high signal to the analog output.
I added a separate Power supply, and the problem went away.
Maybe just maybe, when you restart the machine you are getting a surge on the 24V that is more than the PLC P/S can handle and that is what is making it lose the program?
Good Luck
 
otokdb said:
Just another quick question- with the program offline I verified the project from the tool bar and found numerous errors that were reported i.e. Invalid file type for processor and many unconfigured I/O address'. Could this be an issue when it restarts on power up? However, they are still able to load this program and it runs!
Thanks again for the help guys - I will keep you informed.

When you verify the program, you may see errors and warnings. If it's only warnings, then it can still be downloaded and run. Indirect addressing and shorted branchlegs are two common sources of verification warnings. The I/O issues sound like a show stopper, but usually a processor mismatch can be resolved when downloading, as long as the target processor has enough memory.

RS500_Verify_Results.jpg
 
Update:
We tried over last weekend to resolve this issue by swapping out the power supply without any better results. I did find before powering the new one up that the feed power to the slc was wired backwards - they have a filter before the slc power supply and off of it the line side was on the common and the common on the line. I thought for sure that this would take of the issue of losing the program on power down/up but unfortunately this was not the case. The only thing remaining is the rack itself. We do not have a spare 13 slot rack so for now it remains in the state of "do not power down".
 
The other issue with the program errors was because I tried to download the program into a processor at home while troubleshooting - the rack at home was a 4 slot and the i/o config of the program was for a 13 slot causing “unconfigured i/o errors” in the absence of the cards for those particular i/o.
I hate it when I compound the problem!
 
Does anyone have an idea on the cost of a memory module for the SLC 5/04? I wonder if it is cheaper than the rack itself.

I think I know what you're thinking - but I wouldn't recommend that plan of attack ... if the chassis is indeed "flaking out" - then there's no guarantee that using a memory module to "reload the program" after a power cycle will work correctly ... secret handshake: there's more to a chassis than just electrical connections for the modules to plug into ... look at all those chips on the circuit board and you'll see what I mean ...

now I am NOT saying that the idea won't work - only that I wouldn't guarantee it ... (if the chassis is starting to "go bad" what's to say that it won't "go badder" ? ...

question: do you have TWO chassis of the same size in your plant? ... if so, you might consider swapping them out during a shutdown - to see whether or not the "losing the program" problem follows the original chassis ... I know that's a hassle - but it wouldn't cost any money to possibly nail down the culprit ...

and to try to answer your question, I've got an old 2002 price list here that gives the "list price" of a certain memory module as $290 ... the same catalog gives a "list price" for a 13-slot chassis as $685 ... that might give you some daydreaming/ballpark information - but naturally you'd have to check with your distributor for something more concrete ... be sure to have the exact catalog number, series (letter), and the OS (operating system) handy when you call ... there's a vague mention of a $110 "Memory Module Adapter" being required for some models ...
 
Is the SLC 1746-P(n) power supply ground bonded to the 13 slot chassis at the leftmost bolt hole on the bottom? Is that anchor point bonded to the backplate with the paint scraped away?

Does your filter create a power spike when you disconnect power from it?

Can you break the power between the filter and the 1746 power supply before cutting the power to the filter and see if that gives the same results?

The memory module idea is not a bad one, but like Ron pointed out, if its a bad rack, you're gonna need to replace it at some point anyway.

Anyone can restart a SLC with a memory module, which is good, but you can't write the program to the module while running, and you don't always want yesterday's (or last year's) setup data restored to the machine because of a memory or power glitch...
 
I duked it out with a 10 year old SLC 500, brick style processor a few months ago that lost it's program when it was power cycled. But it wasn't repeatable and seemed completely random whether it would come back up or not. It was the 120V model with the internal power supply.

Never did get it licked. After way too many 2AM call-ins to re-load the program, I gave up and went with a memory module. I don't have the quote, but I think it was less than $150. Worth every penny.

I haven't had a call since, but it still bugs me when I think about it. I even went as far putting in a whole new unit, with the program out of a different, but identical, machine, a brand new battery, and a re-worked grounding scheme (inspired by an Aleric post in another thread). And, yet, it still persisted. :mad:

I hate not finding a root cause.... over a long enough timescale, the problem always returns to bite you.
 

Similar Topics

I have a program that I've used 100 times. SQO settings: File N7:0, Mask 0FFFFh, Dest B3:1, Control R6:0, Length 8, Pos 2. Length & Position...
Replies
48
Views
792
Everyone, i am in the process of purchasing the Slc 500 version of software to support what we have and i have a question. Several of our...
Replies
9
Views
673
In a slc 500 plc I am trying to move data with out using a lot of moves. I want to move data from n7:1 to n7:2 and data that was in n7:2 to n7:3...
Replies
16
Views
1,306
Customer has a circa 2004 SLC-500 PLC. Fieldbus is a 1747-SDN DeviceNet scanner. Customer has SLC-500 file (.rss) with no comments. Has no *.dnt...
Replies
7
Views
512
After I tried wiring, I used computer program communication to read the PLC N value, but received a NAK signal. And the DL3500 CHA light keeps...
Replies
0
Views
399
Back
Top Bottom