PLC5 I'm missing something simple wrt program size

[ptine]

Hello guys, I know this is a simple problem but I just can't find the answer. Yesterday we cleaned up the folder we store our .RSP files to on our PC from various PLC5s. Well we ended up with two programs that appear to be identical when we are offline and online. However, they are substantially different in size.....991 KB and 640 KB. The 640 KB program goes online smoothly. We can get the 991 KB program to go online if we open it and then go online. What we are concerned about is the discrepency between the two program sizes. The processor information is the same wrt the checksum, # of words, Data files, overhead and so. How can we compare the two files and discover what we are leaving behind by using the 640 KB file. Thanks for the help guys.

 

[allscott]

Not sure why the discrepency in size. You can go to the tools menu of Logix 5 and select compare projects. You then select the two projects you want to compare and it will find the differences.

 

[TimWilborne]

Could the descriptor file be much larger in that program?

 

[kamenges]

I've had that happen too. Other than being a curiosity to me I've never worried about it. I have never found a case where it has caused a problem.

I thnk TWControls is on the right track, though. I'm guessing that some actions (saving the file under a different name, for example) causes Logix5 to reorganize the file before it saves it, eliminating space and shrapnel that it has accumulated as a result of Logix5 doing it's thing.

As I've said, I have yet to see a case where this has resulted in a loss of anything.

Keith

 

[Mickey]

I've had that happen too. Other than being a curiosity to me I've never worried about it. I have never found a case where it has caused a problem.

I thnk TWControls is on the right track, though. I'm guessing that some actions (saving the file under a different name, for example) causes Logix5 to reorganize the file before it saves it, eliminating space and shrapnel that it has accumulated as a result of Logix5 doing it's thing.

As I've said, I have yet to see a case where this has resulted in a loss of anything.

Keith

Ditto:
I too have seen this without problems. Except when the file was to large for a floppy backup, but we don't use floppys anymore.

 

[ptine]

Tooo slow of a typist. Thanks guys for the qick responses. Here is what I came up with......think it makes a difference?


thanks allscott. Perhaps a little more detail is in order. When I try to open the 991 KB program a window pops up saying that the file is currently Open or a Read only file. Would you like to go work with a copy? I have to choose yes (no bunts me out) and the 991 KB file opens in the offline state. In the offline state when I go to TOOLS/OPTIONS/System communications in the comm path we have "ROCKWELL!AB_KT-1\16" but when I open controller properties....controller communications in the comm path we have "ROCKWELL_ELEC!AB_KT-1\16". This has to be an issue. The 640 KB file that the processor chooses to go online with has the "ROCKWELL!AB_KT-1\16" in both locations. Also, the 991 KB program was the original program that we used go online with but this morning we uploaded the processor image (the 640 KB program) and saved it into the same directory that the original program (991 KB) was in. Everything seems to be working and when we did a compare between the two the only mismatches were in the data tables wrt discrepencies between data. We have noticed that our several of the .BAK files for the same program are also 991 KB. The rather large difference in file size is causing lots of head scratching. Any ideas?

 

[allscott]

Perhaps a little more detail is in order. When I try to open the 991 KB program a window pops up saying that the file is currently Open or a Read only file. ABQUOTE]

Have you already got the file open with another version of logix, possibly over a network connection?

In short, your sweating over nothing. If you can go online with the small copy of the program than it is current. The only difference could be in the documentation for the program, if it is there (and current) all is OK.

In all my years dealing with logix 5 and 500 I have never really paid that much attention to the file size and wouldn't worry about it.

Having said that if anyone knows why the discrepency I would be curious.

 

[Ken Moore]

I've seen this also. Here's what I think does it, on line editing.
If you edit on line and then save the file, it's bigger, than going on line and uploading the latest file.

I could be wrong, that's just my impression. I have not done any testing to confirm or deny this theory.

 

[Operaghost]

I remember something about the .RSP file contains both an offline image and an online image. It's possible there is a difference in the online images that is causing this difference.

However, I have exported and imported files before and had pretty large differences in the file sizes which I always attributed to basic housekeeping and reorganizing. I never really paid much notice to the file sizes.

As to the differences in the comm settings, I wouldn't worry about it. Just use the Communications and System Comms menu options and those other settings become irrelevant.

OG

 

[OkiePC]

The databases are different.

First, go online with the non-matching filename, and then do a save, and answer yes when prompted to upload everything.

Then do a Tools > Compare and compare the two whole projects. That will point out differences in the I/O config, processor settings, and there should be no differences in the ladder logic.

Next, do an Ascii database (*.csv format) export of both RSLogix files, and you'll probably find out that one of them has more documentation. The other may have been truncated automatically or manually, or cleaned up by deleting spare stuff.

You can compare the data bases side by side and find out exactly. You can do this within RSLogix, but I prefer to open the *.csv files in Excel so I can do a better job of sorting and searching for differences.

The only area of documentation I don't know how to export is the I/O config data. That stuff impacts offline file size, but you'll have to compare those details using the compare utility within RSLogix or by manually checking through all the screens for differences.

Note that each address can have multiple database entries, but you may only see one of them while viewing the ladder. There is the (most commonly used) address comment, and also instruction comments. One address can be described with completely separate bits of text depending on how it'd used in the ladder program.

For example, you can call XIC B3:0/0 "Auto Mode" and XIO B3:0/0 "Manual Mode". If the program contains no instances of the XIO B3/0 address, you'll never know that you are using a few bytes to describe it that way.

I usually try to eliminate instruction comments because they can confuse the reader, and just stick to address comments that will appear the same for any instruction that references them. When I read XIO "Auto" I am thinking "When not in auto, then the rest of the rung will happen...".

Another area that can lead to extra database entries and confusion: Timers and Counters comments: I try to only comment the main element name, not each and every bit and sub-element. If T4:0 is "E-Stop Dwell", all references to its bits will get that same label, and I don't really want to see them described differently in most cases.

Of course, RSLogix will prompt you when entering XIC T4:0/Dn to add a comment, but I just skip it.

I will delete those extra descriptions when I find them too. The built in bit descriptors (like EN, and DN) in conjunction with the base element description are plenty for me to understand the code.

Also, it is easy to end up with stray unnassigned rung comments. You can open those up within RSLogix and delete the ones that have no address or file/rung assignment.

In the left pane, scroll down and dbl click to open the various database icons. You can find dead end rung comments, address instruction comments, and address comments there that will affect file size.

At certain times, RSLogix will prompt you to clean up the database and will elminate the unused addresses and their comments. It doesn't always do a thorough job. I prefer to always answer no to this question, because sometimes there are commented addresses that aren't used YET. I don't want to blow away those comments, or the ones for I/O that are actually wired but not currently in use. That is information I want to keep, but other technicians may not realize that and may answer yes to the prompt because it "seems like a good idea" to them to keep things "cleaned up".

Lastly, when RSLogix "zips" up it's "project file package" before saving, it probably does some housekeeping and orgranizng, maybe leaves some padded space here and there, that may explain why a program with just a few additional comments or rungs is several kbytes larger than another.

 

[OkiePC]

 

[ptine]

Thanks guys. We resolved the problem. I truly appreciate the timely responses.

 

[allscott]

Thanks guys. We resolved the problem. I truly appreciate the timely responses.

No problem on the timely respnses, How exactly did you resovle the problem? And what WAS the problem?

 

[rPraveenkum]

Try ot compare both programs you can find the difference between the two files even there is a mismatch in the descriptor

 

[ptine]

As far as resolving the problem: we compared the two programs with the above mentioned techniques and only found differences in the data tables where identical addresses would have a 1 in one program and the other program would have a 0. However, the Program Files all matched 100% so we didn't sweat the small stuff, renamed the smaller program (640 KB) that went online without asking us to upload/merge and all is well again.

I'm at home now and have a bit more time to absorb OkiePC's post. As I am still fairly new to all this so I am constantly finding new features wrt RSLogix 5. OkiePC has further convinced me that there is always something to learn.

For example, you can call XIC B3:0/0 "Auto Mode" and XIO B3:0/0 "Manual Mode". If the program contains no instances of the XIO B3/0 address, you'll never know that you are using a few bytes to describe it that way.

I had to go and play with my 5 that I have here at home to proof that to myself. As usual, once someone points you in the right direction it is much easier to learn. Our programs at work don't take advantage of this helpful feature. What I have been training myself to do is to look at a XIC as the normal state and treating the XIO as the "NOT" normal state. or vice versa. It works for me but after studying OkiePC's informative post I'm thinking that there are better methods to help you navigate/understand logic. I realize that some of you gurus probably think that this request is rather elementary but if anyone has the time/inclination to go into the Database files......and efficient/effective methods of using them throughout your programs I would greatly appreciate the help. Thanks again guys. I'm not sure were I would be without all your help.