90-30 to RX3i upgrade troubleshooting

Hello!, First time poster here. My name is Steve and I work in a manufacturing facility in northern Nevada.


So I'm pretty new to PLC's. I can do basic troubleshooting and program writing but I am far from being proficient. I'm a mechanical engineer and I was put in charge of a PLC upgrade when the electrical engineer at my company left. Basically what we are doing is upgrading the controls on a machine from the 90-30 series to RX3i. We have built a new machine to add to production and this is the machine I am working on.

The program I have isn't anything too crazy, only a couple hundred lines of code total. There's multiple instances in the code where the exact same block of logic was copy pasted, just using different registers, outputs, etc. I have the original 90-30 program and I have converted it to the new RX3i PLC and quickpanel+ HMI. Older standalone items such as the APU and TCM are no longer supported and they did not convert and I was advised to use a profinet controller by the hardware company.

My biggest problem right now is that nothing works on the conversion! I'm trying to come up with a system for attacking this project and after reading through this site I feel that there are members here with the competency to help me. I can post both versions of the code if anyone would like to look at it.

Go ahead and post both versions. I can take a quick look, although, I only have a laptop with PME version 7.0 available until the weekend.
Make the PLC the active target and perform a validation of the logic. The rungs of code that converted with errors will show up in the feedback window. Double click on each error to jump to the rung.
Timers won't convert cleanly. Don't ask me why. Just about every 90-30 program will have timers and the most frequent conversion from series 90 to Rx3i will start from a 90-30. But the engineers at GE chose not to make a straightforward conversion.
Also watch out for the outputs of all of your comparison instructions ( GT, LT, GE, etc). The converted program will very likely have the output connected to the top right node of the instruction block instead of the middle right node labeled with a "Q". The validation will catch them as an unconnected instruction.

Steve,

I've tried the validate tool and nothing popped out as being wrong. Which was quite confusing since I haven't even done the wiring for the profinet high density terminal blocks for temp control or axis positioning yet.... I would guess that I'm doing something wrong

I'm also trying to find the correct cable so that I can connect to a functional 90-30 series machine and verify that the program those are running are the same as the one I was given.

Are there any good tutorials/videos on the proficy me for the RX3i? All of the PLC tutorials I'm finding online are most for RSLogix 500 and DL105 that I already am familiar with. I get the concepts but the procedures for those PLC's don't seem to carry over to proficy very well. Or am I just missing something?

90-30 Logic attached

RX3i Conversion link (too big to attach 408kb)
https://drive.google.com/open?id=0B20U34rY3A6tSVI3T2VSbTV2eTA

As I mentioned in my earlier post, I only have Proficy Machine Edition version 7.0 available until the weekend. Your project files are both version 9.0

Steve, no worries. I appreciate any help I can get. Take your time, I'll update this thread with any progress I make so that I'm not wasting yours.

Check your hardware configuration. Make sure the physical matches the programmed.
I had a conversion that changed to a 16-slot base rack when the physical only had a 12-slot base.

Check your fault tables.

I hate to state the obvious, but check that the PLC is in run mode. The Rx3i CPU has a switch that must also be in run.

I took a quick look at your converted project. When I performed a validation on the "DUALPLATEN" target I got 116 warnings about duplicate addresses which you should double check to make sure they don't cause any problems.
I also found several instances of the problem I mentioned in my earlier post about the comparison instructions. Rung 78 of the _MAIN program block is the first instance. You will need to move the MOVE_INT instruction down so that the horizontal wire leading to it connects to the "Q" node of the EQ_INT instruction. You will need check every comparison instruction in all blocks and make the corrections as necessary.
I'll check for any more anomalies later.

dYoast, yes I checked the hardware and it is correct. I'll look into the fault tables tomorrow with the other items Steve mentioned.

Steve, I'll work on those over the next few days and repost a revised program after I work through everything.


Thanks again for the help guys!

Steve,

When I click validate for the DUALPLATEN target I get 0 errors and 0 warnings.

I copied the original file and converted and validated again. The only errors I get is
Error 8300: Invalid configuration for Ethernet module.

For the HMI I conversion I get 447 errors which are
Warning 7100: Converting DataType of Variable [RI4089] From [INT] To [DINT]
Warning 7123: Quick panel font not compatible, Arial font will be used
Warning 7102: No equivalent for Password list on converted destination target
Warning 7101: No equivalent for [#System_Contrast] system variable on converted destination target
Warning 7103: Converting [#System_Day] to [#CurrentTime.DAY]


I'm not sure why the warnings you are getting are not showing up on my end.

You can set up PME to ignore, to warn, or to prevent duplicate addresses when validating. Your installation of PME may be set up different than mine.
Zero errors and zero warnings means you will be able to transfer the program to the PLC. Go ahead and send it and monitor the results. Did you correct the rungs where there were instructions to the right of comparison instructions that weren't properly connected to the "Q" node of the comparison?

Steve, I updated my previous post with what happened when I made a copy and converted again.

I am currently working on properly connecting the comparison nodes


Thanks for the quick response

Steve,

I have gone through and changed all of the Q wiring on all logic pages. One thing I noticed is that the names of some variables have changed.

For example, on rung 163 of the "JOBSTOR" logic the variable name changed to LegendPlate326 and the pre-conversion name was AI0250. They both have the same reference address of %AI0250. Should I leave this alone or change it back?

Steve

Edited to add: After watching the program it looks like I want to leave it with the LegendPlate naming convention since the reference variables are working.