What is the best, cheapest, fastest way to migrate from the SLC 500 to a micro 820?
Seriously, since you are asking the question after you some of the surveying is done and you have a roadmap, then I suspect the answer is to open your wallet and hire someone.
TL;DR
Caveat: these are all opinions and WAGs, probably ignorant.
My hourly rates are too high and I am only a hobbyist in this industry, but here is how I would do it.
- The SLC to MicroLogix conversion was straightforward, so start there
- Compare the SLC and MicroLogix PDFs visually, 1-2h
- If it is the same, assume (we have to start somewhere) the MicroLogix will perform identically to the original working SLC;
- I.e. the MicroLogix code is now the baseline for process control
- Understand the process via discussion with owner, 4-8h
- Write a process simulator and digital HMI, e.g. using AdvancedHMI, 10-30h
- It is supposed to be a simple machine
- Add input map to MicroLogix code so it can be controlled from either physical inputs or E/IP from HMI/Simulator.
- Input map may be necessary anyway as SLC and MicroLogix have, I think, I/O scan synchronous with program scan, and Micro820 may not.
- Input tags I:0.0/9, and others, are referenced several times in the current program; it is not worth the time assessing the cases where they change value mid-scan.
- Test the MicroLogix against the simulator 4-16h
- Includes more discussion with OP about proscess and operation
- Develop some test scenarios
- E.g. push this button, set that dial, get result X
- This will probably result in fixes to the simulator, 4h
- Now that there is a working system, port to the Micro820.
- Code already exists, but has errors; resolve them, 8h
- Port process simulation to Micro820, 4h
- Test scenarios from MicroLogix, 8h
- Fix bugs, [?]h.
Total is 43-80h, plus an unknown amount of debugging; double that and add 6 because those are all WAGs. There is still the physical I/O to deal with, but I assume that would be on the owner.
I would bid this hourly with a cap of not less than 20k$, but again I am a hobbyist and my professional rates are probably out of line with this industry. It might come out less (I have a fair bit experience writing simulators), but there are no guarantees. Half of the work is for the MicroLogix case which gets discarded after generating the test scenarios, but going directly to CCW means the code has no benchmark, which will be paid for via more extensive discussions with the owner and the risk of problems, possibly safety-, injury-, or death-related, down the line. I would definitely review my business insurance policy, and consider my inexperience and that risk, before taking such a job.
An alternative to the simulator would be for the owner to hook up a MicroLogix (few hundred dollars off of eBay), and see if it can run the process and generate test scenarios, and do the same with the Micro820 later.
Compared to someone with experience, I suspect my 20k$ is maybe an order of magnitude or more too high, and they would probably be right, so you should be able find someone to do it for less, but this is not a few hundred dollar job.
Other approaches
- Start with the owner discussions and code from scratch.
- Convert 5-10 rungs (maybe?) an hour manually from the SLC PDF to Micro820; there are ~70 rungs so this will still be up to ~15h.
- This assumes the SLC PDF, from the less experienced person at the manufacturer, is correct.
Going back to Post #1, a grand for cables and software to keep the SLC alive doesn't seem so bad, eh? Anyway, my point is that the PLC hardware, whether EZ PLC or Micro820, is probably the smallest part of any conversion cost, unless labor is free.
