changing out el3602 for el3602-0002 and updating link

[c-tsoi]

Description

We added the new EL3602-0002 and swapped it for the EL 3602.

Motivation and Context

The old slice was reading out incorrectly and a scientist requested us to change it out.

How Has This Been Tested?

Kaushik and I changed out the slice, logged into the plc and saw that the terminal was reading back a reason value.
Kaushik and Philip will do a checkout on Thursday to verify that it is working as requested.

Where Has This Been Documented?

Screenshots (if appropriate):

Pre-merge checklist

• Code works interactively
• Code contains descriptive comments
• Test suite passes locally
• Libraries are set to fixed versions and not Always Newest
• Code committed with pre-commit (alternatively pre-commit run --all-files)

[KaushikMalapati]

This looks good to me. There's still an xti file for the old 3602
https://github.com/pcdshub/lcls-plc-kfe-rix-motion/blob/9911fb89691ee5acb4ec32989ddcf7e6e246de11/plc-kfe-rix-motion/_Config/IO/PLC%20Rail%20(EtherCAT)/Power%20(EK1200)/Fiber%20Coupler%20(EK1521-0010)/Term%2034%20(EK1501-0010)/Term%2050%20(EK1122)/IM4K2-PPM%20(EK1100)/IM4K2-EL3602-E8.xti
but I assume that's okay since it's not being used anywhere. I will let you know if Philip and I notice anything tomorrow

[KaushikMalapati]

@c-tsoi did you run and push pre-commit on this?

[c-tsoi]

@c-tsoi did you run and push pre-commit on this?

Yeah, I am pretty sure I did. Are edits that were corrected?

[KaushikMalapati]

I tried inputting some values from the el3602 manual into the IM4K2 custom analog input function block and I think there is a discrepency when plugging in max and min values.

https://download.beckhoff.com/download/Document/io/ethercat-terminals/EL36xxen.pdf#page=150

iRaw for 10V should be 0x7FFFFFFF, or 2147483647. With iTermBits=32, fTermMax=10, fTermMin=-10 (not using fResolution and fOffset for simplicity)
fScale := (EXPT(2, iTermBits) - 1), which is 4294967295
fReal := (iRaw / fScale) * fTermMax, which is 2147483647/4294967295* 10 = 4.999
which is half of what we want. Using iTermBits=31 would give us 10, the correct value.

@ZLLentz @jozamudi, can you tell me if this makes sense since you worked on #68 that added the custom analog input block for im4k2?

One of the sanity checks in #68 (comment) used 2^32-1 as the iRaw for a max voltage of 10V, but this seems to disagree with the manual page I was looking at.

[jozamudi]

I think you're right. It looks like the iTermBits should be changed.

[ZLLentz]

I think you're right too. I've gotten this one wrong too many times! Which means, maybe there's a better way to parameterize these that is less prone to errors....

[KaushikMalapati]

Is there a way to make FB_AnalogInput.TcPOU capable of handling DINTs and INTs? For example, can we make
iRaw AT %I*: INT;
link to a DINT instead and let twincat upcast INTs?
We could also change its implementation to something like https://rosettacode.org/wiki/Map_range but use iTermBits to calculate a1 and a2

[KaushikMalapati]

@c-tsoi once you make my suggestion can you makr this as ready for review and then we can merge?