Square D rack addressing problems

I know I've probably brought this on myself, but here goes:

I have a client who is making some changes to their operation. It's an old Square D system originally installed in 1991. It's Model 600 processors, with multiple LI's and 14 remote racks with RI's. I have both SFW472 and SFW375 to program with, using a NR&D cable direct to the processor.

They have one rack that gets demoed in the project, but there's not enough I/O in it for all the new additions. It happens that the slot/rack/drop just before it is not fully populated, using only two registers from a six-space rack (53 & 54, in this case).

My idea was, then, to add in my new racks where the demoed rack is (registers 57 - 64) and change the rack addressing to take the unused registers from the previous drop (new addressing 55 - 64). I went into the rack addressing on the Model 600 and made the changes - and all RI's and LI's come up with no errors.

The problem is that the program doesn't see any of the inputs that use registers 55 and 56. Oddly, the rest of the I/O in 57 - 64 seem to work fine, even with the actual I/O modules in different slots than the old rack. This behavior makes me think the rack addressing is fine.

So, the question is why the inputs for the previously unused registers won't pass signal into the program? Am I missing a step somewhere in the rack addressing? Could there be some kind of setting in a control register that's blocking these inputs?

-- Dan --

First look at your rack addressing table. There is a line for Module Info, this is a 2 digit hex number unique to each module type. Your "missing" slots should have a number there other than INT 13 or 14.

Also, did you need to change the addressing in the LI module to accommodate your expanded registers? This would only be needle if the the registers moved from one LI to the another.

Finally, I think there were some rack/module combinations that required a minimum number or a specific multiple of registers.

Hi Jim,

Thanks so much for your response. I'm not at the client site right now so I can't check the online rack addressing. I'll go out later this evening and take a look.

Regarding moving the registers, they are in the same LI. What I am trying to do is move them from channel one, drop two to channel two, drop one. Channel one, drop two has a six-space rack with only the first three slots populated, but the original rack addressing was for the full rack (53-56) - with the fault override bit (8156-13 in this case)forced on in the program. I changed it to 53-54, and set the rack addressing for channel two, drop one to 55-61 (14 space rack) with channel two, drop two set at 62-64 (six-space rack).

I fear your last comment - i.e. minimum number of registes per rack. If that turns out to be the case, well.....

-- Dan --

After much swapping of parts around, I have discovered that the problem was a bad rack. The rack addressing was fine; the coincidence that the first four slots were the ones that didn't work threw me off-track when trying to troubleshoot.

Many thanks to Jim Dungar for his suggestions via PM. Now I can finish this off and move to my next problem child - a Modicon 984.

-- Dan --