Indirect Addressing Error After Converting from PLC-5 to ControlLogix

Hello I've already come up with a solution to the problem described below, so this is mainly a curiosity had the situation been different. I'm wondering if there is a clean way to solve this should many different values exist and not just a few as in my situation:

Background:
I was faced with an issue where the PLC-5 code for a PLC5/11 used indirect addressing in a copy instruction with "#N[N251:0]:1" as the source and a length of 50. Using the translate tool to convert the PLC-5 code to ControlLogix, I ended up with "N[N251_0][1]" which returns an error. Since there were only 3 possible values for N251_0 (101, 102, or 103) allowed by the code, I set up 3 EQU instructions (each on their own branch) to copy the value of the appropriate N101[1], N102[1], or N103[1] length of 50. This was a quick fix because there were only 3 possible values.

My question is: What if there were many values for N251_0? Is there a way to make the indirect addressing work as it did in the PLC5 code without having to make major modifications the code that was converted from PLC-5 to CLX?

I've attached screenshots of the before and after conversion, and of the solution I used.

Thank you for any input!

In your pictures it shows that N251_0 is an alias to N251[0]

Thus using N251_0 as an indirect yields essentially N[N251[0]] which has double indirect (two sets of brackets) which isn't allowed.

Another way would have been a MOV of N251_0 into a DINT which is not part of an array (let's say TEMP_DINT). Then use N[TEMP_DINT][1]

Welcome to the forum! I believe the issue is that N[N251_0] is actually referring to an element in an array tag named "N" in a logix based processor, whereas the PLC5 referred to numeric data table files. You might be able to do something similar in Logix if you used a two-dimensional array tag (ex. N[102,1]) but I don't think this would be worthwhile as it may use up a bunch of unnecessary overhead.

I tried looking up a similar situation in one of my conversions on hand and came up empty. Looks like you did the best thing. Others may have a different conclusion based on their experience.

For your particular application, I think your solution is ok.

I think a more general solution would need some rewriting. In the attached image:

a) my_array is declared as DINT[4,50]
b) destination is declared as DINT[50]
c) N250_1 is DINT

You'll need to copy N101 to the first dimension, N102 to the second one, and so on. Obviously, N250_1 only can be 0 through 3.

bernie_carlton said:

In your pictures it shows that N251_0 is an alias to N251[0]

Thus using N251_0 as an indirect yields essentially N[N251[0]] which has double indirect (two sets of brackets) which isn't allowed.

Another way would have been a MOV of N251_0 into a DINT which is not part of an array (let's say TEMP_DINT). Then use N[TEMP_DINT][1]

Click to expand...

Thank you for the input! I tried that with a DINT tag as you suggested, but it still returns an invalid expression.

RonJohn said:

Welcome to the forum! I believe the issue is that N[N251_0] is actually referring to an element in an array tag named "N" in a logix based processor, whereas the PLC5 referred to numeric data table files. You might be able to do something similar in Logix if you used a two-dimensional array tag (ex. N[102,1]) but I don't think this would be worthwhile as it may use up a bunch of unnecessary overhead.

I tried looking up a similar situation in one of my conversions on hand and came up empty. Looks like you did the best thing. Others may have a different conclusion based on their experience.

Click to expand...

nhatsen said:

For your particular application, I think your solution is ok.

I think a more general solution would need some rewriting. In the attached image:

a) my_array is declared as DINT[4,50]
b) destination is declared as DINT[50]
c) N250_1 is DINT

You'll need to copy N101 to the first dimension, N102 to the second one, and so on. Obviously, N250_1 only can be 0 through 3.

Click to expand...

Thank you both! I believe you two have the solution should a large amount of possibilities exist

RonJohn said:

Welcome to the forum! I believe the issue is that N[N251_0] is actually referring to an element in an array tag named "N" in a logix based processor, whereas the PLC5 referred to numeric data table files. You might be able to do something similar in Logix if you used a two-dimensional array tag (ex. N[102,1]) but I don't think this would be worthwhile as it may use up a bunch of unnecessary overhead.

I tried looking up a similar situation in one of my conversions on hand and came up empty. Looks like you did the best thing. Others may have a different conclusion based on their experience.

Click to expand...

Just had the same issue. There is no way in Logix to substitute an array tag (ie N100, N101 etc) with a variable.

I ended up having to have a separate rung for each array.
Using your files: EQU N251_0 = 100 COP X N100[0] etc.

There is a tech note that says if you really want to do that use a 2 dimensional array...no thanks.