RIO troubles

[Williyb]

Jesper, (or any one else)
Can you tell me in your words what this means. This is for the BT's RACK, GROUP and SLOT. I have my RACK at 2, my GROUP at 6 and my SLOT at 0. I'm using 1/2 slot so I now my slot should be 0. But I dont quite understand my group. I understand the Rack is the logical rack as stated in my G file, but does the group in my BT's have to match my group in my G file like the rack does? This is one area I havent understood even when I asked the AB tech line. Alot help they are!

Group

The I/O group number (0-7) which specifies the position of the target I/O module in the I/O chassis.

Rockwell Software 2000

 

[Williyb]

Jesper, (or any one else)
Can you tell me in your words what this means. This is for the BT's RACK, GROUP and SLOT. I have my RACK at 2, my GROUP at 6 and my SLOT at 0. I'm using 1/2 slot so I now my slot should be 0. But I dont quite understand my group. I understand the Rack is the logical rack as stated in my G file, but does the group in my BT's have to match my group in my G file like the rack does? This is one area I havent understood even when I asked the AB tech line. Alot help they are!

Group

The I/O group number (0-7) which specifies the position of the target I/O module in the I/O chassis.

Rockwell Software 2000

Eric,
I been thinking about the rack being bad. If it was I dont think my discrete I/O would be working. It is kinda working but if I disable my logic for my block transfers the "odd operation" of my outputs goes away and I can force these outputs on with no problems. This leads me back to a programming error or dipswitch error. I couldnt check the rack out tonight but I will this weekend providing production slows down so I can.

I hope I dont wear my welcome out with all these questions but I dont see to many topics about RIO on these board.

 

[JesperMP]

Willy (or Jeff ??),

this is how I interpret RACK, GROUP and SLOT:

RACK: 8 input words + 8 output words.
GROUP: 1 input word + 1 output word.
SLOT: Used only for 2-slot adressing.
Slot 0 adresses the first module of two neighbouring modules.
Slot 1 adresses the second module of two neighbouring modules.

Logical rack #0 group #0..#3 must be your physical rack #1.
Logical rack #0 group #4..#7 must be your physical rack #2.
Logical rack #1 group #0..#3 must be your physical rack #3.
Logical rack #1 group #4..#7 must be your physical rack #4.
Logical rack #2 group #0..#3 must be your physical rack #5.
Logical rack #2 group #4..#7 must be your physical rack #6.

Logical RACK #0 group#2 slot #0 must be HSCE in physical Rack #1:
Logical RACK #0 group#6 slot #0 must be HSCE in physical Rack #2:
Logical RACK #1 group#2 slot #0 must be HSCE in physical Rack #3:
Logical RACK #1 group#6 slot #0 must be HSCE in physical Rack #4:
Logical RACK #2 group#2 slot #0 must be HSCE in physical Rack #5:
Logical RACK #2 group#6 slot #0 must be HSCE in physical Rack #6:

And that checks with your program.

I have never had so many block transfers with a 1747-SN, so I am wondering if there is a limit to the amount of concurrent block transfers that the module can handle (just brainstorming).

 

[JesperMP]

Wait a minute, you say that you are using ½-slot adressing.
That assigns two words per slot.
As you have 4 words per physical rack, then the max number of adressable modules is 2. That means that you cannot access the 3rd module, and thats the HSCE !
I think that the racks #1 to #5 are set up for 1-slot adressing.
Check it.

 

[Williyb]

That sounds more logical than any but.. my G-file states that each rack 1-5 is 1/2 slot. Could the ASB's be set to 1-slot? I doubt it but I will check.
I thought the same thing about whether or not I was putting too much data out there on the line. But thats one thing I havent tried yet. This will also be tested when I set my 6th rack up to my 5th rack's addresses and shut my 5th rack down. If this does "fix" my problem then do I need to just run my hard wires to the 5th rack HSCE? It has more than one input and my discrete I/O work just fine? Plus its only about 30 feet away. I have already thought about doing this since this has almost whipped me but I'll double check and let you know what I find. Thanks,


Jeff Williams

 

[Ken Roach]

I've stayed out of this because I'm on the road and unable to do a lot of research, but it's painful to watch Jeff get confused.

The A-B terminology doesn't help; this "Rack, Group, Slot, Chassis " stuff would drive anybody crazy. It comes from a technology that's grown an awful lot since it's introduction in 1978.

First, Jeff, consider calling A-B technical support. Really, they don't bite.

Barring that option, let's make this clear by posting your program. You can strip out everything but your 1747-SN configuration and the block transfer logic.

Your six 4-slot chassis with 1747-ASB's are well within the capability of the 1747-SN scanner. The module can handle up to 32 block transfers total, so you absolutely should be able to do what you need to do.

I can suppose and guess all day long, but until we see what you've done in your SLC program and on the DIP switches of your 1747-ASB adapters, there's no way for anybody to do anything but guess.

 

[Eric Nelson]

Eric,
I been thinking about the rack being bad. If it was I dont think my discrete I/O would be working. It is kinda working but if I disable my logic for my block transfers the "odd operation" of my outputs goes away and I can force these outputs on with no problems. This leads me back to a programming error or dipswitch error. I couldnt check the rack out tonight but I will this weekend providing production slows down so I can.

First off, I know zilch about AB RIO (except what the acronym stands for). I was only taking into account that the ASB and the HSCE both function fine in the 'other' rack. That just leads me to believe that there's nothing wrong with these two items, leaving the RACK as the 'questionable' item (even though it's new).

beerchug

-Eric

 

[Williyb]

Eric, your right, if I dont question this rack I cant rule it out. I'll check it this weekend. I dont know much about RIO either but I'm learning, I hope you didnt take me wrong I was just trying to keep the communication going so I could get a few idea's where to go next.

Ken,
I'll get my program and dipswitch settings tonight.In the mean time I have to have this panel working so I'm going to run a cord to my fifth panel and set up logic to use that HSCE.

I cant believe this is that difficult, but, it is good to have a challenge every now and again!

Jeff Williams

 

[JesperMP]

Jeff,

your G file does NOT say that you are using ½-slot adressing.
Your G file says that you have assigned a ½ rack (4 words) to each device.

I am convinced that you should use 1-slot adressing.
You cannot connect to the HSCE with ½-slot adressing because it would be unreachable.

Whats more, it explains everything!
Its the reason that the block transfers times out, because they just cannot connect to a module with 8 words adress space at the specified location.
And its the reason that the digital outputs in the 2nd module behaves "strangely" when the block transfers are active. Thats because the 2nd module occupies groups 6 and 7 with ½-slot adressing. Your block transfers writes to group 6.

Case closed !

 

[Williyb]

OK hang on, I have set my ASB module to 1-slot addressing and it acted the same. But as I say that I might not have set my dipswitch3-8 correctly so it would accept that change. Hope I said that right.

I'm assumming that's the only place I set for 1-slot or 1/2 slot? Anywhere else? I couldnt find anything about this other than the ASB manual.

I wasnt able to double check my DS settings since they ran production late. But for some reason I remember seeing the other racks set to 1/2 lot. Maybe I transposed 1/2 rack to 1/2 slot.

Ok one more question, since I write to group 6 with 1/2 slot will this group number change when I go to 1=slot?

 

[JesperMP]

½-slot addressing cannot work, because that assigns two words per physical slot.
With half a rack (½ x 8 words = 4 words) for each device you can only connect to 2 slots, and that would make the HSCE unreachable because it is in the 3rd slot.

Wit 1-slot adressing (that WILL work) your physical rack #6 will have the following addressing:
Logical RACK #2 group#4 slot #0 = physical Rack #6 slot #1 (IA4)
Logical RACK #2 group#5 slot #0 = physical Rack #6 slot #2 (OWA)
Logical RACK #2 group#6 slot #0 = physical Rack #6 slot #3 (HSCE)
Logical RACK #2 group#7 slot #0 = physical Rack #6 slot #4 (spare)

 

[Williyb]

Jesper you are absolutely right! Because IT FINALLY WORKS!!! I feel a little embrassed that I transposed my G file 1/2 rack to the 1/2 slot addressing mode. All this trouble for two dipswitches.
Thanks for all your help and everyone else that contributed. I have gained a great amount of knowledge about the RIO from the troubles I had and being able to work through them on this site and with your help! If I could buy you a drink or shake your hand I would. Now all thats left is getting my HSCE to count. Probably some programming bug I missed. I'll let you know how it turns out.