Blue Hose via 1756-DHRIO to L81E

[ruairicussen]

Remote I/O



The occasional times i have had to deal with blue hose/PLC5 I/O addressing it always wrecks my head. Now i am working on a ControlLogix program (L81E v34) using a 1756-DHRIO to integrate existing RIO currently controlled by a PLC5. The CLX local chassis (17 slot) has 1756 cards to connect all the PLC5 Local Chassis I/O (PLC5 was also a 17 slot local chassis). Local I/O is no problem, but i was hoping an expert could look at my control logix I/O tree RIO configuration and tell me if it looks right. I also have some confusion about how some of the "1 Slot" "2 Slot" I/O is coming in in the running PLC5 system. Attached are screenshots of the PLC5 Chassis Table and the Channel B RIO i added to the CLX I/O tree.



Question 1 - PLC5 Local Chassis is listed as 8 slots, actually 17. I assume this entry just doesn't matter for the PLC5 Local Chassis? The local chassis IO consumes I:000 - I:017



Question 2 - PLC5 Chassis 2 is set to "2 Slot" Addressing, but the Input and output data do Not use the same locations in the Input and Output Files [PLC5_Rack2_IO pic attached]


Question 3 - PLC5 Chassis 7-16 are set to "1 Slot" Addressing, but the Input and Output Data DO use the same locations in the 2 IO files ("Span" of only 2 Groups, but there are 2 words each of Input and Output Data) [PLC5_Chassis16_IO pic attached]. Only attached one but they are all similar. These are 1336's via 1336-GM1 RIO adapters.



Question 4 - is generic "RIO ADAPTER" the right adapter to use in the CLX I/O tree under ChannelB of the DHRIO to represent a 1336 with a 1336-GM1 adapter? or is there a module profile i should be adding to studio? I would not be too worried about this if the "2 Slot" addressing of those Chassis in the PLC5 did not seem to act differently than the "2 Slot" addressing of Chassis 2 in the PLC5. There seems to be very little config available on the 1756-DHRIO side of things.



Question 5 - is there any change to LSB MSB order of I/O under the 2 systems, or anything else like that that would be a major mistake to assume the IO data will come in to the same bit in the CLX words as it does in the PLC5 words?



In my confused brain i cannot find a single thing here where one chassis setup vs. result does not contradict another chassis setup vs. result in the PLC5 RIO.



Thank You in Advance for any assistance,

Ruairi

 

[Ken Roach]

Welcome to the PLCTalk forum community (as a poster, anyhow !)

Your PLC-5/30 system condensed quite a bit into the limited RIO capacity (7 "logical racks" of data) of that controller.

And you're right that the mixing of the terms "chassis" and "rack" can get confusing, when applied both to physical and logical arrangements of devices. A physical "slot" is both a single place where a module can plug in, but is also a logical memory space that can be divided into 1/2, 1, or 2 logical parts.

It is my understanding that the late-1970's PLC/PLC-2 had 8-slot physical chassis, so 8 slots = 1 rack is the logical basis for A-B Universal Remote I/O.

Since the physical 1771 chassis where your PLC-5 was installed had 16 I/O slots and was set up for "1 slot addressing", the modules in it were addressed as both logical Rack 00 and Rack 01, even though it's one big hunk of steel.

The RSLogix 5 screenshot you posted that shows the local chassis as a 1771-A2B (physically 8 slots) with 1-slot addressing, and omits mention of Rack 01, is merely incorrect as an organizational tool in the RSLogix 5 program.
Those settings for the PLC-5-resident chassis do not affect the actual PLC program, since the PLC-5 can't tell how many slots are actually in the chassis it's mounted in. If there had been a logical Rack 1 connected to the RIO network, it would have conflicted.

But you've probably got all that handled when you re-addressed that portion of the PLC-5 Input and Output data table to the .Data tags of the 1756 modules in your ControlLogix chassis.

So Question 2 is about "Rack 2", which is configured in RSLogix 5 as a 12-slot chassis, configured in "2-slot" mode.

2-Slot addressing assigns one 16-bit word of Input data and one 16-bit word of Output data to every two slots in a 1771 chassis.

Typically you use it by alternating Input and Output modules. A 1771-IAD and 1771-OAD mounted in the first two slots of a 1771 chassis at Rack 2 would use I:020 for the Input module, and O:020 for the Output module.

That's not consistent with what you're showing in the PLC5_Rack2_IO screenshot. That shows data usage that's more typical of 1-slot addressing, where a module uses only Input data or Output data for each phsical slot. An Input module in the first slot uses I:020, and the output data is unused.

Ideally, take a physical inventory of those 1771 chassis to give yourself some more context. I wouldn't be able to be certain about your system without an inventory of modules and some cross-checking.

Question 3: that's normal for "1 slot addressing" when you're using 1336-GM1's. They emulate a "1/4 rack" device at a minimum, and they might have been auto-detected like they were one of those special 2-slot "1771-AM2" adapters.

The 1st Word of Input is Status, the 1st Word of Output is commands. The 2nd word of Input is the feedback speed, and the 2nd word of Output is the speed reference (scaled 0-32767 = Min Hz to Max Hz, if I recall correctly).

If you want to do more with those drives (and especially if you're using block transfers in your PLC program), we should talk about reconfiguring and readdressing them.

Question 4, you're right on. A Generic RIO adapter object in the 1756-DHRIO scanlist is what you use for 1336-GM1's.

Question 5, don't worry. The LSB/MSB and bitwise order is the same for Universal Remote I/O, from the 1970's to today.

Let's take a bit of a step backward: is this a migration that's in process, or one that you're maintaining and troubleshooting ?

 

[ruairicussen]

Ken,

Thanks for the response and clarification of some of those points. I am glad i am not crazy thinking the Rack 2 Input/Output file usage seemed odd for 2 Slot Addressing. What i am doing on this one, something i have never done before (i believe in new conduit, new wires, new enclosures, new hardware on each and every upgrade project) is using the AB adapter kit things to quickly mount a CLX Rack in the PLC5 rack, including their adapters comprised of an RTB that mounts to the CLX modules and a modular plug at the other end of a short belden cable to plug into their PLC5 I/O Wiring adapters. All of this is AB factory pieces. Makes for i would guess a 30 minute to 1 hour physical swap of the Local Rack hardware.



Reason i am doing it this way in this case is in the next 2 years we will be buidling an entire new line to replace this one. This is a short term bridge to allow controls upgrades in the meantime (vision cameras and Servo drives/motors). I would hate to spend probably $500K or more rewiring a line that will be knocked down in a year or 2.



Except for the Local Chassis and the 1336s, all the other Chassis are actually Flex I/O with 1794-ASB adapters (i missed this when i made them 1771-ASBs in the CLX I/O tree).



Attached is the real world makeup of Chassis 2 that the guys on site put in excel for me. I guess i should have used the generic adapters there too for 1794-ASBs? Obviously the #46 (these numbers are written on the Flex modules with sharpie, and look like they are Rack/Group) module is mislabeled, must have been moved from a different console at one time.

Thank You

 

[ruairicussen]

To directly answer the question you asked, this is a migration in progress.

 

[Ken Roach]

Excellent context and info, thanks !

When you get a chance, take a camera and a notepad and go get all the DIP switch settings from those 1794-ASB modules. Make a note if you can about the Series letter; Series C, D, and E all had new module support and features, but were backwards-compatible with Series A and B.

You need to know which Series you're using in order to set or interpret the DIP switches correctly.

It's my guess that the RSLogix 5 entry for Rack 02 is simply incorrect.

Which does happen ! RSLogix 5's I/O configuration was largely for the programmer's reference, and could be inaccurate if it wasn't the result of Autoscan of the RIO network, or the source of auto-generated BTR/BTW instructions for analog modules.

For Rack 02, the arrangement of the modules, and their address usage, is consistent with the default "1-slot" addressing used by 1794 FLEX on RIO since the early 1990s.

My guess is that you have a "Right-to-Right" 1794-CE1 extension cable between those two banks of modules. When FLEX modules are installed with one of those cables, the slot numbering follows the order relative to the path from the network adapter.

ASB → 0 → 1 → 2 ↓
          4 ← 3 ←

 

[Ken Roach]

One more thing for this evening, before we gripe about the automatic conversion between PLC and Logix: block transfer budget.

If you have more than sixteen analog modules, you should consider a different RIO migration. The ControlLogix 1756-L1 in 1998 had enough buffers to handle about sixteen analog modules through the 1756-DHRIO, and that element of the operating system simply hasn't changed.

Included in that count of analog modules is the 1336 Plus drives if they are using Block Transfers to read and write parameters from the drives.

Because you have more RIO scanning capacity in terms of absolute Rack space on the 1756-DHRIO than you did in the PLC-5/30 (up to 32 logical Racks per channel and/or up to 32 physical adapters per channel) you could set up the drives as Full Racks and move data back and forth using Data Links, rather than block transfers.

Handy discussion of some of the DHRIO limitations (and the 1756-RIO module, now available as an AN-X network gateway):

https://rockwellautomation.custhelp.com/ci/fattach/get/91815

 

[ruairicussen]

Great, thank you. I am a lot more comfortable now, with your answers about the Fact that the Chassis Table does not necessarily have to be correct for things to function (removes my 2Slot problem) and that the 1336s will work with the generic RIO-Adapter.



I assume i will want to change the 1771-ASB adapters in the CLX DHRIO tree to generic rather than 1771-ASB?



I appreciate all the info, it is much more than i have been able to get through distributor channels (they are normally very good but i think, like me, they are much more up to speed on current generation hardware).

 

[Ken Roach]

Configure the network adapters in the 1756-RIO Scanlist as close to their actual device type as you can. There might be some low-level protocol details that make it matter.

I am, both personally and professionally, previous-generation hardware and technology. In-depth network diagnostics and PLC-5 migrations were a big part of my job for a decade, including covering the Olympic Peninsula.

 

[ruairicussen]

There were (are) only 2 Analog modules getting BTW / BTR in the program. One was an analog input module in the local chassis that got current from CTs on 4 large motors. Since Analog nothing is available for AB right now we are replacing the 4 old SMC soft starts with current gen SMCs with 20COMME and getting current over Ethernet/IP. The other is an 8 point thermocouple flex io module on a 1794-ASB. SO i will have only 1 analog needing BTW/BTR in the new program (i guess now they will be a MSG instruction configured for BT, havent looked that far. once i get past RIO ignorance this is a simple process/simple program).

The drives are being used pretty dumb, Start/Stop + direction on some, Running status bit, Frequency command. That is all. I did install new conduit and fiber to Stratix switch at the MCCs, so if one of these drives fails it's replacement will be a 755.



Are you still doing that work now?



Thanks again. i really appreciate the help and knowledge.

 

[Ken Roach]

Oh, good, it sounds like you've got the project well in hand.

I left Rockwell almost twelve years ago, so most of what I do is just one generation off the cutting edge, in the aerospace OEM field.

But I spent plenty of time counting in octal to figure out FLEX I/O addressing !