RIO failure after CLX upgrade from PLC-5

Upgraded a PLC-5/80 today to a CLX. There are 5 physical ASB modules for a total of 10 racks. All 10 are added in the CLX tree under the DHRIO CH. A, at 115.2k baud, which is what the PLC-5 ran at. The main PLC-5 chassis was rack 0&1, so I made the ASB module rack 0 that went in it's place. The PLC-5 was end of the line, so I have a 150-ohm resistor there, and one at the other end (another ASB module). The new DHRIO is in the middle of the network, but the way it was added has me concerned. The nearest node was far away, so the blue hose was run 150' from a control cabinet to the DHRIO card, and then 150' back where it was spliced with wire nuts. I don't like this setup, but the PLC-5 seems to be OK with it. The DHRIO card will flash red for CH. A, indicating adapter fault, according to the manual, but none of the adapters are faulted. Every PLC-5 chassis was power-cycled after switching to the DHRIO card, yet the DHRIO CH. A flashes red. Occasionally, we saw it say CHAT LINK under CH. A, which I read means "babble" on the RIO network. I couldn't get any further info about this. Two Rockwell techs had no idea what this could be. DHRIO card is brand new.

We had to revert to the PLC-5 so the plant could run, but we'll need some options before we attempt this again.


BEFORE:
PLC-5 Rack 0/1 <> Rack 2/3 <> Rack 4/5 <> Rack 6/7 <> Rack 10/11

AFTER:
Rack 0/1 <> Rack 2/3 <150' blue hose> DHRIO module <150' blue hose> Rack 4/5 <> Rack 6/7 <> Rack 10/11

For 115.2 kbps and 150Ω resistors at both ends you should be good up to 5000' (1524m) so the length is not an issue; the wirenut splice maybe(?); PLC5s were/are true workhorses...Especially the 5/80s...

You are also under the device/rack limit (16 devices and 16 racks) for 1771-ASBs...

Just for clarification, why are you using Channel A as a RIO Scanner? Is Channel B also used as a RIO Scanner?

Otherwise, if only one channel is to be used for RIO the user manual specifically states :

"...If only one channel is needed for RIO, use Channel B...."

IMPORTANT at Page 104 of :

http://literature.rockwellautomation.com/idc/groups/literature/documents/um/1756-um514_-en-p.pdf

I'd try it again for Channel B this time if both channels are not configured and used as RIO Scanners.

We tried Channel A initially. We switched to Channel B as a test, but got the same 16#0312 errors on all the racks. I missed that fine print on Pg.104 about only using B. I suppose they could have made it more of a bullet point in the configuration setup. However, since it failed that way too, I don't think that's our issue.

Would you agree that the PLC-5 is more capable of handling RIO communications than the DHRIO module?

phuz said:

Would you agree that the PLC-5 is more capable of handling RIO communications than the DHRIO module?

Click to expand...

PLC5s were designed specifically for RIO communications.

As for the new RIO Error information, have you followed the mitigation suggestions from:

1756-DHRIO / 1771-ASB: Error 16#0312 Invalid link address
65975 | Date Created: 01/05/2010 | Last Updated: 03/24/2017
Access Level: TechConnect

https://rockwellautomation.custhelp.com/app/answers/detail/a_id/65975

Yes. Sure have! I've setup a few of these in the past, and aside from someone rolling the wires, they just work.

I'd say more than 75% of the RA 16#0312 errors I have ever experienced are either media or static/hard configuration setup issues...The rest were hardware related...

Do you have another 1756-DHRIO?...If you do, I'd try it with Channel B as RIO Scanner to begin with...

I really don't know what happens if you try to run a DHRIO with only Channel A connected to a RIO network...If it is 'Important' to not do so then hardware failures might occur......

dmargineau said:

I'd say more than 75% of the RA 16#0312 errors I have ever experienced are either media or static/hard configuration setup issues...The rest were hardware related...

Do you have another 1756-DHRIO?...If you do, I'd try it with Channel B as RIO Scanner to begin with...

I really don't know what happens if you try to run a DHRIO with only Channel A connected to a RIO network...If it is 'Important' to not do so then hardware failures might occur......

Click to expand...

Unfortunately we do not have another DHRIO with us. I'm surprised that neither of the RA techs I spoke to even thought to mention about using Channel B.

I dunno what to tell ya...I have replaced quite a few PLC5s, however, I mostly used the 1756-RIO modules.

The DHRIOs I have installed were mostly for DH+ to Logix communications Pass-Through or Logix Gateways and for both DH+ and RIO setups; never been in your scenario...

I'd try using a 1756-RIO if you don't need a DH+ interface...I am a little weary of 'multitasking' networking automation hardware...

+1 for the 1756-RIO, though all of our experience is with the module in adapter mode.

If you can scrounge up a 1771 chassis, ASB, and some I/O modules, consider a test setup with your CLX system before the next upgrade attempt. You can verify CLX hardware and wiring strategy is OK on the bench. For example, we have been bit on swapping polarity with RIO on non-1771 devices.

Check the wiring polarity carefully.

On the 1756-DHRIO when you are using a channel as a RIO scanner, the order of the conductors should be:

2 - Clear
- Shield
1 - Blue

On the PLC-5, the connectors face the other direction, so it's the same color code and pin but they're reversed top-to-bottom.

1 - Blue
- Shield
2 - Clear

In my experience, the 1756-DHRIO isn't as robust with poor signal quality as the PLC-5 was. It's still good, and within the specifications and standards and tests, but the PLC-5 was great.

Get an oscilloscope and look at the signal amplitude and quality with the PLC-5 in place, then again with the ControlLogix 1756-DHRIO.

I used to have an SST DH+/RIO analyzer mounted in a "luggable" DOS computer. Unfortunately the tweaker who stole it also stole all my hardcopies of my RIO signal analysis notes.

I brought the DHRIO and ASB home to test in my PLC5 rack and CLX rack. It worked immediately so it looks like a signal issue with the blue hose. Maybe the wire but splice? Maybe another point? Who knows. But these modules are not complicated and I guess we just proved how robust the PLC5 is.

PLC5 does seem to be a lot more robust than CLGX cards on RIO and DH+.

It's poorly documented (not consistent), but I've seen Rockwell recommend 82 Ohm resistors, even at 56K and 115K IF all devices can support 230k. See Answer ID 42505 and the 1771-ASB manual (1771-um001 page 2-14)

I would check the following:

Any single strand shorts at any terminations?
Verified all connections are tightened?

Verified all wires are on the right terminals on all the devices? I had a DH+ network that worked fine at 57K, but failed at 115K. Shield and Clear were switched at one of the PLCs (which is understandable because the installer used clear heat-shrink on all the shields). I checked this one by tying a known shield to ground at one point. Immediately an ohm meter showed blue and clear as grounded (blue to clear will normally show shorted to each other as each node has 4 ohm resistance between them, so this is expected). I went around and unplugged devices one by one till the short went away. Dig back the insulation on that node and sure enough clear and shield were backwards.

If 1771-ASB are they daisy chains on the same terminals or in on 1-2-3, out on 4-5-6?

Can you get one node to talk? Disconnect everything but Rack 2/3 <150' blue hose> DHRIO module

then

DHRIO module <150' blue hose> Rack 4/5, (with resistors on both ends).

Verify the ASB at rack 0/1 is configured and wired right. Can you unplug that rack, remove those racks from your CLGX scan list and see it? Can you connect a test CLGX-DHRIO rack at this point to the existing network (perhaps at rack 0/1) to eliminate the 300' of new blue hose.

Power off all racks and power them on one by one till the network dies.

Replace that with breaking the network and extending it one node at a time (being sure the network is always terminated).

I just went through this. Changed out a SLC 5/05 with 6 SN modules!
Had to put the 150 ohm at the ends of all the links for sure. the older RIO modules are a lot more forgiving then the DHRIO. Only other problem I had was block transfers to HSCE2 cards (5, with 3 on one SLC rack). After working with AB, the basic problem is the SLC had a 245ms scan time and the Logix has a 9ms scan time. So using the old code that had the XIO of both read and write enables was just choking the RIO buffers. So I had to use 100ms timers to do the triggering. Also had to do the multi-block configurations one module at a time. Some of the DHRIO had VFD's with 20-COMM-R modules. For those I took advantage of the fact a DHRIO will go up to rack 76 (SLC SN only goes up to rack 3) and just re-addressed them to 1/2 rack each so no block transfers were needed. 62 VFD's later my fingers were killing me from those stupid HIM modules buttons you have to stand on to work.

So my main lessons here were:
1. Use the 150 ohm at each end, even if it's 10'.
2. Sloooow down block transfers.

Jack.

Excellent thought JaxGTO on Block transfers.

DHRIO has a limit of 16 Block Transfer buffers per remote I/O channel, meaning you have to be careful with what you're firing at once, and you definitely have to be careful to avoid firing a BTW and BTR to the same at module at once. I would start by testing a program with discrete IO only to make sure the physical wiring is good, before adding and testing interlocking logic for multiple Block Transfers. I would avoid sending more than one per rack at once.

Somewhere in the KB Rockwell has a sample of how to interlock MSG instructions.

Well, just wanted to update you guys and hopefully you'll get a kick out of this as I did. We were back on site to try this again, and determined that the electricians terminated the RIO wiring at the wrong area and it was never connected to the DHRIO module. Now it works.