Koyo MODBUS help...

This question is probably best answered by Tom (Jenkins), but others may be able to offer suggestions. Tom, after reading a few of your posts over at the A-D forum, I gather that you're familiar with using MODBUS with A-D stuff. I'll probably ask this same question over there, but I'll probably have to wait a week for a usable answer...

I am by no stretch of the imagination very familiar with MODBUS, but I have been sucessful in the past using MODBUS to talk to A-D drives and HMI's, but only because they give step-by-step instructions and examples for these specific combinations...

Now I'm trying to communicate with a Trio motion controller via port 2 of a DL260 CPU (RS-232). So far I've had no luck. I'm positive I've got the wiring right, so I think I just have the communications setup wrong.

As it stands, I have the Trio setup for MODBUS communications at 9600,8,N,1. It's station address is 2. I'm confident that the Trio settings are correct as I've used the Trios before with a Maple Systems HMI, though that used RS-485 (and I had more step-by-step and examples to help).

If it helps, the manual for the Trio can be found HERE. The MODBUS stuff is on pages 3 thru 6 in Chapter 13 (Adobe pages 475-478). Sorry for the huge manual... The pdf is encrypted, so I couldn't extract only the pertinent pages. Although, I DID save you the trouble of registering with Trio to get the manual by putting it in my webspace...

I think my problem lies in the DL260 port 2 setup and/or my use of the MRX command. Here's my port 2 settings:
[attachment]
Do these look correct?

Now that I've used up my ONE attachment allowance, I'll continue on my next post.

beerchug

-Eric

Continued...

Here's my MRX instruction:
[attachment]
As a test, I'm trying to read the Trio's inputs status for inputs 0-9 and put them in C300-309. The Trio manual states that 10001 is mapped to IN(0), 10002 to IN(1), etc. This (of course!) is not working. Many of these inputs are ON, but are not reflected in C300-309 (all are OFF). The port 2 'busy' bit (SP116) turns on/off as the communications retrys after each time-out.

I have also tried reading 10 of the Trio's holding registers (40001-40010) and putting them in V3000-V3009. Same lack of results...

One thing I noticed that may point towards a solution. When I download the program with this MRX instruction, disconnect, then reconnect, I get a mismatch error. The difference is that the "Start Slave Memory Address" has now changed from 10001 to 100001 (or 40001 to 400001 for the holding registers test)...

I have a feeling that this is where the problem lies, but why the heck does it change? Should I be using 484 mode instead?

Any insight on this will be greatly appreciated! I can live without this communication, but it would make my life much easier if I can read/write to the Trio.

Thanks,

-Eric

10001 and 100001 are exactly the same address as well as 40001 and 400001. If you analyze the protocol strings sent from the PLC, you can see the strings are exactly the same (0x0000) for the address whether you have 10001 or 100001 as well as 40001 or 400001.
You may want to look at the exception response v memory (V400 - V402 in your setup) and see what the error is. That may indicate something to you. The error byte will be something like 8x the x being the actual error.
You can either go to modbus.org website to get the modbus manual to interpret the error or simply post in the forum and we can tell you what the error is.
If there is no data in V400 - V402, then the Trio device is not sending any data back at all, which would still seem to lead towards configuration setup or wiring.
Maybe some of this will help...

Your main change from the previously sucessful implementation with the Maple Systems equipment is going from RS-485 to RS-232. Are there any lights on the Trio to verify that it receives the request and/or transmits a reply. If there is no reply check for the possible need for RTS-CTS hookups or possibly there is a setting in the Trio which will cause it to ignore the RTS/CTS states and reply anyway. And secondly, the shorter address is the 484 mode, the longer is the 984 mode (these units have more memory space). Your instruction adressing is merely being corrected for you free of charge.

The last time I used Modbus on the Koyos the special MRX commands weren't successful. I used standard RX and WX commands. I'll dig out that project and try to post some info tomorrow (I'm bidding a job today!). In the meantime, try reading the data into a V-register instead of directly into the coils on the Koyo end. Then try a modbus word into a Koyo word. The idea is to see if you can at least verify the communications protocol is working, and then worry about correct register assignments.

All good info...

Sorry I forgot about you Bernie. "Otto" avitar = Bernie can help too!...

I have the RTS/CTS jumpered at the DL260 end, but there's no connections for them at the Trio end. And, of course, there's no comm lights on either end... That's one thing I like about the DL06 units...Lights!...

Thanks for the 'exceptions' tip, Jackson. Now that I know what it's for, I'll have a look at the buffer.

You have a good point about the MRX/MWX commands, Tom. This is my first time trying them, so maybe they haven't got all the bugs worked out yet? I'll try going back to RX/WX and see if THAT works any better.

Thanks for the tips guys! I'll play around some more tomight and report back with the results.

beerchug

-Eric

Progress!...

I took your advice Tom and went back to the good ol' RX instruction and am now getting data!. I still have to decipher this data, but at least NOW it seems to be communicating... :nodi:

I'm too busy with other stuff right now to spend any time on this, but at least now I know I'm on the right track...

Thanks guys!...

beerchug

-Eric

P.S. to Jackson... When I ran the MWX instruction, the 'Exception Response Buffer' contained the hex value 'FCFC'.
Any clues?...