[Solved] Warning messages on trace lines

http://www.merg.org.uk/merg_resources/cbus.php

[Solved] Warning messages on trace lines

Postby BrianW » 23.03.2014, 21:32

I have a small layout running MERG CBUS with 3 sets of points operated by TowerPro 90 servo motors via MERG Servo4 boards.

Whenever a points servo motor operates (either from a Rocrail command, or direct reed switch command) the following line appears once, or sometimes several times, on the trace.

date/time stamp r999W cbreader OCBUS 1505 no start byte [0xFF]

Exactly the same message (except date/time) for each of the 3 servos every time.

If another CBUS message is transmitted at the same time, it seems to get lost.

As two of the servos are operated by track reed switches and magnets on locos (no connection to or control via CBUS), the 3 servos are on searate baseboards with separate Servo4 boards, and the same result occurs whether connected to Windows XP desktop or Windows 7 laptop, both running different releases of Rocrail, I suspect the errant messages may be noise pickup, either power supply or RF. (Power supplies are well smoothed with decoupling capacitors on all modules).

At the moment it's a minor niggle, sometimes resulting in a sensor off event being missed, which eventually stops the layout until the sensor is manually reset. However as I intend to expand the layout to more than 20 points, the problem is likely to get worse.

Monitoring the CBUS signals with the FLiM utility shows all sensors working every time and no apparent 'noise', however Rocrail seems to be picking up something that FliM doesn't recognise.

Does the trace line offer any clues or has anyone experienced problem, or have any ideas?
Last edited by BrianW on 25.03.2014, 19:44, edited 1 time in total.
BrianW
 

Re: Warning messages on trace lines

Postby rjversluis » 23.03.2014, 22:06

I have only CAN-GCA Modules and do not have these problems.
You have to provide Rocrail server traces with BYTE level activated in wich such a lost/0xFF appears.
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBus - CAN-GCA ] - [ 0: RocNetNode - GCA-Pi ]
rjversluis
Site Admin
 

Re: Warning messages on trace lines

Postby BrianW » 24.03.2014, 14:51

Hi Rob, thanks for quick response. 2 trace files and xml attached.

1st trace file rocrail.009
Points LTP1 to thrown by mouse click on Rocview diagram. Rocview server lists no start byte [0xFF] twice
Points LTP1 to straight by mouse click on Rocview diagram. Rocview server lists no start byte [0xFF] twice
Points LTP2 to thrown by actuating reed switch. Rocview server lists no start byte [0xFF] once
Points LTP2 to straight by actuating reed switch. Rocview server lists no start byte [0xFF] once
(the number of listings of ‘no start byte’ varies each time the sequence is run, minimum once, max so far six)

2nd trace file rocrail.001
Train started in auto from Lower platform 2. Rocrail sets route but train actually operates LTP2 via reed switches as it approaches LTP2 and as it clears LTP2 (so the ‘no start byte [0xFF] timings will not coincide with Rocrail outputs).
Occupancy detector LTS16 does not reset as train leaves section. When tested without actuating points it sets and resets perfectly every time.

xml file for reference. If you need any more information please let me know

Regards,
Brian
You do not have the required permissions to view the files attached to this post.
BrianW
 

Re: Warning messages on trace lines

Postby LDG » 24.03.2014, 15:11

Hi Brian,

is LTlayout.xml really your current plan file ? If yes: The revision you are using (4338) is dated November 2012. Update as soon as possible to a recent Rocrail revision and try again :wink:

Regards,
Lothar
LDG
Site Admin
 

Re: Warning messages on trace lines

Postby rjversluis » 24.03.2014, 15:12

Hi Brian,

I cannot fixed that because the interface did receive some binary junk.
If you are using USB make sure that no transformer or other electromagnetic field is nearby the USB cable.
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBus - CAN-GCA ] - [ 0: RocNetNode - GCA-Pi ]
rjversluis
Site Admin
 

Re: Warning messages on trace lines

Postby rjversluis » 24.03.2014, 15:14

Yes! You must update first before you may complain again. ;)
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBus - CAN-GCA ] - [ 0: RocNetNode - GCA-Pi ]
rjversluis
Site Admin
 

Re: Warning messages on trace lines

Postby BrianW » 24.03.2014, 18:45

Hi Rob,

Thanks for the reply.
The 'binary junk' you refer to confirms my suspicion that it is probably 'noise' from the servos.
Will investigate this further. I wanted to make sure so far as possible that it was unlikely to be a simple Rocrail problem that I had missed before I started investigating in detail elsewhere.

Surprised how out of date my Rocrail is - it had been working very well until recent problem detected so didn't want to risk problems by changing it.
Will now update to latest version.

My comments were not intended as a 'complaint'. I'm very pleased with Rocrail and the help that's available via the forum.

Many thanks

Brian
BrianW
 

Re: [Solved] Warning messages on trace lines

Postby BrianW » 25.03.2014, 19:48

Hi Rob,

The 'binary junk' that caused my problem was eventually traced to a slightly loose 15 pin D plug!
After unplugging boards 2 & 3 the fault was still present on board 1. I spent the afternoon checking board 1 cable connections – all secure, then found the plug between upper and lower sections of board 1 slightly loose – pushed it home, problem solved.

It's often said that the least reliable parts of any electrical or electronic system are the mechanical components, certainly proven in this case.

My thanks to Lothar and Rob for suggesting an update to latest version of Rocrail. I didn't realise how quick and easy this was. It didn't solve the problem, (even the latest version can't deal with 'binary junk'), but I'm now up to date and running again, that is until the next time I get stuck!

BrianW
BrianW
 


Return to mergCBUS

cron