Surge Reprogramming C-Bus

I have a job that is 2 years old and the job has been working fine. 3 weeks ago just before the client headed overseas 2 of the relays (which are on their own phase) inexplicably turned all 12 channels on. The rest of the job including several other relays and dimmers were unafected however they were not on the same phase as the 2 relays in question. When I have scanned the network I could communicate with the relays, change the programming however I could not turn any of the channels on or off from the software or the switches. On further inspection it appears that there has been a surge of some sorts and despite the fact that there is 2 stage surge protection on all phases the theory seems to be a surge. Now the explanation that has been floated to me by our CIS rep is that there has been a surge on that phase and that surge has altered the programming of the relays to the attached GUI. I find it difficult to comprehend that a surge can create a new group address and engage the logic as well as change the default from and to or as well. Has anyone got any ideas and can this indeed happen?

Sorry wrong attachment. Aplicable attachment.

 

G'day Adrian
Sad state of affairs about the CIS rep suggesting that C-Bus is capable of reprogramming itself...
But rather about your network, is it possible that the remote override facility (the green pair) is forcing them on? Have you checked the cabling (disconnect from relay) and see what happens. Dosen't explain the other relays not also on, but that depends on how it's all wired together.
Regards
Matthew

 

I had a similar thing happen to me. At complete random times, lights from one of my relays would just switch on. I couldn't even switch them off using the override on the unit. I also thought it might be a cable issue but here in SA we don't even wire the remote override.

I disconnected the cables from this unit and ran the rest of the network to see if it was something on the network that caused it. After many days and weeks of struggle I finally came to the conclusion that the relay must be faulty and started the RMA process. As I started to take the relay out I found a few connections not very thightly screwed in. As I moved the relay it started making a faint buzzing sound and the thing would go into this weird state, and yes, I saw groups assigned where they were not supposed to be. I decided to tighten the loose connections and checked to see if it still gave these weird random light issues. This was more than 3 months ago and it still runs fine.

 

Hi Matthew
Yes mate I have checked the green pair and have made a mini network with only one of the relays my pci and nothing else as well as changing over all leads and tightening all terminals. We actually changed the phase it was on just to see if that helped. I then changed the relay unit over to a new one to see if there was any difference.Unfortunately it seems as if whatever happened has happened previously and changing all of this may be after the horse has bolted. It is now really anoying me that it has somehow changed without any real explanation. I don't have enough hair to keep pulling out

 

Hi Ingo
I did actually tighten all terminals on the relay when I had a look at it and possibly that may have contributed to the initial problem. I will give it a go and see if anything changes. Thanks for the advice.

 

Hi guys

I have been meaning to post about this for a while but keep forgetting.

I have experienced the same thing on at least 5 different sites, always following a report on site of a "surge" causing one or more breakers to trip or an electrical storm and associated power outage.

In all cases, except one, a relay or dimmer channel had been "re-programmed" to Group 0, which we never use.

The most recent incident was slightly different. We were called in to look at a house that was nearing the end of construction but the owners were living there - over Christmas of course - and most of the C-Bus was up and running. The electrician responsible for the job had gone overseas after programming all the gear that had been installed.

The reported fault was that "we had some issues, a breaker tripped and after we reset it some lights won't respond to the switches, which used to work alright."

I scanned the network and found that one dimmer was set to "Application 175". It had groups in it and when I changed it back to the Lighting Application, normality was restored.

Anyway, after those long winded stories my point is that I belive power "surges" can cause C-Bus relays/dimmers to "be reprogrammed"...and I'm glad I'm not the only one to have experienced it!

Cheers

Matt

 

I concur. I have returned to a couple of jobs after similiar circumstances, ie storms, power outs were GA's had moved themselves around. There was one job in particular that was on the same floor as the A/C plant room which i suspect was the problem. It also changed primary GA's as well as logic GA's and logic and/or tick boxes. So its good to hear im not the only one. But we haven't heard anything from them for a while know knock on wood.

 

There are over a million C-Bus units installed all over the world, and up to the time I read these posts, I was satisfied that in any cases of 're-programming' occuring, that there was a satisfactory explanation for it. These complaints are few and far between, but you have me concerned.

C-Bus has been designed and tested to withstand all manner of 'surges'. All units comply with the European EMC Immunity requirements, and these standards require surge, transient amd electrostatic discharge testing. In none of the tests I have attended, did I ever see any 're-programming' occur, so I don't expect that a simple surge is the cause.

Intermittent connections due to loose terminal screws can generate high amplitude electrical noise very close to the C-Bus microprocessor. This could possibly be a culprit, but I would expect more general corruption of data, One or two random bytes at a time.

Learn mode in units, if activated, is likely to create a "group 00" where none existed before, as it attempts to assign group 0 if it is available. Learn mode is entered by holding down a key on an output unit for longer than 10 seconds, and it is possible that in conditions of high humidity or with geckos or vermin entering the enclosure (or a person leaning against a switchboard full of units), that this might be simulated without human contact. Learn mode times out on its own accord, but if something like a gecko wandered past it is just as likely to activate the same input a second time and thus complete the operation of re-programming a terminal to group 00. Learn mode can be disabled in units, and I suggest that if Learn is not being used, it should always be disabled in all units programmed by an integrator with toolkit.

If there are any units on a network which are running C-Gate or other software which regularly interrogates units in turn, there is an opportunity for data corruption to result in commands being interpreted by units incorrectly. This is a long shot, because all data packets incorporate checksums sufficient to detect multiple bit errors, so such corrupted packets would most likely be discarded by any units (resulting in re-transmission and eventually successful data tarnsmission).

A corrupted program counter in the C-bus microcontroller could in theory point to some code that may be able to orchestrate changes to EEPROM parameters (where the non-volatile programming is stored), but this possibility has been considered, and an extensive net of "watchdogs" has been implemented to reset the program counter to a correct value. You would have a better chance of winning the Lotto than having this occur, and from the comments, I think this can safely be ruled out.

Finally, it is possible that some corruption or bug might exist in a PC or other unit that has the capacity to issue programming commands to C-Bus. There is always also the possibility of unintentional re-programming by a (lay)person using such tools ('finger trouble').

I would like to know in all of these installations that people have experienced 're-programming' if the networks that the units were on at the time were simple C-Bus networks, or if they were connected to a PC or could have been interfered with by a programming tool such as Toolkit. It would be a relief to know that if there is a problem, that we can eventually work out a solution to further improve the overall reliability of the system.

 

Don,

I've had units reprogram themselves as well.

However I don't recall if it was after a storm.


Peter

 

I chip in as well and say that I think I have had one do it as well. I say I think it has because there was just a group number assigned to an output and I have never just used group numbers. It has only appeared recently as well (last two months). No idea exactly when and if it was storm related.

Mick

 

My network has HomeGate and Minder Pro connected. It's still like that without any problems. I think mine was caused by the electrical noise of a few loose connections. I must say it's fairly robust, we have very regular power failures and the system can handle all that without any issues.

Ingo

 

Hi Don
Sorry for the late reply. The network in question is a simple network that does have the first colour touch screen that was released. I did updated the card and the firmware at the time. We have had no issues with it's operation for the 2 year period after changing the card.