Clamp on CT bizarre occurrance. 3

[jraef]

Had a doozy of a wierd occurrance today. I can't see how I had anything to do with it, but it was eerily coincidental if I didn't.

I was testing resistances on some SCRs with my digital VOM, which has a spring jaw clamp on ammeter built-in to it. To keep it at eyesight level while I worked, I opened the jaw and placed it around a piece of seal-tite flex conduit containing the line leads feeding 480V into the unit I was working on. HOWEVER, the upstream circuit breaker was locked off, and I had already done a voltage tick-trace to ensure that all circuits were dead, so the conductors inside that conduit had no potential on them whatsoever. There were no control wires or any other conductors inside that conduit.

The very instant that my clamp-on got around the flex and the open jaw made contact with a conduit strap holding it to the wall, the adjacent (1 foot away) Building Automation Control System panel went completely dead, shutting down ALL of the HVAC loads in a 37 story high rise office tower! Needless to say, everyone turned and looked at me as the culprit! No matter how much I explained that there is no way in hell that an open jaw of a clamp-on ammeter could have any effect on a dead power line inside of a flex steel conduit, they would not believe me. They want me to supply my insurance carrier's contact information to recover damages (I don't even know what damages would have resulted from having no HVAC for 30 minutes).

When they went to restart, the computer kept saying that it had lost communications with the control panel, even though all of the individual CPUs were functioning. The comm line to the central PC was a CAT-5 hard wired system, Ethernet I think, and they claim there were no power line modem communications. It took 30 minutes to accomplish a cold reboot of the system and get it to re-establish comms and get back on-line.

I am convinced that there is nothing that I did that could have had that effect, that it was just a "Murphy's Law" coincidence. Eerie, but coincidental nonetheless. Any opinions?

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P³[/blue]

 

[itsmoked]

I agree. That clamp-on can't affect anything inside a steel conduit. Bad luck there.. sheesh.

 

[ScottyUK]

Any capacitances in or around the equipment you were working on that could have held charge?

Was there an RCD (GFCI) or RCBO in the panel which tripped? If the load already has a high leakage to earth it can be easy to tip them over the edge through the most innocuous actions. If that was the case you could probably argue that the system was vulnerable to start with.


----------------------------------

One day my ship will come in.
But with my luck, I'll be at the airport!

 

[Barry1961]

If what you did had anything to do with what happened the system is very unstable. The only thing I can think of would be the ethernet is screwy. There are a lot of people out there trying to do industrial ethernet with stuff from Office Depot instead of something like N-tron.

If their ethernet is that bad or if it was just Murphy playing tricks, I would try to repeat it during off time hours. If it does repeat you can make some money fixing their problem.

Barry1961

 

[VEBill]

"...there were no power line modem communications..."

Hmmm...

This barely makes sense, but since the entire event barely makes sense, this qualifies as being worth mention:

Maybe the clamp-on jaw acted as a common mode ferrite choke and blocked some sort of DSL-like 'power line modem communications' (probably in the MF band, around 1 MHz)?

Perhaps the power line comms normally go through the opened breakers, and when opened the power line comms was able to couple onto the conduit somehow...

Even if this is what happened, you should not have to bear any legal responsibility because there is obviously something seriously wrong with the design or condition of the system if clamping a conduit would block the communications. No way such a system would be considered as being normal. Even if that is the way the system needs to work (for whatever crazy reasons), then it should be plastered with warning lables and perhaps the critical conduit should be placed inside, hmmm, another conduit (?) for protection.

The above doesn't make much sense - but the mention of 'power line' comms, combined with the mental image of a amp-clamp (looks like a ferrite clamp-on choke) made this a theory worth mentioning.

 

[jraef]

The power to the control panel was not interrupted by a GFCI or anything that I observed, just everything apparently quit communicating all at once, causing the total shutdown.

I'm the only one who asked about power line modem comms, because that was the only possible remote connection I could draw to anything I was near. But now that you mention it, I wonder if I could have caused some sort of common mode spike up through the ground plane of the control system's comm ports? But how the heck could an open jaw ferrite core cause that from a dead set of cables?

No big caps anywhere that I could see, just the small caps in the dv/dt RC network of the soft starter I was working on and maybe some miniscule residual line capacitance on the 25 feet of 4ga cable. Hmmm, I wonder if by having the jaw open when it touched the conduit clamp, it punched a spike from that little bit of line capacitance onto the ground plane? I was moving the jaw at the time, which I suppose could have been akin to passing a magnet through a magnetic field, IF one existed.

You all make a good point though, if it was that easy to affect it, it was a vulnerable design to start with.

I responded this morning with a letter explaining my actions, but suggested they get a PE with experience in power quality and communications systems to investigate their system. If he can describe to me how this could possibly happen, I'll get my insurance carrier involved. It will be interesting to see if they find anyone who will take it seriously.

Thanks for the comments.

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P³[/blue]

 

[stevenal]

Communication ground loops through the conduit? Fiber might be a good idea around high current stuff.

 

[VEBill]

I'd interpreted the phrase "...and they claim there were no power line modem communications" to mean that they claimed that they had LOST the power line comms (not that there was no such thing to begin with...). Now I can see the intended meaning. Sorry.

 

[Barry1961]

Sounds like they are trying to defraud your insurance company to cover their negligence on the operation and maintenance of their communication system.

Barry1961

 

[jraef]

Well, I wouldn't characterize it as fraud just yet, but certainly the potential is there. I just think that the bldg. engineer was immediately looking to cover his butt with the owner when they hear from all the tenants about lost work, revenue etc. This was in San Francisco, where every time the utility power fails, the lawyers go out and order new Mercedes sports cars in gleeful anticipation.

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P³[/blue]

 

[Skogsgurra]

Had several power line communication problems lately. One is where a VFD input filter seemed to shunt the 86 kHz carrier and made remote metering imossible. Hope to hear more about it soon ;-)

Gunnar Englund

http://www.gke.org