Bus Bar Air Space Separation & DP Redundancy
Introduction:?It’s Wednesday evening and 35 minutes before I should post an article, it’s time to desperately cannibalize weekly work.?This week, we will look at main switchboard air space separation.
Bus Bar Boom:?If you have an electrical background and you know that bus A and bus B of the main switchboard cannot be lost together, it is often disappointing to look in the back of many low voltage switchboards and see the bus bar arrangement.?They are often sharing the same air space and just begging for an arc fault to ionize the air, turn into a flashover, and blackout both buses.?Nice, well-meaning engineers will tell you that the bus bar distances are in accordance with standards and the arcs are like fires and fires are outside the scope of DP2.?But aren’t shorts inside DP2 criteria??Don’t closed bus ties need to reliably open to save one side of the power system according to the rules??How can that possibly work when the bus bar air space has become a metal box full of plasma??The plasma is in parallel with the bus tie breaker and the circuit won’t break if the breaker trips.?Cue experienced engineers with stories of splattered copper.
Hmmm:?There is plainly a problem here.?We aren’t allowed to lose both buses, yet we have a valid DP2 failure mode that can cause this.?There are lots of low voltage DP2 switchboards built like this and some rules allow it.?Shouldn’t all these air spaces be isolated from each other to ensure redundancy??Yes.?Yes, they should.?Thank you for your time.?You may go.
Reasonableness:?Less enjoyably and more seriously, it is good practice to separate the air spaces containing different redundancy group main bus bars.?This is required by some class society main class electrical rules, found in most medium voltage switchboards, and is good engineering practice.?It isn’t required by everyone, because there is an assumption that rational actors will regularly shutdown their switchboards and clean the bus compartment to prevent the build-up of metallic dust (burning/grinding), salt (it’s a ship), and tracking.?If this bus bar maintenance and the associated breaker maintenance is performed well enough and often enough, then the risk of arcs becoming flashovers is probably below DP2 criteria.
领英推荐
We So Smart!?They obviously don’t know us well.?Maintenance is sometimes a dirty word in the offshore, and people will throw away buckets of money to avoid doing it.?Other industries might make money off of their maintenance practices, but we resent the time and money spent on the investment.?This isn’t always irrational, as owners need to make money, while the sun is shining on an often cyclical industry.?The assumption of maintenance is less rational than expected, and without it, we are back to flashovers.
ROI:?We don’t like spending money or losing time, but we also hate losing equipment, people, and time.?We can’t always pretend that these events were acts of god, we knew the odds, or the fault of the people that we wouldn’t let do maintenance.?If we don’t like doing maintenance, then let’s use a design that is less dependent on maintenance.?This makes a cheap investment of proper air space separation a good DP2 redundancy investment.?This is best done when switchboards are specified, but can be retrofitted with sturdy barriers (like the Bakelite below).
At Last, A Diagram:?The left picture shows a bus tie breaker with the Port redundancy group bus bar and air space separated from the Stbd redundancy group bus bar and air space by a sturdy non-conductive barrier.?The two middle pictures show the advantage of this.?A major fault on one side is not a major fault on the other side.?(Yes, I can complicate this, but let’s keep it simple.)?The air spaces reflect the redundancy groups, so an arc fault is limited to one redundancy group.?Some power is lost, but there is no total blackout.?On the right, we have an arc in a common air space becoming a common fault.?They don’t always become common faults, but they can.?Separate air spaces are a worthwhile investment.
Application:?I’d like to say that this is only a problem with some class societies, but it isn’t.?I’ve been the class surveyor stuck with an unseparated approved switchboard, because the people who approved it didn’t know the rules.?I’d like to say that this is only a low voltage problem, but it isn’t.?I’ve seen medium voltage switchboards built like that and you would not believe the stuff I used to do for a living.?There are lots of old vessels where this is a problem.?I’d like to say that DP3 doesn’t need to worry about this, because of their explicit requirement for separation.?Does anyone remember the bad DP design contest??One entry had a DP3 ship with one switchboard.?Then there is the little considered field equipment, which is more likely to be dual fed.?Automatic changeover switches can cause a similar problem if connected to main power.?Won’t the supply breakers solve the problem??When were they last calibrated, tested, and maintained??As we drop in voltage and power, the problem diminishes, so 24Vdc needs a lot of juice to be considered, while 480V main bus bars are usually a valid concern.
Conclusion:?Bus bar air space separation should be seriously considered on DP2 vessels.?It is required on some DP2 and all DP3 vessels.?It is good practice where arc risks are significant and redundancy is required.?It reduces risk and the dependency of the system on maintenance by making flashover blackouts far less likely (I’m an engineer - nothing is ever perfect).?Vessels without this need to take a serious look at their maintenance to ensure risk mitigation.?It should be a no-brainer in new-build specifications and in upgrade wish lists.?What does your switchboard have?
--
8 个月Si
Engineering Management Professional | Experienced, Practical, Registered Professional Engineer | Dynamic Positioning Subject Matter Expert (DP SME)
1 年My apologies. I was unfortunately too busy to properly reply to people who commented, and people made good comments that I wanted to reply too. Open discussion is vital. Russell is right that a capacitor or power supply that pops can take everything in the same box with it. You don't have to put redundant systems in a separate box, but why would you want to depend on luck? It's good practice vs. economics & space. Good practice usually wins, but there are strange counter examples. Manual is right that arcs like to go to ground, but electricity is creative, copper splatter travels in straight lines, dust/objects fall down, & ionized air just expands. If they connect to something interesting then the problem expands. Arcs are normally limited but we can all think of examples that weren't. Why let them? Russell and Manual are right that cleaning is vital. I'm a little concerned about hearing talk of vacuums & no talk about using liquid cleaners. Many of the cleaners that I used to use are now forbidden, but there are modern equivalents. Insulators need kept clean & brushes can't get some places. Vacuum then clean. We had a 1908 Edison Electric open bus switchboard in my first place of employment. Generations of men kept it clean.
A career path diverted by essential roadworks
1 年This is not an issue which applies only to switchboards. It can also apply to redundant control equipment. More than once I have pointed out that two computer systems in the same cabinet are subject to simultaneous failure if, say, a power component burns out and starts a fire in the cabinet. See HSE RP 195 for a particularly infamous incident (if you have a memory as old as mine). On each occasion the response has always been: compartmental failure only applies to DP3. If cable routing can apply to DP2, surely the classification societies can extend co-location in cabinets. At least one DP computer manufacturer has addressed that issue. As to cleaning switchboards and cabinets - give me a break! Men don't get out a vacuum cleaner and clean the dust (and even welding slag and bits of cable left over from the build ten years ago) - That kind of thing is Women's work. Besides, its not in the PM schedule anyway. I've seen people complain about the unreliability of controllers then opened up the cabinet to find layers of dust over the control cards. What do you expect? Housekeeping is an important part of maintenance (remember fan filter cleaning?) - unfortunately too many consider it to be SEP (Douglas Adams: 1952-2001).
Fleet Management Specialist at Bureau Veritas Marine & Offshore, but my views are my own.
1 年Comment 1 of 2: Although I agree with you in broad terms, something needs to be said. Most LV switchboards are separated into a port side and starboard side. In between the two halves, sitting inevitably in one room for DP2 vessel, there is a buscoupler panel. This basically separates the two sides. It is correct that there still mostly is an open channel for an arc to travel through in the back of a board, but arcs usually find the path of least resistance to earth (outer cladding of the switchboard) more preferable. Two things are more a danger, and both involve owners. Firstly, keeping a switchboard clean is not that hard. Just switch it off, make sure it is dead, take your industrial hoover out and your dust cloth and within 30 minutes or so that thing is spotless. People tend to make this seem harder and more complicated than it really is. Train your people to shut down an electrical powerplant safely and you are good to go. There is always an opportunity during a port call to do this.
Marine Manager & Technical Superintendent at James Fisher Subtech
1 年Hi Kerr, some switch boards do have 2 breakers as bus tie, this is slightly better. However bus bar air space seperation is best