Vacuum CB's potentially dangerous? 1

[Bill99]

SF6 breakers can monitor and trip on low gas pressure.
What if a vacuum CB was opened under load with no vacuum in bottles? Can this be monitored similar to SF6 bottles?

 

[dpc]

You should read manufacturer's literature on this, since this question has been asked since the devices were introduced.

There is no method of verifying integrity of the vacuum bottles while in service, or at least no economical method that has ever been employed.

A compromised vacuum bottle can fail if opened under load or fault conditions, but generally they do not fail catastrophically. I'm sure someone will report their war stories shortly to contradict this. Stay tuned.

There are only a few manufacturers of the actual bottles and reliability vacuum seal has been extremely good. The weakest component is the metal bellows.

These are very simple, rugged, reliable devices. For the voltage range covered by vacuum devices, they are always my preferred choice over SF6.

 

[jraef]

Typically if a vacuum bottle were to lose vacuum, opening under load would essentailly mean it would not open; the gap is so small (typically only around 3mm) that current would arc over. Catastrophic failure as a result of that is totally dependant upon the conditions at the time the breaker was trying to open.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[davidbeach]

That's what breaker failure protection is for. You do include breaker failure in all of your relaying schemes, don't you?

 

[Electic]

Regarding Breaker failure: we've just received a proposal from major switchgear manufacturer that does NOT include breaker failure function.

Upon questioning they say
1)the old gear didn't have it
2)a main breaker stands as second line of defence (but is not wired to BF of the feeder breakers.

I don't like their reasoning, but are there occasions where Breaker Failure is not a requirement?

 

[davidbeach]

Traditional selectively coordinated time overcurrent provides back up protection in case of breaker failure, but that can add considerable time. Dedicated breaker failure relaying will trip the backup breakers much sooner; a 10 cycle delay from the original trip time is typical.

I think serious consideration of arc-flash hazards will greatly increase the use of dedicated breaker failure protection. If you don't include breaker failure time in your arc flash calculations you are assuming that the only thing going wrong is the arcing fault. Very seldom does only one thing go wrong at a time. Something to think about.

Now, with breaker failure available essentially for free with numeric relays, why wouldn't you? Also, breaker failure protection that trips a bus lock-out relay will cause more investigation than two apparently tripped breakers. What you really don't want is that failed vacuum breaker being closed again if it managed to stay together while waiting for the backup protection to operate. If you don't test the downstream breaker anytime two trip for the same event you don't know if it is safe to close the one that should have tripped alone.

 

[jraef]

Very seldom does only one thing go wrong at a time.

One of the most under represented tenets of good engineering practice. I never ceased to be amazed when people just reset a breaker or replace a fuse without investigating not only the cause, but the colateral damage that may have ensued.

I once worked on a 2300V pump station where they had a 1600A main breaker trip on OV from the protection relay. The operator thought "Huh, wonder what caused that? Oh well, better get back on line..." and reclosed the breaker. Turns out every MOV in the downstream gear had vaporized, so the ionized gas left behind allowed flashovers to occur in the motor starters, which then caused a 3 week shutdown as they were cleaned and repaired.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[Electic]

Thank You DavidBeach

You have provided some great scenarios.

RE: Now, with breaker failure available essentially for free with numeric relays, why wouldn't you?

This has been a six month sparring match with this manufacturer, and what I consider 'base' has needed constant negotiation. The supplied equipment has been good but the contracted 'design' has been very lacking.

 

[dpc]

I agree that breaker failure protection is a good idea, and that it will become more prevalent in the future due to arc-flash concerns.

However, at present, breaker failure schemes in metalclad switchgear in industrial facilities is extremely rare in the US. It is certainly not anything close to a standard practice.

If you don't specify it or design it in, I would not expect any supplier to provide it on their own initiative.

 

[wolf39]

Vacuum breakers are usually installed in 3phase systems. If two breaker phases open properly the third phase goes out without problems under normal load conditions regardless of the defunct vacuum bottle.

 

[swn1]

We never see BF in industrial systems. We'd like to, but the systems are usually pure radial and the breakers are more reliable than the never-exercised transfer trip circuits would be in practice.

The single-failed-bottle scenario won't clear a ground fault on the bad phase. That means that if you clear a three-phase fault by opening a vacuum breaker with one bad bottle you will probably be left with a SLG fault burning.

We had a customer lose half their generator building to a fire when one phase of an oil circuit breaker welded. Single-phased the generator for long enough to blow up the neutral grounding transformer, spraying burning oil all over one end of the building.

Haven't found a bad bottle yet. Still testing

 

[JBinCA]

1. I completely agree with DPC. I do field engineering in the US and have NEVER come across a metalclad line-up (or any other for that mater) that used BF.

2. I also agree with Davidbeach that many numeric relays come with the algorithm included standard. Even so, I've never seen it incorporated in the protection scheme.

3. I either don't understand what wolf39 is saying or I disagree with him. The contact separation in Vacuum circuit breakers is on the order of 1/2 inch (Jraef, are you sure your 3mm clearance isn't "wipe"?). If that BKR is tripped open with a leaking bottle (no vacuum means full of air) I expect the arc from the fault to be maintained until cleared upstream.

4. Combo time! Failed vacuum bottles are indeed rare. But so are operators / electricians that wouldn't have reclosed Jraef's 1600A main BKR. I think I find more failed vacuuum bottles (2 in five years of testing) than I do properly cautious operators / electricians.