Theoretically, the pressure from the fault would cause one or more rupture discs to blow.

A few years ago I was commissioning an AIS ring bus with 345 kV SF6 breakers. All of our breakers were open. The remote end had energized the line and had soaked for about 5 minutes. I went outside to check the voltages at the CCVT & VT box for the line and we heard what sounded like a flag pole "clanking". My CCVT voltages disappeared. The CCVT/VT secondary boxes were about 12 feet from the failed breaker.

The center phase had flashed over phase to ground. Fault records from the remote end (about 1 mi away) indicted ~28kA, with ~ 3cycle clearing time (411L relays, 2 cycle breakers). Locally, some reported seeing the center phase of the line gallop. The voltage dip was so pronounced that a large automotive plant across the way tripped off line (fed from 138kV).

After verifying that my underwear was clean, I inspected the GCB. The rupture disc was intact, with no visible damage to the GCB. Compared to the other phases, the cement pad appeared darker, which we speculated was due to exterior moisture on the tank being thrown off from the forces of the fault. An SF6 sample was drawn with about 200 ppm SO2.

The OEM replaced the failed phase and took the failed unit to the factory for analysis. About eight months and 15 emails asking for the analysis, the OEM reported internal contamination as the root cause. These particular type of breakers are shipped from the factory with the bushings installed and a small positive SF6 pressure. FWIW, Doble tests were performed on each breaker as part of the commissioning tests and the results from this breaker were no different than the sister units.

In the least, your story is comforting lol. :)

Now, THAT's a good story. And well told. I stay away from those things. 690 V is enough for my body (and soul).

Gunnar Englund
www.gke.org
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

Being able to put your hand on an aluminium tube and being 6" from a quarter million volt busbar is unnerving.

I used to hate being in the breaker hall when our generators closed onto the system because if something was going to go wrong that was probably when it was going to happen. The day one of the shift managers manually closed a machine onto the bar about 40° out of phase scared the, ahem, poop out of me. That's nearly 3 o'clock on the synchroscope, after we'd talked about how we'd close just before twelve o'clock with the machine running slightly fast. The auto-synchroniser and checksync relays were both out, and we were closing up the old way. Not something many of us had done a manual sync 'for real', and this guy had never done it before:

"Close! Close!"
"When, now? Oh, OK"
"No, wait. No! NO!! F--k!"

Apparently the angry growl I heard in the breaker hall - and which seemed to lasted *forever* - was nowhere near as scary as being on the machine floor where the whole turbo-generator train rocked on its foundations. Some old hands came down white as ghosts after that.

I remember now. Summer job in a hydroelectric 3x40 MVA station. Midsummer. That is sacred in Sweden. And there I was "assistant control room assistent" - which meant reading water level and feeding pig's ears with white fat worms to the German Shepheard that was our defence against sabotage). So, being the youngest, I got a short training in how to synch the units before the guys went off with "gals and bottles". But they forgot to tell me about the slow motor mechanism (no air, no springs, just a motor and a gear that took the breaker above the trip point. It took several seconds from ON until the breaker closed. When the scope was at noon, I pressed the button and it wasn't until six P.M. that I heard the bang.

Turned out to be a much better time than 3 o'clock. That would have caused mechanical damage. But now, we had lots of current but not much forces.

Gunnar Englund
www.gke.org
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

In the event of  internal flash-over occurred within a section of GIS the following potential issues should be considered:
. a)High frequency fast transient overvoltage could be caused unexpected failure on the insulation of other equipment connected to it and the grounding system.
. b) Gas overpressure can damage the vessel, rupture disk or other internal components of the GIS.
. c) If the GIS is installed indoor, there is a risk that the overpressure damage the building

Adding special pressure relieve louvers, could mitigate the potential damages to building and personnel For additional information in this matter, see the excerpt below from typical installation in 15 GIS substations for a client in the US.

Awesome! Can you increase the pic in size, its blurry on my end.

Hi Mbrooke: Per your request, enclosed is a pdf version of the information sent earlier.

FYI, below is a quote and link on an incident report of a 66 kV GIS failure triggered by a breaker pressure relief valve.

Quote (Damage: The pressure from the release of the safety relief system was sufficient to blow open the switchroom door. First responders on arrival at the switchroom noted the doors open and smoke emanating from the building.)

https://dpti.sa.gov.au/__data/assets/pdf_file/0009...

• https://files.engineering.com/getfile.aspx?folder=a923f2c6-56c6-4b3b-83e8-a

Thank you- again. :)


Can such a blast injure someone from over pressurization of the room? Even with vents?

Although the blast effect of sudden release overpressure has been some incident it is a rare event, however, there is potential for injury. Should be noted that there is no known injuries incident due to blast wave effect.
Injuries are a function of the levels of protection stand-off distance and predicting blast effects. Specific blast design concerns and mitigation options should be evaluated on a case by case basis for GIS project.
Will be prudent to take certain precautions and evaluate the probability of this type of fault and the consequences associated with this event.

Like tripping on low-low lol...


Hope it never happens outside of that.

Have seen burn-through of bus duct, along with deposition of a highly corrosive white powder, presumably hydrogen fluoride crystals; off-potential-only haz-mat clean-up of switchgear externals without violating limits of approach was the first step so that isolation and establishment of work protection could be undertaken, following which necessary dismantling and final clean-up could be undertaken. Once chemical hazards were eliminated, the usual assessment, ordering / acquisition of parts leading to repair / replacement took place, along with a more detailed fault analysis.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

Was this SF6? (I'm thinking something like GITL, so forgive me if you had classical bus duct in mind) What was the cause of failure?

Quote (Free bouncing (dancing) particle in GIS and GITL one of the key factors which affect their reliability. Free metallic particles can adhere to the insulating spacer surface of the GIS, thus creating serious defects. As a consequence of such defects, partial discharges (PDs) can commence in the SF 6 insulation. Such PDs are very dangerous to the insulation integrity of the GIS. ...)

Yes, classic SF6 switchgear; free dancing metallic particle contamination leading to PD as per cuky's description.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

How do hydrogen fluoride crystals form?

I'm not a chemist, but I've been told it's something to do with very high levels of energy dissociating the SF6 just as it's vented to the atmosphere. Reportedly the fluorine has a high affinity for the hydrogen found in water vapour, which it will dissociate to get the H needed to yield HF and free O2.

Some of the energetic sulphur may also combine with hydrogen to yield H2S; maintenance people attending to faulted SF6 switchgear have reported both the white powder and the smell of rotten eggs.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

I've heard that the byproducts or SF6 after it extinguishes an arc are toxic, but don't know to that degree.


How toxic are the crystals btw?

I don't know the answer to that.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

SF6 is inert during normal use and It has been identified as one of the greenhouse gases and in small quantities causes the voice to become significantly deeper, due to its density being so much higher than air.
Breathing SF6 is not recommended ==>> https://www.youtube.com/watch?v=u19QfJWI1oQ

Quote (HUMAN EXPOSURE AND TOXICITY: SF6 is a simple asphyxiant. it is chemical inertness and it is very low accumulation potential support the low concern for the toxicity of this substance.)

However, when electrical discharges occur within SF6-filled equipment, toxic byproducts can be produced that pose a threat to the health of workers who come into contact with them. (see below).

Quote (Degeneration products can be toxic, causing nausea and vomiting, pulmonary symptoms, and transient atelectasis. It may be contaminated with other fluorides of sulfur, such as sulfur pentafluoride and disulfur decafluoride, which are extremely toxic and are respiratory irritants. Inhaled as a mixture of 80% with 20% oxygen, it produces tingling, excitement, and altered hearing and is a mild anesthetic.)

For indoor GIS, it is recommended to properly design the HVAC system to remove the concentration of SF6 close to the floor or the building basement. Below is an illustration of the effect of the toxicity of the SF6 courtesy of
Dilo.

@crshears: its ok, neither do I lol.

@cuky2000- wow! Leaned some chemistry today.


Is the risk the same (of any toxins) of pure air or CO2 gear?


https://www.siemens.com/content/dam/internet/sieme...

The HF is one of the big problems, because the moisture to produce it is present in normal air: it forms one of the nastier acids. GIS facilities usually have showers provided to enable thorough rinsing after any accidental contamination.

Makes me wonder if its worth considering alternatives to SF6, which sadly seem limited.

Like this ? https://www.3m.com/3M/en_US/novec-us/applications/...

I learned of it here @ eng-tips . . .

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

Yup- like that. As long as it won't produce toxins when subjected to breakdown arcing.

Hi Mbrooke,

None of the dielectric gases produce toxins. However, all of them are toxic under arcing. The toxicity of the SF6 is larger than the alternative gases mentioned above.

Vacuum is not poison and is widely used in MV but with limited applications up to 69 kV (possible to 138 kV).

Quote (A toxin is a poison produced by a living organism. Venom is a toxin injected from a living organism into another. A venom, therefore, is a toxin and a toxin is a poison, not all poisons are toxins, not all toxins are venoms. In short, toxins and toxicants are narrower definitions for the word “poison.”)

I know SF6 is safe, but doesn't it produce toxic byproducts when subjected to arcing?

Yes, SF6 and other alternate gases as well produce toxic byproducts when subjected to arcing

Thank you, you and others have given me plenty to think about. I will research the toxins and breakdown properties of SF6.


Are there any resources of published or documented GIS failures?

One good source is the CIGRE Report on the latest international survey on high voltage gas insulated substations (GIS) service experience. Below is a glance of the main failure causes in GIS collected from various publications.

When a GIS fails, is everything typically contained or does it burst the rupture disk with lots of smoke/discharge?