Ask me about industrial electricity 2: electric boogaloo

The Something Awful Forums > Discussion > Ask/Tell > Ask me about industrial electricity 2: electric boogaloo

«‹›4 »

grover: Jan 23, 2002; PEW PEW PEW

Anti-Hero posted:

Three-Phase, what kind of hoops are you going to have to jump through to get a PE with a BS-EET?

For VA, BS-EE requires 4 years experience, and BS-EET requires 6. Which isn't all that big a deal, all things considered. If it's a non-accredited BS-EET, 10 years experience is required. All of which still require applicants to pass both the FE and PE exams.

They'll waive the FE requirement if you've got a doctorate or 20 years experience and let you sit just for the PE, though.

grover fucked around with this message at 14:34 on Jul 12, 2013

# ¿ Jul 12, 2013 14:30

Adbot: ADBOT LOVES YOU

# ¿ May 17, 2024 02:05

grover: Jan 23, 2002; PEW PEW PEW

PE is extremely important in construction and a few others areas, but not really used at all in manufacturing (which undergo different testing/certification processes). A licensed engineer's stamp on technical drawings is required for most commercial construction work. Also, in a lot of states, you're not legally allowed to call yourself an "engineer" if you're not a licensed PE, and companies have to have a licensed PE on the staff to advertise engineering services.

Most civil engineering fields are in areas where PE is important, thus most CEs are PEs. The opposite is true for EE, where most work in fields where PE isn't required. I picked up my PE license a few years ago, but have yet to actually use it. It does seem to instantly garner a lot of professional respect, even if someone has never met you.

grover fucked around with this message at 01:31 on Jul 13, 2013

# ¿ Jul 13, 2013 01:28

grover: Jan 23, 2002; PEW PEW PEW

Here's one of the Argonne scientists loving around with the high-powered magnets in a Klystron during a test many years ago.

# ¿ Jul 14, 2013 14:23

grover: Jan 23, 2002; PEW PEW PEW

Some Guy From NY posted:

Ha, I have people do that all the time to me.

I will be infront of a high voltage test set performing a high-pot, with the cable going to the G&T device with everything roped off with danger tape and signs...and guys will still go around/under the tape, step OVER the cable which is putting out 25KV to come ask me if "I'm busy".

and these are career substation workers.

I bolded the important bit. I think the problem is that caution tape applies to other people, laymen, not trained professionals who often set up their own caution tape. I widely ignore the tape, too.

Part of the problem is that we don't have clear distinctions between "hey, we're busy here, public keep out" and "DO NOT loving CROSS THIS LINE OR YOU WILL DIE" kinds of tape.

# ¿ Jul 19, 2013 14:30

grover: Jan 23, 2002; PEW PEW PEW

Three-Phase posted:

Yellow/black or yellow tape (caution) versus red/white, red/white flagged, or red tape (danger).

Can't say I've ever seen red tape used on a jobsite. Shame- makes a lot of sense.

# ¿ Jul 19, 2013 21:19

grover: Jan 23, 2002; PEW PEW PEW

Three-Phase posted:

It would need circuitry to detect the loss of the grid ("islanding" or loss-of-connectivity) and fault out. It's not simple to do because you can be operating in different conditions with the inverter synchronized to the power grid:

- Heavy load in the house, using power from both the alternative and grid source
- Light load in the house, sending power from alternative to grid
- Equal loading of the home and alternative source, no power from the grid

There's an article here that talks about detecting islanding conditions.

Grid-tie solar systems are legally required to cut out if there is a loss of line power for this very reason.

# ¿ Aug 1, 2013 00:36

grover: Jan 23, 2002; PEW PEW PEW

AmbassadorTaxicab posted:

How are brownouts managed, in the context of load-shedding? Ontario had one a number of years ago, and this was the reason given.

Not sure how they're managed at the "rolling blackout" level, but as a more basic protections level, there are undervoltage relays designed to open breakers if power level drop below a certain voltage to prevent damage that can occur at low voltages. (Motors burning up are the most common occurrence.) If enough breakers trip open on undervoltage, the load is often reduced enough that normal system voltage can be restored. This is what happened in Texas when they experienced a sudden loss of wind energy when the winds unexpectedly died down, and industrial sites all switched to emergency diesel generators.

Breakers with undervoltage relays are often designed to automatically reclose when proper voltage is restored, which could lead to a vicious cycle in a brownout. Power companies will very often cut deals with large customers that have emergency generators, where the customer is charged a lower power rate in exchange for agreeing to voluntarily take themselves off the grid during brownout conditions. Which is generally a win-win, as that sort of customer would be taking themselves voluntarily off the grid anyhow in brownout conditions to avoid damage or risk of power interruption.

e: This can occur even at the residential level. Anyone who's used a large power tool at the end of too many extension cords has noticed how easy it is to stall the tool; this is because voltage drop on the extension cord leads to very low voltages. Stalled motors are essentially short circuits, and can cause the breaker in the house to trip. In brownout conditions, this can lead to motors in air conditioning units and refrigerators tripping breakers, taking those loads off the grid.

grover fucked around with this message at 16:52 on Aug 9, 2013

# ¿ Aug 9, 2013 16:30

grover: Jan 23, 2002; PEW PEW PEW

TyroneGoldstein posted:

I use these types of devices for data infrastructure rather than power, but yeah...what were you expecting?

I still have an ancient yellow fluke and it works great to this day. Spend the extra coin.

I'm a big fan of Extech. Every Extech meter I've used has been every bit as good as a Fluke, but without the Fluke tax attached.

KaiserBen posted:

I was at a plant in Ohio where they had a florescent orange label on each cabinet detailing the incoming power sources (voltage, current capability, arc flash hazard, and a lockout point) on every cabinet. Struck me as a decent idea, but nobody else seems to want to implement it.

It's legally required now. Well, most of that at least. "Nobody else" really don't have much choice in the matter.

grover fucked around with this message at 17:31 on Sep 13, 2013

# ¿ Sep 13, 2013 17:28

grover: Jan 23, 2002; PEW PEW PEW

Three-Phase posted:

Does Extech have a good process calibrator like the Fluke 726? Something that can do milliamp loops down to microamp precision?

Not sure; I don't do any calibration.

Quick google search shows Extech has a pretty good range of calibrators; looks like their process calibrator also goes down to 1μA precision. For about 1/8th the price of a 726. http://www.extech.com/instruments/categories.asp?catid=3

If I was doing something where I absolutely needed accuracy to 1μA... I think I'd pay the extra 3 grand for the Fluke, though. Just for the CYA factor of having used a "fluke".

grover fucked around with this message at 00:42 on Sep 14, 2013

# ¿ Sep 14, 2013 00:39

grover: Jan 23, 2002; PEW PEW PEW

DaveSauce posted:

What I'd really love is something that can do clamp-on 4-20mA measurement, but I'm sure that's an expensive accessory or separate unit. I'd need to measure 4-20mA somehow at the least.

I've got an amprobe AC50A AC leakage clamp-meter I use for grounds and other low current stuff (the small clamp makes it nice for small circuits in tight panels, too). Works pretty good on AC; not sure how it would be on DC controls. It's also conveniently small, though it just feels cheaply made. Runs about $300, though; if you don't need to go down to 0.1mA, you can probably find something cheaper.

http://www.amprobe.com/amprobe/usen/Clamp-Meters/Specialty---Leakage,-Recording,-Analog/AC50A.htm?PID=73058

grover fucked around with this message at 14:24 on Sep 15, 2013

# ¿ Sep 15, 2013 14:21

grover: Jan 23, 2002; PEW PEW PEW

kastein posted:

Can't use a clamp meter on DC wiring because: loving electromagnetism, how does it work?

Not all clamp ammeters work on DC- the cheap ones are AC-only. My other two clamp meters work on both AC and DC; I use them all the time for DC. But that particular Amprobe has "AC" in the name and no DC listing on the datasheet, which probably means it doesn't work at all for DC, which was all I was saying. I haven't actually tried to see what would happen.

grover fucked around with this message at 22:03 on Sep 17, 2013

# ¿ Sep 17, 2013 22:00

grover: Jan 23, 2002; PEW PEW PEW

SSH IT ZOMBIE posted:

What kind of education goes in to becoming an industrial electrician?
Is it guild\union based, or usually college?

This stuff is absolutely fascinating.

I was hoping someone would pipe up with actual specific knowledge about this, but a high school diploma is all you really need. Most of the industrial electricians I know have had some type of formal trade school (vo-tech or comparable) supplemented with a period of apprenticeship, either as a formal apprentice or as an electrician's helper. While you're doing this, work on your licensing requirements and study up for the journeyman's exam. Once you pass the journeyman exam and get licensed, you're officially (and legally) an electrician.

Technicians typically have some formal college education, and get to do a lot the more technically challenging jobs, like setting up or troubleshooting complex industrial power systems, controls, etc. The engineers who design those complex industrial power systems have at least a 4-year degree.

The engineer knows what to do; the electrician knows how to do it. The engineer might want a 1200A feeder cable; the drawing might specify a certain number, type and diameter of conduit and number & type of wire, and show how it's to be connected; or it may just draw a line and say "1200A". It's up to the electrician to figure out how to route that conduit, what fittings to use and where, how to support it, where to put the knock-outs for the equipment, how to pull the cable in, and so on. So the electrician has to be up on code as well as knowing the techniques for doing things like pulling in large industrial cables, and other incidentals like how to order cable, how to know how much to order, what the best cable lube is, and so on. Then the technician will come in and throw a hi-pot on it and bitch about how you did it wrong and exceeded the tensile limits of the cabling, and the engineer will look up and say you routed the conduit where the HVAC guy needs to run his duct and you have to move it and OBTW you used the wrong fittings... and the project manager makes you do it all again. Because that's the life of the industrial electrician

grover fucked around with this message at 00:42 on Sep 28, 2013

# ¿ Sep 28, 2013 00:26

grover: Jan 23, 2002; PEW PEW PEW

Liberty ships had some crazy scary switchgear. 120VDC, 20kW.

Click for bigger:

this ship (SS Jeremiah O'Brien) still goes to sea, and still uses this switchboard for main power

grover fucked around with this message at 11:30 on Oct 2, 2013

# ¿ Oct 2, 2013 03:05

grover: Jan 23, 2002; PEW PEW PEW

I think I might have found the problem...

arstechnica posted:

The Army Corps of Engineers is overseeing construction and promised to make sure the data center is "completely reliable" before allowing it to go online.

I've been thoroughly unimpressed with ACOE's MILCON power engineers. When you add in low bid contractors, you get problems up the rear end and people getting their hands burnt off.

# ¿ Oct 9, 2013 02:53

grover: Jan 23, 2002; PEW PEW PEW

Early industrial electrical poo poo is all so :black101:

It's also so, basic, just fundamental applied electrical theory without no covers or control boards to mask it. You can look at every piece in there and make a pretty educated guess at what it is, just because there's so little hidden. Amazing how we got from open air transfer switches to having to wear PPE just to reset a modern 20A circuit breaker.

# ¿ Oct 14, 2013 12:14

grover: Jan 23, 2002; PEW PEW PEW

Three-Phase posted:

"Jones is dead! He bumped against an exposed electrical switch."

Protection against the tragic dangers of the open knife switch! Which were banned by code, effective Jan 1st, 1921, apparently.

# ¿ Oct 15, 2013 01:44

grover: Jan 23, 2002; PEW PEW PEW

Pvt Dancer posted:

Do people in the US actually wear PPE for switching MCBs? I do arc flash assessments in Europe mostly and it's always a good laugh when I tell workers that this in their corporate safety policy. I try to hammer on doing risk assessments as well (requirement of NFPA 70E no less) to get better acceptance but the US safety people always recoil at the though initially. Really, what?

Real answer? Depends who's watching.

Nobody suits up to reset a 20A breaker, ever, though. Don't know why that poo poo's even on the books.

# ¿ Oct 15, 2013 10:56

grover: Jan 23, 2002; PEW PEW PEW

There are UPS for every application; you can get them in multiple megawatt arrays. If it's an important enough process or interruption would result in a lot of damage or financial loss, you can put an UPS in. Run-time depends on how large the batteries are, but it's easy to size batteries to give enough time for generators to start up and pick up the load. Or if the UPS would cost $3M and a power failure "only" ruins a few thousand dollars worth of parts each time, could be cheaper just to let it fail.

My favorite, though, are superconducting electromagnetic energy storage. Literally a coil of superconducting wire sitting in a tank of cryogenic helium that can dump out a SHITLOAD of power in a fraction of a second.

# ¿ Oct 16, 2013 22:42

grover: Jan 23, 2002; PEW PEW PEW

squeakygeek posted:

This is a really interesting idea--haven't heard of it before. I work in a research MRI lab and was just thinking about how much energy might be stored. Do you have any interesting references off hand?

The largest holds 20MWh, which is pretty damned big.

http://en.wikipedia.org/wiki/Superconducting_magnetic_energy_storage

# ¿ Oct 17, 2013 00:31

grover: Jan 23, 2002; PEW PEW PEW

M_Gargantua posted:

So transformers hum. Normal. Today I had a heater controller buzzing itself to the misery of everyone around. Got told not to replace the failing contactor. (Relay. Anyone else call them contactors?) wound up just hand cycling it a few dozen times and adjusting the mountings. Quieter now but still noticeable.

Any of you have any sort of tripwires for when stuff should be replaced? We've got a few 1500A molded case breakers that I'm starting to worry about too. The managements view is run it till it fails.

Yep, transformers hum. In other circles, a giant electromagnet attached to a resonator is called a speaker. When transformers are new, they're usually pretty quiet, but as they age, the insulation and lacquer can crack or loosen and allow it to vibrate- generally perfectly safe, just annoying. It's usually louder when the transformers are lightly loaded.

Contactors are a specific kind of relay; generally bigger than 5kW and mechanical. When you're talking about industrial electricity, circuits will often have a number of purely electronic "relays" looking for different fault conditions, and will actuate breakers or fused contactors if they detect one.

# ¿ Nov 3, 2013 00:17

grover: Jan 23, 2002; PEW PEW PEW

Michael Scott posted:

Hi, long shot here:

A few years ago I saw a video on YouTube where some guys, in Russia or somewhere, approach a full-size (I believe AM) radio transmitter mast outside. They take a branch/stick, bring it close to the tower and start to touch it. An electrical arc forms between the stick and the tower, and the arc gives off a sound that perfectly replicates the music and speech that is being broadcast from the radio station. I think it's due to some sort of short or fault in the tower itself, I don't think that normally happens?

Anyway, I lost that video and haven't been able to find it. If anyone knows what I'm talking about I'd like to find it again.

That's pretty much how an AM radio receiver works- AM is just analog audio imposed on a carrier signal. Since the carrier signal's frequency is inaudible, you'd hear the audio frequencies from it.

grover fucked around with this message at 02:50 on Nov 22, 2013

# ¿ Nov 22, 2013 02:48

grover: Jan 23, 2002; PEW PEW PEW

Three-Phase posted:

I was initially skeptical of this. However I thought about it and called my dad:

I'm actually more surprised the protective systems didn't shut down the transmitter within milliseconds of the arc occuring. Or that someone was allowed to get close enough to the antenna to make it arc. That poo poo wouldn't fly in the US. :ussr:

# ¿ Nov 22, 2013 02:59

grover: Jan 23, 2002; PEW PEW PEW

Crankit posted:

I've seen the video, it's kinda cool but incredibly stupid in my opinion. I don't know how you could have the transmitter shut down when the vegetation caused an arc though.

There's not supposed to be any arcs, anywhere, ever. Or any vegetation close enough to cause it. High power transmitters generally have protective circuits designed to detect arcing and momentarily shut the transmitter down because arcs = damage, and damage to the transmitter is really expensive.

http://www.americanradiohistory.com/Archive-Catalogs/Gates-Harris/Gates-Harris-Cat-No-100.pdf

On page 7, describing Harris' 100kW transmitter:

The "crowbar" they speak of is a mercury ignitron switch which, when fires, provides a dead-short between the antenna to ground, to take all that energy away from whatever the fault is. Newer transmitters replace the tubes with transistors, but the concept is still the same: arcs=bad.

grover fucked around with this message at 13:19 on Nov 22, 2013

# ¿ Nov 22, 2013 13:16

grover: Jan 23, 2002; PEW PEW PEW

Crankit posted:

What I meant was, how could you detect an arc at the antenna itself, also I don't understand how an arc(a small one like in the video) at the antenna would be more damage to the transmitter than a dead short.

Optical sensors detect arcing within the transmitter and tuning components (they're rather large at these wavelengths- as in, you can walk around inside them); the circuit would be looking for the electrical signature of arcing for open-air antenna faults. Not sure how they did it back in the day, but they have all other sorts of electronic wizardry nowadays that can sense very small arcs.

e: here's a photo from inside the old tuning house at WABC; with the energies and voltages (15kV) involved, you wouldn't go inside the tuning house while in operation, or you'd become the arc path.

Apparently, the magnetic fields in the tuning equipment were strong enough that the components vibrated and acted like speakers. Here's a short audio clip:
http://hawkins.pair.com/wabcnow/coiltalk.ra

grover fucked around with this message at 16:29 on Nov 22, 2013

# ¿ Nov 22, 2013 16:18

grover: Jan 23, 2002; PEW PEW PEW

kastein posted:

Worth noting is that the copper pipes in grover's pic are probably the conductors - skin effect uber alles!

Yeah, they quite literally use copper plumbing pipe for a lot of this stuff due to skin effect and economics. Which is likely what we're seeing in this photo.

There's other special cable where they wrap a sheath of conductors around a non-conductive core.

# ¿ Nov 22, 2013 18:51

grover: Jan 23, 2002; PEW PEW PEW

FrozenVent posted:

Large-ish radar sets have waveguides between the magnetron and the antenna - basically a copper pipe. Is this what we're looking at? (Although I'm not sure I can reconcile waveguides with omnidirectional antennas, but I just use the drat things, so )

In this case, it's not waveguide, that's literally copper pipe used as an uninsulated conductor. Waveguide at these frequencies would be hundreds of feet wide. I've seen "coaxial" used for long distance runs, but it's not recognizable as such, because it's pretty much a copper-lined tunnel (that you walk through) with a conductor suspended in the middle.

Radar sets work at much higher (1000x higher) frequencies.

grover fucked around with this message at 20:04 on Nov 22, 2013

# ¿ Nov 22, 2013 19:58

grover: Jan 23, 2002; PEW PEW PEW

12MW diesel/natural gas generator taking a 1500km land journey on a 216-wheel flatbed journey from the port to a power plant in Australia:

In case you were wondering how they did it, neat 5-minute video about the trip (and some other info about the engine):
http://www.enginelabs.com/news/true-story-behind-massive-engine-in-facebook-page-photo/

# ¿ Nov 29, 2013 18:24

grover: Jan 23, 2002; PEW PEW PEW

FrozenVent posted:

Maximum power: 108,920 hp at 102 rpm
Maximum torque: 5,608,312 lb/ft at 102rpm

Two stroke engines are big.

The Australian one I posted was a 4-stroke. They're big because they're low-RPM. Low RPM reduces wear and maintenance intervals (which is very important for high-powered engines that run continuously for ships, trains, power plants, etc.), but since horsepower=torque x RPM, they need a cubic fuckton of torque to put out that power. So everything is super-massive to be able to be able to handle such high forces.

The 12MW one I posted above runs at 500rpm. The one you posted is 80MW at 102rpm. A 12MW packaged generator unit that runs at 1800rpm would be probably half the size, but would require a major overhaul over 2 or so years. So, you see "high speed" engines like that more in places they don't run continuously, like construction equipment and emergency generators.

grover fucked around with this message at 19:00 on Nov 29, 2013

# ¿ Nov 29, 2013 18:55

grover: Jan 23, 2002; PEW PEW PEW

helno posted:

I deal with some really big motors. Primary heat transport pumps are 11,000 hp at 6.6kv with across the line starting and you have to start all 4 at the same time to prevent flow reversal in parts of the process.

took a second before I remembered you work at a power plant. That's an insane amount of energy! I take it the plant isn't black start capable?

# ¿ Dec 20, 2013 23:03

grover: Jan 23, 2002; PEW PEW PEW

DaveSauce posted:

I also never understood why everything has HP ratings, contactors, breakers, etc.

Aren't we beyond that by now? Code says everything must be HP rated, but it seems like there are instances where that's not sufficient to size things properly. I usually size things based on nameplate FLA, and even that's flawed since your inrush curve is different for different motors.

Seems to me that the HP or kW rating are missing gobs of important information on the electrical side.

Yeah, I always found a simple "HP" rating on a breaker or contactor to be less than useful for engineering purposes. OK, great, x HP, but xHP with what kind of motor? Not that kW rating would be any better. Best to calculate both FLA and locked-rotor and ensure the protection is sized for both. Under modern code, there needs to be a fault coordination and arc flash study anyway, so the actual currents and curves are going to be known; rules of thumb aren't really saving anything.

# ¿ Dec 21, 2013 17:20

grover: Jan 23, 2002; PEW PEW PEW

helno posted:

It can keep going without the grid but cant start. PHT pumps take more power than the standby generators can deliver. We need around 40 Mw to run up.

Very few power plants can darkstart. I worked at a hydro plant that could, they had a battery box with jumper cables you could hook directly to the brushgear and a mechanical governor.

How does the grid recover after a widespread blackout?

# ¿ Dec 22, 2013 03:36

grover: Jan 23, 2002; PEW PEW PEW

Guy Axlerod posted:

I think Tons of Cooling has BTU beat as far as archaic units go.

Blasphemy! Tons is not only the first, but clearly the best unit for measuring heating and cooling. That's why every virtually AC unit you buy in the US today is marketed in "tons".

Two Finger: it represents cooling power associated with freezing (or melting) a ton of ice per day. BTU was created so that 12000 BTU/hr = 1 ton.

grover fucked around with this message at 15:35 on Dec 24, 2013

# ¿ Dec 24, 2013 15:28

grover: Jan 23, 2002; PEW PEW PEW

Best layman's explanation I've seen:

# ¿ Jan 17, 2014 03:21

grover: Jan 23, 2002; PEW PEW PEW

Are there any generators at this site? If so, are they open or closed transition? If your napkin calculations are coming out that close to the MPCB rating, you may have to bite the bullet and call in a pro do to a full fault current analysis & coordination study. Honestly, arc flash analysis is code required now anyway.

# ¿ Jan 23, 2014 22:51

grover: Jan 23, 2002; PEW PEW PEW

atomicthumbs posted:

I love the lamps and synchroscope

What's that noise in the background when they try to do the transfer? Generator being loaded down, or something else?

The first noise you hear is the 60Hz hum of a big transformer. You can hear a generator starting up (typical diesel engine rumble) and then falter- the white lights on one of the switchgear panels on the left go out when the gen dies. The clunk-whir-whir-whir-whir-whir-click is a motor operated breaker tripping open (the initial clunk), and the whir/whisle is a motor recharging the closing spring, and then a softer click at the end as the spring is fully charged. You can also hear the tie switch close to bring the lights back after the test failed- watch the green light (safe/open) on the tie breaker cabinet turn to red (energized/closed).

If you listen very hard, you can just about make out the sounds of poo poo hitting cotton when the lights go out. (Incidently, there are batteries that keep the switchgear going- that's why the meters and lights on the switchgear are still lit and breakers able to recharge, even when the power was lost.)

This was the successful test of the same system: note that the tie switch closes before the generator breaker(s) open.
https://www.youtube.com/watch?v=Qo2PqtSwWtk

grover fucked around with this message at 15:56 on Feb 1, 2014

# ¿ Feb 1, 2014 15:42

grover: Jan 23, 2002; PEW PEW PEW

some texas redneck posted:

Lets talk about what it takes to avoid rolling blackouts.

I know ERCOT will ask industrial customers to drop loads in the early stages of an emergency.

How does this work? I can't quite picture someone at ERCOT picking up the phone to call every plant in the state saying "hey you guys, we're like, out of this electricity thing, can you shut everything off for a bit so grandma can cook her waffles? THANKS!". I also can't quite picture an entire facility suddenly being shut off with no warning; that seems like a fantastic way to destroy equipment (and cause injuries).

The only personal experience I have at home with power emergencies is rolling blackouts. And those don't get implemented until "oh poo poo, guys we're so out of electricity that the entire state grid is about to go into meltdown!". In a commercial setting, I've seen about 3/4 of the lighting get shut off on a college campus out of nowhere, along with a good chunk of the HVAC at the exact same moment (as a student).

How's this stuff work? It can't be all manual, and I doubt it's 100% automatic.

Some of the stuff is automatic. When the grid is overloaded, distribution voltage (and/or frequency) drop, causing a brownout. When distribution voltage drops below an acceptable tolerance, emergency diesels kick on and the building transfers to generator until the utility voltage comes back to within tolerance and stays for a while (usually between 15-60 minutes). This occurs whether an industrial site has an agreement with the power company or not- gives time for the power companies to fix their poo poo and get more gens online. This is, incidentally, exactly what occurred when the wind suddenly stopped in the Texas panhandle and caused Texas to lose the appx 4% of its power generation capacity it was getting from wind power and suffer a massive state-wide brownout; the only thing that stopped Texas from experiencing a massive and catastrophically power outage were expensive dirty emergency diesels kicking on to save the day.

Some sites will enter into agreements with the power company that they agree to run off generator during times of emergency in exchange for paying a reduced electrical utility rate the rest of the time; in these cases, it really is a matter of the power company calling these customers and asking them to come up on gen, because they've entered a contract and are legally obligated to. The customer walks to the gen controls, pushes a couple buttons, and they're on gen a few seconds later. [Win-win for most of these customers, because they'd be up on gen during the brownout or rolling blackout anyhow.] In some cases, where they're configured right, companies can come in parallel with the grid and sell energy back; otherwise, they just operate in island mode, taking their own load off the grid and easing issues. Residential ~~suckers~~ customers that opted for the poco remote thermostat will find their air conditioning remotely disabled, too.

grover fucked around with this message at 11:58 on Feb 5, 2014

# ¿ Feb 5, 2014 11:52

Adbot: ADBOT LOVES YOU

# ¿ May 17, 2024 02:05

grover: Jan 23, 2002; PEW PEW PEW

some texas redneck posted:

Further stupid questions.

Is any kind of sync required when switching between power sources? Or is the delay from whatever breakers/relays interrupting power to switch between gen/poco enough? (I know you need to synchronize if you're generating while still taking poco power, but...)

I know an emergency backup generator only kicks in after the power has failed, but I have no idea how even those switch back to poco beyond a few switches. I know the generators we had at the grocery store I worked at took about 15 seconds to start providing power to lights/registers after a poco failure, but it literally powered about 1/50 of the store (handful of lights + all registers + electric doors + a couple of outlets in each dept - no refrigeration).

This is probably what that grocery store's generator system looked like. Big loads (lights, air conditioning, break room) would be on the main distribution panel, and emergency loads (cash register, fire alarms, emergency lights, etc) fed from generator via the emergency distribution panel.

The automatic transfer switch (ATS) is the brains; it monitors utility power and makes sure it's in tolerance. If power fails, it waits a few seconds to make sure the power is REALLY out and not just a blink, then tells the generator to start. Usually a small emergency diesel will be on and up to speed within 1-2 seconds. The ATS monitors the generator voltage/frequency, and as soon as it's in tolerance, switches power over to it. (TVSS is transient voltage surge suppressor) The whole process can be as quick as 5 seconds total after power failed to bring the lights back on, but it depends how it's programmed. There is often an intentional delay of 10-15 seconds to allow time for motors to spin to a stop before power is restored.

The most inexpensive form of ATS is "open transition", or "make before break" which basically means the generator will disconnect before utility power is reconnected. This happens in the blink of an eye- typical transfer time is about 100ms. To do it successfully without blowing every running motor, though, the generator has to be sychronized with the utility power- in other words, the 60Hz peaks of both waveforms must align. The old-school method was to use a syncscope and manually pull the lever, but modern ATSs do this automatically. They don't usually control the generator at all, just watch and wait for them to drift fairly close in phase with each other.

More sophisticated ATSs will be "closed transition", or "make before break", which places the generator in-line with the utility power. In the simplest form, they synchronise as above, but make the 2nd connection before breaking the first, and will be in parallel for about 100ms. This allows the return to utility power to be absolutely seamless. This also allows a site to transfer to generator if an outage is likely (storm anticipation mode).

The most sophisticated ATSs are "soft transition", which bring the generator in full parallel with the utility to slowly transfer power back and forth by carefully regulating generator voltage and phase angle; this eliminates the dip in voltage often associated when generators have full load quickly dumped upon them like Three Phase described, and is less stressful for the attached equipment. It's also expensive, complicated, and require a lot of sophisticated protective relays and controls. It's more common on very large systems because the controls involved scale up well, but don't scale down well; it's easier to absorb a $300k controller into a $5M project than a $100k project. This is what you're seeing in that switchgear video with the generator failure.

grover fucked around with this message at 02:55 on Feb 6, 2014

# ¿ Feb 6, 2014 00:40

The Something Awful Forums > Discussion > Ask/Tell > Ask me about industrial electricity 2: electric boogaloo

«‹›4 »