19 million Amps

deblau writes "On July 27, scientists at the National Nuclear Security Administration's Nevada Test Site said they generated a current equal to about four times all the electrical current on Earth. During the few millionths of a second that it operated, the 650-ton Atlas pulsed-power generator discharged about 19 million amps of current through an aluminum cylindrical shell about the size of a tuna can. Official news release is available from the DOE (PDF)."

Pure nonsense

[Anonymous Coward]

they generated a current equal to about four times all the electrical power on Earth.

Sounds like apples and oranges:
units of current = Amps
units of power = Watts

The statement is pure nonsense.

What?

[callipygian-showsyst]

...Test Site said they generated a current equal to about four times all the electrical power on Earth.

...

During the few millionths of a second that it operated, the 650-ton Atlas pulsed-power generator discharged about 19 million amps

Um....unless things have changed in the 25+ years since I took a college physics class, we measure POWER in WATTS, and CURRENT in AMPS. So the number you quoted in AMPS that you claims is eqaual to four times the POWER in amps doesn't make any sense. Of course, that never stopped our /. Editors before!

Two points

[TildeMan]

1. Current != power. Power = I^2 R, or any equivalent formula.

2. They did this on Earth, so it was actually only 80% of the electrical power (or insert appropriate noun here, see point 1) on Earth. Assuming it was four times the normal power levels without this extra current.

Re:current == power?

[stinerman]

Sure.

I = V/R
If R->0, I->INF.

Its certainly possible.

Re:And this benefits us how?

[vontrotsky]

This is part of the nuclear stewardship program. The US has a few thousand nukes that need to be maintianed, but not tested due to treaty restrictions. Therefore, intricate computer simulations are used to run virtual weapons tests.

The "tuna can" in this experiments is being subjected to high stresses, and measuring its response lets the researchers validate their simulation's predictions. If the simulation predicts the behavior of the can, it's more likely to acurately describe a nuclear device.

Jeff

Comment removed

[account_deleted]

Comment removed based on user account deletion

Exploding apples with capacitors

[Cyclotron_Boy]

This is a fun project. I was able to get about 18kA repeatably through a variety of objects [niell.org] from a small cap bank using low inductance leads and vacuum triggered spark gap. Lots of people do fun projects like this at home in their garages

For example
Bert Hickman's coin shrinking [205.243.100.155]
Thaltech's capacitor experiments [thaltech.com]
Sam Barros's Power Labs page [powerlabs.org]
Bill Beaty's webpage [amasci.com]
and many others...

Re:Coherence ?

[Anonymous Coward]

In theory, you could push an infinite current through a perfect superconductor.

In practice, you can't--all real superconductors have a "critical current density"--drive the current above a certain threshold, and it ceases to be a superconductor. It's a "density" because the exact current at which a superconductor stops superconducting is proportional to the cross-sectional area of the wire, but you'd need a very large wire indeed to drive 19 Mega-amps through a superconductor.

1.21 Gigawatts

[MikeyToo]

Given: 19MA generated(That's ninteteen megaamps as opposed to ma which is milliamps for those of you who avoided engineering). 1210MW (Again that's megawatts, or 1.12GW for you Back to the Future types) Then using Ohm's Law (E=P/I) They needed to work at 63.68MV (mega again). I wonder how long it will take them to get all this equipment packed in a DeLorean.

Re:11?

[Silver Sloth]

For those who didn't see 'Spinal Tap' - Marshall once sold an amplifier where the knobs instead of being marked 0 - 10 were marked 0 - 11. Many musicians, not normally noted for their technical savvy, assumed that they were that little bit louder

Re:The other questions

[CosmeticLobotamy]

From a Z-machine article, which claims that its 290 trillion watt output is 80 times world production, world production is 3.625 trillion watts. Times 4 is 14.5 trillion watts. Divided by 19 million amps (wattage is voltage time amperage, right?) is about 760,000 volts. But we don't need that number, just the 14.5 terawatts. Which is 11,983 times 1.21 Gigawatts. I'm estimating that the weight of a DeLorean is about a ton? Which means this thing, with appropriate flux capacitance, of course, can send about 12,000 tons back in time 30 years, or one ton 36,000 years (of course assuming that energy required is linear and proportional to weight and not size. If it's dependent on vehicle contents, then the same numbers work if you have access to 12,000 Marty McFlys, which this author assumes that you do).

My math might be off by a digit or two, so if you're going to be sending an aircraft carrier back in time to invade 1975 France (I'm looking at you, Mr. President), you're using these numbers at your own risk.

Re:current == power?

[InvalidError]

From later in TFA: "During the few milionths of a second that it operates, Atlas generates electrical energy roughly four times the Earth's entire energy production."

This is almost technically right except for "Atlas generates"... Atlas is only a huge capacitor bank, it does not magically "generate" energy, it only stores existing energy.

Now, if worldwide production is something like 25GW and the pulse lasts 10us, we have 25GW * 4 * 10us = 1MJ, a balievable finite quantity.

Power Calculation

[superstick58]

If we consider the resistivity of Aluminum as 2.82x10^-8 Ohm-meters and the dimension [pccrafts.com] of a soup can is .2159m length by .0889m diameter, we can calculate the approximate resistance of the aluminum and therefore the power.

resistance = resistivity*length/area

It turns out that the resistance is near 1 ohm at .981 Ohms. This means that the power would be found with the following equation.

P = I^2*R

Therefore we can estimate the total power to be a huuuuuge amount, 354.14x10^12 Watts.

19 million Amperes is chicken feed

[Anonymous Coward]

My home electric stove+oven has 2x 50A circuit breakers; my electric water heater, 2x 40A; my electric clothes dryer, 2x 30A (all 230V service in US). There are at least 15 million houses in the US with similar electrical service. Some industrial plating baths use 6000 Amperes at less than 3V. So 19 million amps is a serious underestimate of the current being used in the world.

MY EYES!

[imsabbel]

1.Dont use bastard childs like TkW
2.Power =! Work. So its Watt. Not Watt/s. or anything. WATT. So the Power rating wont change if you make it shorter.
3. Scientific notation, growing out of your ass: 5.61161e-12 TkW you write... well, thats just 5.61kW... maybe you mean something different?!
and 2.36e-12 Trillion Volts... well, thats 2 AA cells, definitively archivable ;)

Re:Current != Power

[yecrom2]

so, if there are 86400 hours a year

There aren't 86400 hours in a year, at least not on this planet. There are 8760 hours in a year. (other than leap year)

There are 86400 seconds in a day.

Re:Current != Power

[Binestar]

According to the CIA World Fact book, the world uses 15.29 trillion kWh of power a year, so, if there are 86400 hours a year, then we use 1.769676e-4 Trillion kW a year.

*blink* *blink* Typo? We would use 17,696,760 kW/hr (I'm human, I don't mind rounding long numbers when the answer doesn't need to be perfect)

This computes to 5.61161e-12 TkW a second.

295,945 kW/sec

So, if this thing ran for .02 seconds (I think they said for "milliseconds" then, they would need to generate (4x) 1.12232111e-13 TkW to make this thing work.

5,918kW/.02sec

So the voltage used would have to have been [4*1.12232111e-13e-13]/19000000 = 2.36278128e-12 Trillion Volts.

0.0236278128 volts? I may have misplaced a decimal point, because that looks like a pretty small amount. But then again something to e-12 is small, even if we're counting it in trillians (e10)?

Now, please take MY numbers with a huge grain of salt, I'm definately a layman in this, but I just thought his choice of not converting to layman human readable numbers was a obfuscating method of displaying the information.

Also, just punching your numbers into google shows that the final number should be 2.36278128 × 10(^-20). So one of us is way off here, and I'm not an electrical engineer, so there is a good chance it's me.