The most awesome startup I have ever seen. | Hacker News Reader

The most awesome startup I have ever seen.

danshapiro.com

44 pointsdanshapiro19 comments

It's a cool building, but I was a little less thrilled when I read about their tech. It's true that an isothermal process is a reversible process, and isothermal air compression is therefore a great way to get high efficiencies, but compressed air just doesn't seem like it will scale well. The work formula for isothermal compression of an ideal gas is:

    W = - ∫ P dV = - n R T ∫ dV / V =  P₀ V₀ ln(P₁ / P₀).

A shipping container holds 38.5 m³ according to the Wiki. At atmospheric pressure this would mean P₀ V₀ ~= 1 kWh. Cheap hardware might get you to 10 atm pressure or so -- but let's go crazy and suggest that you could get 100 atm, since it's only logarithmic in P₁ anyway. You would still store less than 5 kWh per shipping container, no? That could run a window-unit air conditioner for about a day, only. If you pay 15 cents per kilowatt hour, you could fill up the shipping container with under $1 of electricity, only. My cell phone battery stores an amp-hour at around 3.5 V, so if the above calculation is right, to store the same amount of energy in 100-atm compressed air you'd need 27 liters of air -- two backpacks or so. I'm saying this to guesstimate that it's about a factor of 1,000 less energy density than modern battery technology -- and that that's a fundamental limitation to the medium.

Given all that, I'm really interested to see how they'll scale compressed air up to handle the sheer amount of energy that they want to store.

(On the other hand, the factor of 1000 might not matter too much: it means that if they can build a shipping container air storage unit cheaper than electronics companies can build a battery the size of three backpacks, they could indeed be cheaper to store energy en masse.)

Dude, I like your analysis and eager attempts to calibrate but you messed up your calculation somehow.

I'm pretty sure your problem is that your formula misestimates the amount of gas stored. It is compressed, you know. You of off by about a factor of 200 - we use 200 atm air. It's 1.1 MWHr per high-boy shipping container

The energy stored is

efficiencyOutgoing * P_top * V_tank * (ln pressure ratio - 1 + 1/pressure ratio) -- you can look at our patents for more detail.

Sorry, you're pretty much right. The equation I wrote is correct, but I naively plugged in a bad number: yes, P₀ V₀ = n R T = P₁ V₁, but I used the known number for V₁ (the final air in the container) as V₀. The proper formula is indeed P₁ V₁ ln(P₁ / P₀).

You have my apologies for my confusion.

You have to account for the ~22% loss due to the packing density of cylindrical tanks within the container.

So that 38.5 m^3 is closer to 30 m^3 effective volume, maybe another 5% loss due to end-caps and plumbing.

This is just about the correct estimate of usable space.

Super stuff you're doing Dani, keep it up :)

Definitely rooting for you, compressed air has been tried many times but you're bringing a bunch of interesting new takes on the idea to the table, it would be great to see this nasty little problem solved.

Possible side-line/byproduct: UPS for datacenters / small businesses.

I think after the horrible results with the rotational UPSes at 365 Main, non-battery UPS technology is kind of hated by most datacenter administrators.

But, batteries are much more expensive than shipping crates and also wear out with the sort of usage cycles that this sort of tech requires by design.

I was hearing recently about a similar energy startup that was based on cooling air to liquids as a power soak and then shunting that back through turbines to re-extract the power. Similarly less efficient than batteries for power density, but also longer lifed and much cheaper.

Who knows what the best solution is? Not me, I haven't studied physics since I was 16. But I'm glad people who are better equipped than me to find out are investigating, and I look forward to the day when we can use this sort of tech to provide the smoothing effect on renewable power sources rather than relying on burning fossil fuels.

Note that the tanks also "wear out". Their life-time depends partly on how much pressure they use relative to their "maximum".

Also, wet air has corrosion issues, not just in the pumps, but in the tanks.

These issues are $olvable.

Our material system doesn't react with oxygen at these temps, so there's basically no degradation...

Poking around suggests they are using a modified natural gas compressor. That doesn't give the pressures they are working with, but CNG, when used as a vehicle fuel, is stored at closer to 200 atm (which again, doesn't suggest that they are working at those pressures, but 100 atm isn't crazy anymore).

also, dont forget pv=nrt: temperature rises linearly with pressure, so the air gets pretty hot as you compress it from 1 atm to 100 atm.

A) temperature normally rises as the 0.4/1.4 th power with pressure in an adiabatic process

B) our water spray makes the process near isothermal

"middle school-dropout"

Not the words that I'd use to describe someone who starts studying for a degree at the age of 12.

Me either, but that's how she describes herself.

http://daniellefong.com/about-the-author/

"Danielle started her PhD at 17" -- Thanks for ruining my monday!

Just kidding, awesome idea. We're so used to IT startups that we often forget about other engineering. Very cool stuff.

Wow. These guys rock.

Thx :-)

Found additional details here: http://daniellefong.com/2012/01/22/green-dreams-life-in-the-...

Reply on news.ycombinator.com