another air conditioning proposal: engine-driven compressors

A recent thread discussed the idea of pulling cold air down from the upper atmosphere for cooling office buildings; it was shown not to be viable.

But it got me to thinking about this one. Air conditioners typically use an electric motor to drive a refrigerant compressor. My understanding is that power plants typically operate with a thermodynamic efficiency of around 50%, but the cross-country transmission of electrical current wastes quite a bit due to resistive/reactive losses, with the net result that the electricity delivered to you represents about 25% of the energy released from the fuel that was burned to create it. In converting the electricity back to mechanical shaft power (to drive the compressor), the electric motor introduces another inefficiency. It’s not bad (95%?), but it’s one more step.

When it comes to converting chemical energy into mechanical shaft work, large-displacement engines operated at a substantial fraction of full load can be considerably more efficient than this, i.e. over 40%. Has anyone started using air conditioning systems in which the compressor is driven directly by a combustion engine (either natural gas or diesel)? Engines are of course considerably more expensive than electric motors, but I wonder if the long-term savings of buying diesel fuel or natural gas (instead of buying electricity) can pay for this.

Note that I’m not talking about the small units placed in your home’s back yard; I’m talking about the multi-ton cooling units suitable for meeting the demands of a skyscraper or casino.

This is not true at all. This would represent 50% transmission losses. The actual number is around 7%.

If transmission losses were as high as you assume your proposal would make sense for all sorts of applications, although you would still have issues with emissions, noise, etc.

With transmission losses and peak-demand charges it can sometimes be cost effective for some businesses to generate their own power. I know of one feed-mill that did such. They got a good deal on re-built Diesel genset, and hit break-even after around 4 months.

The thing that makes this less common than it might be, and would apply to the OP case of running mechanical loads directly from an engine, is that most businesses don’t have two brothers as owners who are retired gear head/top-fuel dragster mechanics like the feed-mill I mentioned above. At a minimum, you have to change the oil regularly, slap on a new filter, and occasional actual repairs will be needed, and when that happens your business may be shut down until it is fixed, and it is all your problem to deal with…not just a matter calling the utility company and demanding they restore service.

You have to schedule fuel deliveries, and store that fuel in EPA approved ways, and hope you can afford what it does to your fire insurance premiums. There is also noise and smell of running big diesels, and dealing with the neighbors. If it is not so reliable, then there is a fair amount of expense associated with retaining the utility as a backup.

You know how hard it is to find a good, honest, efficient automobile mechanic, right? Now you need to find one that makes house calls as well.

Oh yeah, that feed-mill had about an 80% reduction in motor rewinds after they went off-grid. Enough that the motor shop inquired about the loss of business.

There are industrial uses for engine driven compressors, one that I’ve dealt with a few times is ice skating rinks. They tend to use engines for their main pumps for compressors and cooling towers.

I think this has more to do with on demand high hp than it does efficiency. Most the rinks in my area are older and may not have been able to bring in larger electrical services necessary to power equivalent hp electric motors.
I think for residential applications engines would have to be incredibly more efficient and reliable than electric motors, which they aren’t.

This will take too long to catch the edit window:

I am not sure of current practice, but small engine driven refrigeration systems were once rather common on rail cars and semi-trucks.

With regard to AC loads in particular, electrical drive has a huge sealing advantage. Mechanically driven compressors, like on cars, require a seal at the shaft to prevent leakage of refrigerant. These seals are never perfect, and the system will eventually require topping-off of the refrigerant.

Small electric AC compressors have the motor and compressor inside hermetically sealed housings (cans) , and only non-moving wires need to be sealed. These systems are normally leak-free, and will hold refrigerant until something fails. Larger compressors are “semi-hermetic” meaning they are still sealed, just not contained in a can, and can be disassembled, repaired and/or rebuilt. In both cases the refrigerant and oil is used to cool and lubricate the motor…so it is a very holistic approach.

Driving an AC compressor electrically is such a winning solution that people keep looking for ways to make it practical for automotive applications. Sadly, the motor would need to be impractically large and heavy, as well as requiring a huge alternator.

Why the reduction in rewinds ?

Interference from other devices connected to the power grid - perhaps at other nearby businesses - can distort the waveform of the AC voltage, which is nominally a perfect sinusoid. An AC motor is designed to operate most efficiently when driven by a perfect sinusoid. When driven by a distorted sinusoid, it will draw more current, and this may lead to overheating and reduced lifetime.

The general term for this type of this is power quality:

Suspected far fewer spikes. Especially from lightning, but it might also be less harmonic content on the waveform. If the motor is not overloaded, it is insulation of the windings that is the weak link. Spikes can arc that, and harmonics cause vibration that mechanically stresses the insulation.

Yanmar make internal combustion engine driven refrigerant compressor heat pump (refrigerant air conditioning and heating) units.

Another thing would be power factor. With in generation you can overexcite the generators to bring the power factor from lagging to ner unity. There for the motor will use less amps for same KW,

For large applications onsite power generation makes a lot of sense. The University I went to had its own steam plant and the U wasn’t really all that big. You start to lose that economy of scale when you’re talking about a window air conditioning unit. Then add it the inconvenience of having to have fuel on hand to run it; one of the great advantages of the grid is always (well mostly) on power whenever you need it. Don’t discount distribution as a key factor; even if it costs more the convenience may be something people are willing to pay for.