Exhaust Heat Recovery - Steam Power

[jdgFirefly]

At least this gives a rough idea of how much steam will need to be generated.

What CFM is your die grinder rated at?

[Christ]

Originally Posted by Daox

Well, apparently youtube isn't liking my video file. I've uploaded it a few times now and I just don't get sound.

Anyway, the die grinder could not spin the alternator fast enough with the field coil powered up. As you can see, the voltage slowly drops on the battery as the video progresses. So, we need something capable of flowing more air or higher pressure, or both.

Ok, well, you can use an air ratchet, impact gun, a larger air grinder, air drill, or any of several other tools.

You can also try porting the die grinder to allow cleaner air flow (requires a flow meter and a good, constant stream of air). It's like porting a head, only on an air tool.

Are you sure your compressor and hoses were capable of flowing the required CFM for the tool to operate at the max PSI?

120PSI at 30CFM is far different than the same at 60CFM. It's not really the pressure that matters in the equations, as much as the mix of pressure and flow volume.

You probably won't want to use an impact wrench, because of the impact part. When it gets loaded at all, it will start to slap. You'd more than likely want to try out an air drill or air orbital sander, both of which are capable of the 10k RPM that may be needed, and are normally required to produce fair amounts of power from compressed air.

Also, keep in mind that steam flowing is much more dense than air, and heated water (still in water form) is much more dense than either... density = more potential to produce power.

Mix your steam water 50/50 with coolant. The coolant has a lubricant in it that might protect the air tool so you don't have to remove it, oil it and reinstall it every time you run to the store.

Capturing the "exhaust" from the air tool will prove somewhat difficult, since most of them vent randomly through a series of holes in the body. You'll probably have to break it apart and change the discharge area so you can recapture the steam to recondense it into coolant/water using a condensor (radiator) in the system. This would be an example of a closed system that could be effectively used, as well. As the coolant mix in the heating area is heated, it will build pressure, and move away from the heat element, flowing through the air tool, creating power. As it flows through the exhaust port of the air tool, some of the energy of the mix has been extracted, so it's going to cool some there alone, but then, to further cool it, you install a radiator in the circuit to waste off the excess heat that wasn't used in the production of electricity. Directional flow control comes via a simple check valve, like those used in coffee pots, only much larger. The check valve's spring face would also ensure that the water is sufficiently heated before being allowed to flow, ensuring that any movement in the fluid is due to pressure created from heating it.

Running an alternator at it's peak efficiency all the time will net more power production than is actually necessary, as well... it would be great to have the battery capacity to store it all, but in the event that we don't, the excess power should go where? Do we just turn off the alternator field and allow free rotation when power generation isn't necessary? I don't personally like that idea, but it may help to get the project started, then fix the minor follies as they are addressable.

[Piwoslaw]

Bearleener once tried recovering exhaust heat to mimick the gen3 Prius, but failed.

Originally Posted by Bearleener

I tried to build an exhaust heat recovery unit but it didn't work. It's probably too small. It's basically thin copper tubing (5 mm) bent into a double back-and-forth loop so it has an overall length of about 14 cm and a total contact length to the exhaust pipe of about 50 cm. I attached this to the exhaust pipe just downstream from the header (i.e. upstream of the catalytic converter) using 2 hose clamps and let water flow through it and thin rubber & tygon tubing. The idea was to pump the heated water through tubing wrapped around a steel pipe on the engine through which the coolant returns from the heater core to the water pump.

About 30 seconds after starting the cold engine the exhaust pipe was definitely too hot to touch. But after several minutes of water through my heat exchanger (at a rate of about 500 ml/min.) the water was lukewarm at best. After about 3-4 minutes the normal coolant tubing was already quite hot.

That is, my heat recovery doesn't work. Either the thermal contact is bad, the contact surface is too small, or it's difficult to extract lots of exhaust gas heat through the exhaust pipe wall. Also, I'd probably need to use much thicker tubing, larger contact area, and circulate the coolant directly rather than using an intermediate water circuit. It may also help to use the exhaust header directly (greater heat capacity --> warms up more slowly but doesn't get cooled down as much by the heat exchanger.

[Christ]

Regarding how to actually extract the heat from the exhaust - Try this:

[Christ]

It's probably not the cheapest solution, but it has water inlets and outlets already, and is designed to run water through the ports to cool intake air for turbo/supercharging applications, however, if you were to pipe your exhaust through it, you could replace one or more of the mufflers/resonators on your car, as well as extract a good bit of heat from the exhaust into the coolant mix via the series of small passages in the body of the heat exchanger, which is specifically designed to remove heat from the passing air and deliver it into water.

It's practically designed for this application, just in a slightly different sense.

That particular unit, from the linked website (pic is link) runs $169. You can probably find/make something cheaper, especially if you can weld aluminum and find a junkyard with turbo parts. DSM cars typically came with small side-mount intercoolers that might work effectively if the cooling air portion were blocked and bungs installed to flow liquid cooling medium through the heat exchanger.

I also have no idea what type of restriction there would be in such a setup... however, I expect minimal, again because of the designed intent of the item. Pressure drop across an intercooler is usually very little, even after cooling.

That is a topic to which we should be paying attention, though. Engines are designed to run with a specific exhaust velocity and volume, and the volume that the exhaust occupies will be less once its' cooled, as well as the velocity having slowed. This could have an adverse effect on engine/emissions controls and performance. Something to think about/keep an eye out for.

[Piwoslaw]

I found a prototype home furnace with a 1kW stirling engine running a generator (more info at EcoRenovator.org). If this would finally go mainstream, then after a few years there would be used stirling engines at scrapyards An engine with a 1kW generator for the price of scrap metal (120kg of scrap metal). I'm sure that if this caught on there would be smaller models available.

[Daox]

Originally Posted by jdgFirefly

At least this gives a rough idea of how much steam will need to be generated.

What CFM is your die grinder rated at?

Unfortunately, I don't have any idea. Knowing that would probably be very helpful though.

[Daox]

Thanks for the ideas Christ. I did think about the steam being more dense. I wish the sound worked on my video. If it did you could tell I really need a fair amount more power unfortunately. I'm pretty sure the tank/lines are up to the task of powering the grinder. I'll just have to try a different tool. I had considered a drill, but apparently they don't really spin too fast. I just checked a few on ebay and they were all less than 1k rpm. The random orbit sounds like it might be more promising. I'll take a look at the CFM rating of some tools and see what kind of tools generally take more CFM. I'm guessing those tools will have more power behind them.

Of course I could always go back to the hand made turbine if none of this works too. That would be very easy to tweak as I could just use a pipe plug and drill it out for a jet. We'll just have to see.

[Daox]

Yikes. While looking at air tools I found this:

Ingersoll Rand Air Vertical Polisher — 7in. Dia. Pad, 1 HP, Model# 318 | Air Polishers | Northern Tool + Equipment

Its a 1hp polisher. I figured, oh, that'll be good to look at to compare. 22 CFM! Gonna have to recapture a LOT of heat to power that thing.

Doing some rough calculations:

1hp = 745W

With alternators only being ~50% efficient, thats about 375W. Plenty to power the car, but achieving a steady 22 CFM sounds like it'll be quite difficult.

[Christ]

More than likely near impossible. There are surely more than 22CFM of air flowing in the exhaust for direct capture, but again, at what pressure? You'd choke the engine out trying to power the tool directly, I fear.

However, if the system were closed loop, never allowed to actually become steam, water is ~800 times more dense than air of a similar temperature, right?

22/800= .0275CFM of heated water flow.

That might not be correct, and I'm sure someone else knows how it should really be, but maybe it's at least encouraging?

What I'm suggesting is that pressurized water will take FAR less volume to produce the same power for a given pressure.

If you figure that your tool takes 22CFM at 90PSI, it *should* only take less than .25CFM of water at the same pressure. That means you can lower the pressure significantly while upping the flow rate, and you'll achieve the same end.

So, if you can't get the pressure up to the 90PSI that most air tools require for max load/power, you could just flow more volume at a lower pressure, like 1CFM of water at 60PSI or 2CFM of water at 30 PSI (not accurate figures, etc.)

Also, since you mention that your grinder felt lacking in power -

1. It wasn't meant to be strong, by any means, so no news there.
2. The power curve of air-powered tools is not even close to linear. It spikes somewhere in the chart, and that's about where it's most efficiently run for the power/air pressure/volume charade.