Rotary Heat Exchangers - Shearing and Centrifugal Force

Hi,

if you use ambient air for heating you are wasting a lot of cold that should be used to cool down the air to be liquefied.

So change your system design.

As the cooling down will need previous drying of air (have a look on absorption dryers) your problem will evaporate.

If still you stick with the original design please clarify why this waste of energy is making sense.

If ok, then I have some ideas what to do.

RHABE

Geeehh!! Wrigth wrigth..... This migth sound something like a turbine mechanism itself. No bad at all. The thing sound almost like an centrifugal rotary unit which are a good design itself. Yes you will need to make out some good transfer surface material between those surfaces to make it nicely efficienttly at same time. You have the technology! It'll come to you eventually by a little trial and error approaching I'll bet.

Don't rush into nothing and easy does it my friends.

TurboTech,

MC

Hello RHABE:

Although I am at the "other end" turning already liquefed air back into energy through evaporation into ordinary air, I would very much like to hear your ideas on efficient methods of air liquefaction.

My only near term relationship to liquefaction of air is to pay the manufacturer to deliver and fill a Dewar flask. How he makes it is no concern of mine. Our next project, however, will be to design an inexpensive, small, and long lasting device for liquefying air, and we will appreciate any new ideas.

Until something better comes along, there is a proven design using the Stirling principle on the market.

There is also what here in the USA is known as the "ELECTROLUX'' type household refrigerator which used a gas flame, water cooling, and ammonia as a refrigerant. It seems to me, although I have performed no calculations, that this principle could be combined with a mechanical compressor and utilizing the heat from compression of the air to further cool the incoming air.

Back to my original question: do you have any idea how to calculate the power requirements to keep frost shorn from a disk through centrifugal force? I suppose I could calculate the weight of an individual frost crystal and calculate how much energy was required to accelerate it to the speed at which it would break free from the cold plate. Then I would have to estimate it's weight, and the various quantities available from the air at various humidities. Then I would have to guess at what percentage of the frost attached itself to the spinning disk and what percentage formed in the air like snow.

It seems to me that this is a truly insignificant factor, but I suppose I have to prove that it is. Any suggestions?

bill michaels

Using a spinning hollow disk containing liquid air as a heat exchanger with ambient air certainly is a novel idea. A energy flow balance must be achieved. This will require cooling a tremendous volume of ambient air --which somehow must be placed in contact with the spinning disk. The disk will have to act as a free air centrifugal blower wheel with ducts conducting warm ambient air near the hub where it gets entrained by vanes,grooves, lands ? on the rotating surface of the disk, and gets accelerated outward and discharged away from the warm air intake.

One can estimate the energy needed to accelerate the required amount of air (a delta T on the air side will have to be specified) as well as estimating a pressure drop through intake ducting and to raise the discharge pressure above ambient to insure remote discharge. An efficiency of such a free air compressor/blower wheel will be low--perhaps less than 50%.

It is not a trivial or simple design when you start considering just SOME of the unknows and factors that must be determined -friction coefficients, viscosity, heat transfer coefficients on the air side fins (you were planning to use extended surfaces weren't you) and so forth.

Keith E. Bowers:

Anticipated design values simplicity, economy of construction, and long life. Motor initially is anticipated to be in the 5 hp range. Maximum efficiency is of lesser concern. Due to lubrication problems at cryogenic temperatures, it is designed to be "one piece" except for the governor mechanism.

Yes, I hope to make the heat transfer areas of the rotary heat exchanger smooth for better frost stripping and less noise. In present design, ambient air enters through the hole in the "washer" and is moved by what I call "boundary layer attraction" between two disks maybe .030 inch apart or less like a "Tesla turbine" air compressor. It would seem that sufficient turbulence even with a smooth disk at this spacing would suffice to keep it from "layering". I plan to add however many disks it requires to achieve desired power output.

There is a lot of temperature differential available between ambient and -312F, and I hope the boundary layer factor will move quite a lot of air in close contact to the heat exchanger surface, and that my calculations for heat exchanger areas are correct, and that it is as "doable" as it seems to me to be.

The design calls for the pressurized air from the evaporated liquid air being discharged at the periphery of the spinning heat exchanger (think Hero's engine) resulting in a "one piece" heat exchanger/turbine design. Liquid air pump will simply be a standpipe conducting the liquid air from a port in the axle and extending almost to the rim causing centrifugal force to cause a flow, and also serving as a safety valve by allowing pressure to back up into the supply tank upon exceeding the pressure obtainable through centrifugal force times the moment arm. For starting, a pressure cap on the supply tank should suffice. A governor will have to be employed to prevent overspeeding.

In dealing with cryogenic power storage, what represents a negative value often becomes a positive one. Friction, momentum, heat transfer or radiation to the atmosphere which are clearly losses in a heat engine operating at over ambient temperature often become plusses. Then there is what I call "the heat pump factor" whereby if you can steal a little heat from the "warmant air" by compressing it a little you would seem to have decreased the natural inefficiencies of the system through slightly raising the working pressure of the turbine.

While I have been working on this design for quite some time and still consider it "elegant" don't be too polite to point out any errors in my reasoning.

thanks for answering

bill michaels

Hi,

A.: why rotary, did you think stationary - much simpler and fewer problems. Conducting heat from cold to warm, connecting to finger-structures to facilitate the fingers to air heat transfer, blowers to keep the temperature of fingers above freezing, drainage to get condensed water out of fingers.

B. If rotary B1: keep surface warm to prevent ice, B2.: make surface flexible and introduce bending or vibration to remove ice - as in aircraft wing deicing.

B1: Calculate heat transfer from surface of metal to air heat exchangers over heat conducting structure (metallic or ceramic? or switchable?, to surface in contact with liquid air. Liquid air boiler surface to be structured to enhance boiling quality.

B2: Rubber structure (Viton or silicone) subject to vibrations or bending will break up ice-layers and throw off ice chips.

RHABE

there are (4) different types of shear forces.

Rotational, Linear, telescopic, dammed if I can think of the other one right now.

Now thats getting into rheology of the product. Oh I remember, the fourth is axial, I do not believe thats the type of shear your talking about, I believe the term you may be thinking of is gyoscopic forces.

Throwing something at you, have you considered scraped surface heat exchangers?

phoenix911

Thanks Pheonix 911:

Ok, how about educating me a little. If I have a disc mounted on an axle and I have something bonded to this disc with an adhesive of which I know the tensile strength in pounds per square inch, the amount of force necessary to break it's bond by pulling in a line parallel to the axle should equal the surface area multiplied by the tensile strength.

Now if I force the "something" sideways (perpendicular to the axle) until the bond is broken is this force equal to the force in the first instance?

What about if the "something" is at least as tall as it is wide, and centrifugal force is increased until the bond is broken. Can I assume that it will break at half the force of the first instance on the theory that as long as the compressive strength exceeds the tensile strength, one half of the stress is in compression and what is on the other side of the neutral axis is in tension? Must we take the "vertical" center of gravity of the "something" into consideration? A trig function perhaps?

Since I can increase the centrifugal force all the way to the point of disintegrating the heat exchanger disk, I believe this to be far beyond anything required to displace the frost.

My problem remains of how to convince the "peer review" person that flinging an occasional ice crystal which had momentarily bonded to the heat exchanger surface into the exhaust stream of the "warmant air" is of neglible additional energy cost as the weight of the the air (including the moisture which had made the frost) and the energy costs of moving it had already been calculated. It also seems reasonable to assume that most of the ice crystals formed in the air stream at the point where the temp dropped to below freezing near but not touching the cold heat exchanger and simply blew out like snow. I either need to convince him that it doesn't matter or show a calculation of it, which noone I have met seems to have any idea how to do considering the many variables..

Yes, I first considered and rejected the use of scrapers for several reasons, including space requirements, wear, noise, friction and the fact that in the real world it would be necessary to have sufficient clearance that a thin layer of frost would remain impeding heat transfer and causing balance problems. They also could be expected to freeze to the disks when shut down requiring large amounts of starting torque to be available to break them loose.

bill michaels

Have you considered a surface treatment that would propagate the creation of larger crystals? The larger mass would take lower rpm to induce the energy to overcome the adhesion.

There is a larger problem than simply the mass of a single crystal. It will build up crystalline structures that are very light weight and relatively strong which will support each other.

You have the rpm range to disintegrate your disk. What is the range? What is the failure rpm? Given a slim safety factor of 80% of your top rpm before failure and the diameter, the available forces in relation to a crystal mass should be straight forward. It should tell you quickly if it is even possible. This, of course, will only apply to a specific diameter range on your disk. The centre of your disk will look like a snow cone.

Also there are a lot of anti-frost surface treatments. Even nano-tech that you could look into.

Gdevine:

Thanks for the response. I wasn't aware that frost crystals came in different sizes. I had assumed that a given temperature and timeframe would produce similar sized crystals regardless of surface texture; just more of them on a very rough surface. Can you educate me a little in this respect?

I had checked available surface treatments, and concluded that a very thin layer of teflon would decrease the force required to shear the frost from the substrate by almost a factor of ten, and at this point jumped to the conclusion that it was easily possible to shear frost within the maximum centrifugal force which could be applied to the spinning substrate. This assumption, in retrospect may not have been valid.

There seems to be very little data available applicable to shearing frost from spinning substrates; about the only being fifty year old results from an experiment conducted by Cambridge and they weren't using centrifugal force.

Ice (which varies in characteristics to a surprising degree depending upon the conditions under which it was formed) can clearly be calculated as to the point where the forces exceed the adhesion values, but what about frost? I assume it is too good to be true to hope that it compares to ice as the respective weights for a comparable mass. It is clearly lighter, but it's surface area to be considered in evaluating adhesion values would be much less also. It would not surprise me if squares or square roots were involved in the comparison. I am obviously in over my head here.

As to the center of the spinning substrate collecting frost, there is no center. The first prototype will be an 8 inch disk with a 4 inch hole.

I am beginning to wonder if we have here an instance of a beautiful theory being destroyed by running into an ugly fact.

Thanks

bill michaels

If you use a wiper, be careful not to compress the frost layer. If you do, it hardens into ice (as any good snowball maker would know).

A teflon layer should be effective in reducing the sticking of the frost to the surface.

As I recall our LOX tanks on a mine I worked at, we got rid of surplus frost (and ice, which tended to form if the frost layer got too thick) by hosing it off. This may not be feasible in your case, so you would need to use a scraper.

I've never known a rotary heat exchanger to rotate fast enough to generate significant centrifugal force, but you appear to be using this almost like a turbine, in which case you may be able to have it drive itself and produce some surplus energy.

To aid in your efforts?

-very fine surface treatment to reduce adhesion?

-surface treatments?

-wiper?

-ultrasonic de-icing?

...just throwing darts here.

Would not a simple test platform answer your question?

Hello gdevine:

Current plans include a polished surface coated with a very thin layer of TEFLON.

Yes, a simple test platform would be great, but costs money to build. In order to get the money to build this test platform, I need to already have the answers to what I wish to prove as a prerequisite to getting the research money. In other words, give me the results of your research up front and maybe we will give you the money to conduct the research. I have been trying to raise $5000 for this purpose but no foundation has a category for portable storage of power for small devices.

We have a "not-for-profit" research foundation classification plus being classed as a "private charity". Until Indiana changed the rules, Purdue had done app $5000 worth of research into adhesion and tensile strength values of ice for us, but now such research requests are subject to "peer review". I haven't attended Purdue since 1957, and certainly have no recent connections or credibility. My training was in civil engineering besides.

I have contacted a few professors and proposed a partnership whereby they help me get the grant without payment, with the possibility of consultant work once the grant has been approved, but have received no answer.

My current plan is to learn enough to present a coherent "complete thermodynamic analysis" myself to continue the grant application. My fallback position is to write a book on the subject and hope to generate research funds in that way.

Ultrasonic de-icing would be expensive, require electric power and present space problems with the closely spaced disks.

The application closest to my heart is for such a motor be used to power a garden tractor or tiny truck running on liquid air enabling solar or wind power to be used instead of oil as a means to self sufficiency. This means maximum simplicity and low cost.

Thanks for your help

bill michaels