> You're probably right that reducing excess ambient air flow coming up between the windscreen and pot maybe have helped by keeping that annulular air space hotter. In which case, probably better to just reduce the windscreen diameter a bit.

Missed this last night.

I think that reducing the windshield diameter might have the opposite effect to the one you're trying to achieve; it may simply increase the air velocity due to the chimney effect, thus sweeping the hot gas out of the stove before it's given up its heat to the pan.

My thought on the vanes is that they might slow the air flow by fluid dynamics effects; getting in the way of the 'sticky' air.

Well it looks like David has a little more time to put into this than I do, but if I don't see any updates from him, I'll try to do an experiment.
Plain fosters can, fosters can with fins, black painted bottom fosters can, and fins + paint. I'll use a thermometer to check start temp, and time them all to 212*F. I'll do several runs and take the average, then post the results.

Of course, that's IF I can find some time to play around with this.

Nick,
If you take the temp readings to 200F you avoid the the majority of the vaporizion changes that occur just before the boiling point. Lids become quite important for testing. I live at low altitide, ~1000feet, so, 212, is out anyway. Water boils at 211.4 or something depending on the weather. 200F is safe enough for sterilizing any water, anyway. It also makes any test repeatable across the vast majority of the USofA. Jerry Gollier (backbackgeartest.org) does the same. So, it is easy to compare results, with others. Besides, 200F is a nice number to remember. As always, your choice.

Nick: I aim to repeat with the can bottom painted black as well, but maybe not this weekend as we're taking the kids skiing (X-C on Saturday, Alpine on Sunday). I also want to play with the metalic tape on the side as VGs.

Regarding air velocity: In general, don't think of it as "time to transfer the heat". Think of turbulance as good for heat transfer. Like wind chill on your skin - higher air speed makes for more heat transfer.

Sure, I guess I could go to 200. I'm more or less doing this to see if the fins or paint get us to that temp faster than just a plain pot, so i don't think it will matter what temp I go to. As long as my results show a significant decrease in time or not. If my tests show that you save a minute or more with the fins, it will show that they are worth using regardless of what temp you go to.

OK, now that is cool. The big barrier to HX pots is the weight of the HX. Typically the weight of the exchanger is greater than the weight of the fuel saved by the exchanger, except on extended trips. Reduce the weight of the exchanger, and suddenly you've removed the barrier to using HX pots, and you can have real efficiency gains. NICE.

HJ
Adventures In Stoving

Edited by hikin_jim on 01/16/2012 09:40:07 MST.

Here's another possible vane configuration, which provides a 'windshield' to the heat exchanger 'tunnels'. It's from one of the 'JetBoil scribbles' I mentioned earlier. I've just drawn a pretty SketchUp version of it applied to a Rexam King Can (1000ml).



The idea is that the foil is folded back in itself to create triangular chimneys. By designing the inner and outer fold lengths correctly, we can form a continuous wall outside the chimney, thus enclosing the hot gases entirely. The number of vanes shown here is purely to make the SketchUp drawing easy (20 degrees per segment).

ps. I have no idea if this might be beneficial; it's just an idea... The outer segments may be dead weight, since they're so far from the can, and unlikely to drag heat back to the can; more likely to radiate it away...

Edited by captain_paranoia on 01/16/2012 11:30:15 MST.

So here is an idea that I was tinkering with a while ago (BTW, I thought about this design after seeing the Squeeze Box concept – great idea). The idea was that channeling the air close to the can may be as beneficial as adding fins. That way, you would place the design burden on the windscreen and use a generic pot. This is a windscreen that was designed to wrap around the mug. The idea was to channel the exhaust gasses close to the mug surfaces. The shape also rolled up and fit inside the mug. The profile is not symmetrical as I was experimenting with creating inlet ports that would enter tangentially to the stove with the hopes keeping the flame steady.

The CAD Layout


The prototype using a Snow Peak 700 mug




The windscreen is made if flashing (full hard) and was not too difficult to make. You could shape it such that you would have a pretty snug fit between the mug and the windscreen. I think that the tangential inlet ports were working pretty well. The system worked pretty well, but I think that I needed a better center burn stove to work with this design. My stove had too high of an output and the flame fronts occasionally moves up to the exhaust ports. Bet regards - Jon

Kevin: I like the larger contact area with the pot walls a lot. I realize it is a conceptual drawing. My guess is that a slightly finer spacing - 10 to 15 degrees per triangle - would add more delta heat gain than it cost delta in weight.

I see the main benefit of the outer segments as containing the hot air closer to the pot and the radial fins. Also, if by their extenstion below the pot or their outer diameter, they may also serve as a windscreen to the pot / fins. They do add weight, but the assembly is also "smoother" than mine - not so many corners to snag when packing. If you were to not epoxy it but remove it between uses, you'd something connecting all the fins and the outer segments serve that purpose very well.

How would you envision forming it?

And/or is there any commercial product from which one would scavange something similar?

>"The idea was that channeling the air close to the can may be as beneficial as adding fins."

Jon,

Yes, keeping the hot air close to the pot absolutely helps heat exchange. As does the small "pipe" size (thinking of the triangular sections as pipes carrying air). Flow in smaller pipes is more turbulant and that greatly helps heat exchange.

I like how you can fit a generic pot in there and I assume the windscreen wraps to fit within the pot for storage? That's a big plus.

Is it possible to get it snug enough that it adds stability to the pot? I'm thinking that the outer point on the windscreen gives a considerably larger footprint than the stove base on the ground or the pot base on the stove.

What kind of an air gap or air holes do you have at the bottom to allow for combustion air?

You've got me thinking that maybe there is an overlap in the two ideas: HX fins and pleated windscreen. Is there a tighter spacing than yours but much, much larger than mine that would handle ALL the exhaust gases but also be in firm, thermal contact with the pot?

An advantage of your approach is that you can use very thin (light) material. My heat fins are pipes carrying heat so they need some thickness to do that.

"I like how you can fit a generic pot in there and I assume the windscreen wraps to fit within the pot for storage? "

Yes, and luckily, most mugs are close to the same diameter.

"Is it possible to get it snug enough that it adds stability to the pot? "

In my mind, I was going to put pop rivets at three points to support the mug.

"What kind of an air gap or air holes do you have at the bottom to allow for combustion air?"

I was using 6 1/4" diameter holes on the bottom all on the downwind side of the windscreen.

BTW, the design for the fins around the can is begging to be made by aluminum extrusion. Maybe someone could start a Kickstart project. The die itself should only run $500 to $750. Most of the cost would be in any secondary machining. Best regards - Jon

Okay, I am not an engineer or scientist. But there is a lot of surface area in the Dave's original fin design, which probably helps move some of the heat to the surrounding ambient air. Seems like the triangle Kevin posted might help prevent this, or even wrapping a band around Dave's first design. Also would be interesting to see the boil time if something like the cloth plumbers use when soldering copper pipes in a house to keep the wood from catching fire might be even better... then analyse decreased boil times versus additional weight. Also doesn't brass retain heat better? How much more would brass fins weigh?

Well, so far it's not looking good for my fins, or the black paint. But, my original stove crapped out on me, so I had to switch mid experiment. So I need to re-do my unaltered pot tests. If I'm lucky, I'll get some time to finish the experiment this weekend.
Who knew it would take so long to boil 9 pots of water? If I have time, I'll also play with the fin configurations.

> Also, if by their extenstion below the pot or their outer diameter, they may also serve as a windscreen to the pot / fins.

Yes, that was part of my thinking too.

I'm now part way through extending the HX height, adding exhaust ports & blocking sections, with a view to adding Thermawrap above this, making an HX & insulated pot.

Simple aluminium flue tape could be used to surround the HX, to add a bit of stability if needed.

And I've been looking at high temperature doubled-sided tapes for bonding the HX, e.g. 3M 9640PC tape.

[edit] d'oh! whilst it's easy to add exhaust ports to the outer triangle, it's not so easy to add exhaust ports for the inner triangle. So I'll abandon that cunning plan...

ps. how to make it? tediously, I suspect... That was one of the downfalls of the SqueezeBox, which had two, opposing folds so was even harder. But I think a simple concertina fold should be possible using conventional packaging manufacturing techniques, even if I can't point you at exactly how to do it... I have a sketch with six pins in front of me; make folds around pins, shift & repeat; two pins provide alignment, and the other four are used for the four folds required per section. I've thought about moving blocks, too, hinged around the 'can centre'.

Edited by captain_paranoia on 01/17/2012 11:03:39 MST.

Keep in mind that a real heat exchanger is constructed to be in contact with both elements. In some of the protoypes there's not much contact at all --> that would be more of a windscreen/flame tunnel.

Very nice idea though, will follow the progress. :)

Both air turbulence, AND donduction are important. Here is a quickie sketch modeled after some heat exchangers I was working on for a commercial application. Same principles, really. A simple stamping process to make the actual disks (shown assembled, ie after forming the cone.) Then press fitted over a can. It might be a little awgwarg to carry fully assembled, but the disks could be slipped off and carried seperatly. Probably about 2oz, total weight. I would gues about a 50%, maybe more increase in heating the water in a can, based on the number of disks that were used, of course. This induces plenty of turbulence, captures heated air from the previous exchanger, funnels it back to the can via conduction, convection and radiation. Simple to design and manufacture, not so simple to make at home, but certainly possible. Again, no patents, please. Open and free for use to all.

I see a lot of good designs for fins. I think most of them will work fine at absorbing the heat. The fins then need to transfer the heat to the can. I don't think the fins will conduct to the can very well, though, without some sort of adhesive that is conductive or has fair amount of surface area in contact with the can. I would love to hear some ideas on ways to manufacture a good contact between the fins and can. That's where I am struggling.

It's certainly true that good thermal contact is necessary in order to ensure that the HX fins couple any energy they pick up to the pan, and thence to the pan contents. But, given that David used glue (and probably a fairly thin layer at that, and I'm looking at high temperature tape, these thin layers will have a failry small thermal resistance; certainly a lot better than air.

Bear in mind that heatsinks for CPUs/large FPGAs are bonded using either thermally-conductive epoxy or thermal tape. Also, have a look at the quoted thermal conductivity for the 3M 9460PC tape I linked to earlier. Hmm, on the other hand, don't; it's a terribly vague specification...

The other issue as mentioned earlier is the thermal resistance of very thin foils; whilst the fins may get got (indeed, so hot they melt), the heat cannot get to the pot due to the high thermal resistance (which is why the fins may melt; they are isolated from the 'cold' contents of the pot, which, if containing water or wet food, won't get much hotter than 100C).

Conduction is one heat transfer mechanism. Another is radiation. If we can trap the hot gases within the HX so that their transit time is longer, they have more time to radiate energy to the pot (and the HX). It may be that by the time the gases have reached the HX, they're no longer incandescent, so the radiation may be much reduced.

>"Another is radiation. If we can trap the hot gases within the HX so that their transit time is longer, they have more time to radiate energy to the pot (and the HX). It may be that by the time the gases have reached the HX, they're no longer incandescent, so the radiation may be much reduced."

Kevin, You're right that radiation is another heat transfer mechanism but only solid objects (and incandescing gases, as you note) radiant heat (in the form of visible or IR photons). Invisible gases don't radiant heat nor absorb IR photons. Hence, solid objects cool markedly below ambient air temperatures on a clear, windless night - they radiate heat into deep space. Look for dew/frost on the lawn, your car and other objects and consider their "view angles" of clear sky versus solid objects like trees and your house. It can give you a good sense of how radiant heat transfer does and doesn't come into play.

I'm going to try a different geometry of fins next with an aim to

1) increase contact area (fin base to can)
2) reduce weight of fins
3) sllow for easy fabrication*
4) maybe include an integral shroud this time

Editted to add footnote: *by easy fabrication, my goal would be something you could make with a pocketknife, using materials from a trash can plus maybe a tube of epoxy from the store.

Edited by DavidinKenai on 01/17/2012 11:05:16 MST.

The atmosphere is thin, but it is there, and it radiates back at us. Take an IR thermometer and point it at the sky. It won't read 3K, which is the usual assumption for a black body radiating into 'deep space'. On the OM thread that mustn't be mentioned, someone (Len Novak) tried that experiment, and posted results; near the horizon, it read about -20C during the day. As elevation increased, the meter went out-of-range. It may be that the meter was 'seeing itself', but this NASA link suggests not. This link also looks useful, but I'd need to digest it.

All matter radiates, regardless of its state (solid, liquid, gas, plasma), provided it's above absolute zero. Gases don't have to be incandescent to radiate. The question is: is this radiation is at a thermally-useful wavelength? I'd have to dig a bit deeper to get the answer to that... but my guess is that it probably isn't, or is vanishingly small.

Anyway, back to the HX...

Talking of IR instruments, it would be instructive to use an IR camera on an HX to see what is going on with the temperatures; should be able to see the effect of high thermal resistance in the fins. I'm sure we have one at work. I wonder if I can borrow it for a weekend...