Wilcox Caloric, 1860, open-cycle, regenerator

[spinningmagnets]

A

StirlingWilcox3.png (187.26 KiB) Viewed 4816 times

As the Stirling was patented in 1816, this engine of 1860 hoped to make some improvement on the established benefits of the Stirling. I found it to be interesting and perhaps worth discussion. There is no cooler, and it simply exhausts the used hot air through a regenerator to the atmosphere.

It does employ a "flip/flop valve" to control the flow of incoming air and exhausting air.

Before considering the drawings, understand that there are two vertical pistons with a crankshaft above them. The piston on the left is the "power piston" and its cylinder has an open top with a conventional rocking connecting rod on a conventional crankshaft throw. Both pistons have radial seals (piston rings) and neither one is a loose-fitting displacer.

This brings us to the fascinating piston on the right. It has a linear rod with a stationary rod seal, and its cylinder has a sealed top that the rod cycles through, up and down in a straight line. Perhaps we should call this the "gas cycling piston"?

The pistons cycle 75-degrees apart, similar to the well-known 90-degree Alpha. The gas piston on the right leads ahead of the power piston. When the gas piston is at the top, and begins to cycle downwards, the flip-flop valve is aligned so that cool outside air is drawn into the rod side of the piston. When it reaches the bottom of its cylinder, the power piston is about 2/3rds of the way down its cylinder on the left.

Then, the flip flop valve shifts, opening the rod-side of the gas piston to the top of the regenerator. As the gas piston rises, it forces the trapped air to make a U-turn and flow downwards through the hot regenerator, from the top towards the bottom. As that air passes through the regenerator, it comes out being focused like a gas-jet onto the hottest part of the "hot plate" that separates the working gas chambers from the firebox heat.

As the gas piston and the power piston are both moving up, the gas flowing through the regenerator and hitting the hot plate is steadily raising the pressure. As the gas piston reaches the top, the power piston still has 1/3rd of its stroke to harvest power at the point of highest internal pressure.

The flip flop valve shifts and allows the hot internal gasses to exit out the top of the regenerator in a reversed flow, bottom towards the top. This valve has two independent passageways, and as the hot gasses exit the regenerator, cool gasses are flowing into the rod-side of the gas piston.

As with the "soon to be invented" Rider-Alpha, the power pulse lifts both vertical pistons, and the weight of the pistons falling on the far half of the cycle (along with the flywheel) make this engine run smoothly with a relatively small flywheel. My apologies if I mis-understood some part of this.

B

StirlingWilcox2.png (168.76 KiB) Viewed 4816 times

C

StirlingWilcox1.png (296.26 KiB) Viewed 4816 times

[Tom Booth]

Not exactly, but kind of similar in principle to an idea I was exploring a while back, I think maybe:

Resize_20230930_122523_3261.jpg (130.3 KiB) Viewed 4800 times

viewtopic.php?t=5578

[Stroller]

I wonder how many of these ever got made.

If an oven door had been fitted to the cavity on the left, the operator would have been able to keep his lunch warm while he waited for the beast to get up to working temperature...

[Tom Booth]

The description of the regenerator found here:

http://hotairengines.org/open-cycle-eng ... ric-engine

Seems very different from what is depicted in the illustration above.

The regenerator design is quite interesting in that it expands radially outward from the center where cold air enters through a central tube. The idea being to provide more room as the gas expands outward from the center and when air travels in the opposite direction, less and less space is available for the gas as it cools.

"allows no more space for air than is absolutely required...making the spaces either wider or thicker at the hot than at the cold end."

In a sense, the geometry of the regenerator assists with the compression of the gas as it cools without the need for any moving parts.

[staska]

Lost discussion. If one just remove valve - ones got nothing less, than hot end connected gamma stirling.

Valve itself is fun thing to substract some air cushion which lowers our compression.

Expelling air is a good way to minimize cooler even more.

[spinningmagnets]

I have a lot of spare time to kill at my job, and I've been running a lot of thought-exercises. I am pondering the performance of various configurations, but also...pondering how difficult it would be to build something that I am interested in. I place a high value on design compromises that are affordable and are also easy to make, if...any loss of performance is minor. A "good" device that is buildable is better than a great device that is difficult and expensive to build.

I understand that it can be fun to make something that spins and has a few shiny brass pieces in the mix. That being said, when a design is being scaled up in size to perform work, some of the designs that were fun when they were desktop sized, might not perform well enough to be worth the trouble.

The thought of being able to eliminate the cooling portion of a Stirling has a certain appeal, but...that is not my motivation concerning my interest in the Wilcox. By going to an open cycle, every incoming breath of air is at room temperature. When we compare the two designs, I am tasked with defending a Gamma Stirling by trying to get the working-gas temperature down to room temperature on the cold side. I suspect the hot ends and regenerators on both would have a similar effectiveness, but...the cold end of the Stirling would struggle to shed heat when it is continuously run under load.

For me, this is the heart of the performance difference between a Stirling and a Wilcox. Both of them use a regenerator to create a large amount of surface area to transfer some of the cycle-heat into the gas-heating process. Of course, if we are comparing a thoroughly modern Stirling with several atmospheres of pressurized gas, and also adding the performance of using helium, there is no contest. However, adding those features will complicate the construction and cost "just enough" that such a design becomes less desirable for promotion (in my view, just an opinion).

My interest is in a solar dish that uses concentrated heat to generate electricity. Solar-dish Beta-Stirlings with pressurized helium exist, but they are terribly expensive. I would be happy for a modest device with fairly low output, and my list of candidates has been narrowed-down to a Gamma Stirling, or an open-cycle Wilcox. I am now leaning towards the Wilcox.

I suspect a well-made Gamma (or possibly a dual-Gamma) would perform as good as a well-developed Wilcox, and a part of me is interested in the Wilcox simply because they are not well-known. By that I mean that...if the two types performed the same, I would still be more interested in the Wilcox. I'm only saying this so my motives are clear, and I am not claiming that a DIY Stirling is now obsolete. I still retain a fondness for the Gamma.

I think the most pressing question is going to be concerning the seals on the hot end. I don't know if a bellows is possible (or even desirable), and if I use a conventional piston in the power cylinder, what kind of piston rings have worked the best so far? Does anyone know if the piston rings on the steam-engine forums work on hot-air that is dry?

[matt brown]

I have a lot of spare time to kill at my job, and I've been running a lot of thought-exercises.

Lucky dog, do you guys need any 'help' doing nothing ? (I slept nearly 12 hrs last night due to my day job)

I am pondering the performance of various configurations, but also...pondering how difficult it would be to build something that I am interested in. I place a high value on design compromises that are affordable and are also easy to make, if...any loss of performance is minor. A "good" device that is buildable is better than a great device that is difficult and expensive to build.

Most of these forums are loaded with starry-eyed impossibilities that ignore these 2 basic realities:

(1) there's very little power with ambient charge pressure
(2) there's very little power with low volume ratios

Simply dreaming that a higher thermal ratio while gaming sq/cube rule will increase output is bogus since 10 x 0 = 0

The typical LTD model has a volume ratio around 1:1.025 where when PP=1 then DP=40. It's a vacuum engine where the ambient 'compression' limits the volume ratio equal the thermal ratio when an ideal cycle, but this likely differs slightly due to out-of-phase issues. Nevertheless, scaling this up will still require a balance between the volume and thermal ratios or it's a no-go (will stall).

Meanwhile, choosing a cycle with or without regen while be a major consideration since regen will increase efficiency, but add cost of complexity in both design and operation. If no regen then open cycle may simplify everything, but this would likely require a total redesign, yet this might allow better phasing.

I understand that it can be fun to make something that spins and has a few shiny brass pieces in the mix. That being said, when a design is being scaled up in size to perform work, some of the designs that were fun when they were desktop sized, might not perform well enough to be worth the trouble.

Yep, a small hole can sink a mighty ship.

The thought of being able to eliminate the cooling portion of a Stirling has a certain appeal, but...that is not my motivation concerning my interest in the Wilcox. By going to an open cycle, every incoming breath of air is at room temperature. When we compare the two designs, I am tasked with defending a Gamma Stirling by trying to get the working-gas temperature down to room temperature on the cold side. I suspect the hot ends and regenerators on both would have a similar effectiveness, but...the cold end of the Stirling would struggle to shed heat when it is continuously run under load.

Everything else equal, the value of regen increases with the thermal ratio and decreases with the volume ratio. Putting everything else aside, the gamma has a lot to offer when (typical) cold PP simply due to various reality checks.

For me, this is the heart of the performance difference between a Stirling and a Wilcox. Both of them use a regenerator to create a large amount of surface area to transfer some of the cycle-heat into the gas-heating process. Of course, if we are comparing a thoroughly modern Stirling with several atmospheres of pressurized gas, and also adding the performance of using helium, there is no contest. However, adding those features will complicate the construction and cost "just enough" that such a design becomes less desirable for promotion (in my view, just an opinion).

ECE have a lot of wiggle room that's rarely explored such as an "open cycle" can be 'open cycle' to LP reservoir vs ambient whereby charge pressure exceeds ambient AND can be your gas of choice.

[staska]

I suspect a well-made Gamma (or possibly a dual-Gamma) would perform as good as a well-developed Wilcox, and a part of me is interested in the Wilcox simply because they are not well-known. By that I mean that...if the two types performed the same, I would still be more interested in the Wilcox. I'm only saying this so my motives are clear, and I am not claiming that a DIY Stirling is now obsolete. I still retain a fondness for the Gamma.

I would oppose this a little. Gamma, especially cold end connect, have lowest ratio of expanded gas two totall cc of an engine. And if one just remove valve from Wilcox engine - he will get hot end conncted gamma stirling. And valve itself just isolates work fluid in expanding state from vold cylinder. Which is nice feature too.

All other being equal - practice and simulation show, you may overcome extra/dead space cc by raising pressure to - which itself makes portion of non working volume bigger. But if one isolate some volume while engine is in expansion stroke. It will need less pressure for same output.

[spinningmagnets]

Thank you both for the thoughtful replies, you have given me much to think about.

Matt, I am near retirement and I've been shuffled by management into a maintenance position where I do a lot of grass mowing in the warm months on a riding mower. Nice gig if you can get it.

When pondering the Wilcox engine, I understand that the cool air passing through one part of the valve can prevent the valve from getting "too hot" from only hot air passing through the other half of the valve. That being said I wonder if the air intake could be a separate "suck valve" which is a lightly-sprung poppet valve and as the piston pulls down, the valve simply gets sucked open. Its a primitive type of check valve, using an off-the-shelf part from the ICE engine parts bin.

By separating the one stock valve into two separate valves, I think it would be easier to build. The hot air exhaust would still need to be timed by some cam or crankshaft.

[matt brown]

I suspect a well-made Gamma (or possibly a dual-Gamma) would perform as good as a well-developed Wilcox, and a part of me is interested in the Wilcox simply because they are not well-known. By that I mean that...if the two types performed the same, I would still be more interested in the Wilcox. I'm only saying this so my motives are clear, and I am not claiming that a DIY Stirling is now obsolete. I still retain a fondness for the Gamma.

I would oppose this a little. Gamma, especially cold end connect, have lowest ratio of expanded gas two totall cc of an engine. And if one just remove valve from Wilcox engine - he will get hot end connected gamma stirling. And valve itself just isolates work fluid in expanding state from cold cylinder. Which is nice feature too.

All other being equal - practice and simulation show, you may overcome extra/dead space cc by raising pressure to - which itself makes portion of non working volume bigger. But if one isolate some volume while engine is in expansion stroke. It will need less pressure for same output.

Guys, the difference between cold vs hot PP must be carefully analyzed. Consider this graphic

hot vs cold PP gamma.png (22.17 KiB) Viewed 3903 times

where the various dead volumes (conduits and clearance) are not within PVT values. Note the 300-600k thermal ratio is 2:1 and results in a 2:1 volume difference between similar cold vs hot PP. The hot PP will have 2x heat input AND 2x the work output vs the cold PP, but FROM THE SAME GAS MASS (since the cold PP volume is 1/2 that of the hot PP volume). However, if we increase the cold PP volume by 2x then cold PP and hot PP engines and output are now equal.

Indeed, the same gas mass produces more power when hot PP vs cold PP, but if we have equal size PP, we can equalize output (cold vs hot PP) via increasing the gas mass and DP volume relative PP volume proportional to the thermal ratio. Therefore, in this 300-600k example, if the cold PP volume equals the hot PP volume, then the cold DP volume (and gas mass) must be 2x the hot DP volume for the same output.

In this manner, the cold PP wins, since both have equal input and output from same size engine (same PP volume). The only major difference is that the cold PP requires a larger DP volume and gas mass while nixing all the nasties of hot PP (sealing, lubrication, distortion, radiate losses). One minor difference is that the cold PP will require a larger regen than similar hot PP and this will increase total regen inefficiency proportional any hot vs cold PP volume differential.

I've never seen anyone compare cold vs hot PP this way, but I suspect that all famous SE gurus did exactly this comp early on and dare not mention this out of embarrassment that they elude the obvious...

[VincentG]

For what it's worth, the Essex mostly solved the hot piston sealing issue. The piston dome was thin brass and only .015" or so smaller than the bore, but protrudes up a few inches, preventing the majority of the hot gas from reaching the piston sealing area.

[staska]

Indeed, the same gas mass produces more power when hot PP vs cold PP, but if we have equal size PP, we can equalize output (cold vs hot PP) via increasing the gas mass and DP volume relative PP volume proportional to the thermal ratio. Therefore, in this 300-600k example, if the cold PP volume equals the hot PP volume, then the cold DP volume (and gas mass) must be 2x the hot DP volume for the same output.

In this manner, the cold PP wins, since both have equal input and output from same size engine (same PP volume). The only major difference is that the cold PP requires a larger DP volume and gas mass while nixing all the nasties of hot PP (sealing, lubrication, distortion, radiate losses). One minor difference is that the cold PP will require a larger regen than similar hot PP and this will increase total regen inefficiency proportional any hot vs cold PP volume differential.

And one more nail - gas shuttling is not a freebie. So to move twice amount of gas - we will have twice pressure loss. And even more dead space in wrong areas. Plus heating and cooling twice amount of working medium will add some thermal losses too. and not small one. And power is beeing produced again by same amount of working fluid. Which will adiabatically compress/expand and additionally load our heat exchanger.

If you could correct your draing to have equal size of power piston - it would show this situation.

Btw - extra gas movent between displacer and hot power piston will isothermalise/reheat working gas, contrary to parasitic cooling in cold connected power piston.

[matt brown]

Yes, there are various issues with both cold and hot PP, but the out-of-phase gas dynamics is probably greater (adds to all issues). Most DIY schemes favor a 300-600k cycle, so cold vs hot DPvol/PPvol = 2. Even within this cycle limit, any hot PP 600k engine will likely radiate more heat than 2x cold DP vs hot DP. And for the same output, a larger heater offers more surface area.

My main concern is regen issues where cold PP regen is 2x hot PP regen for this 300-600k cycle. Meanwhile, this graphic eludes Q vs W where DP volume is considered as 2 volumes during heating. In this manner, cold PP only heats 1/2 DP gas volume while forcing other 1/2 of gas volume thru regen (Cv) to PP. (yikes Fool, how about that for slight of hand - LOL)

cold vs hot PP Q vs W.png (27.46 KiB) Viewed 3787 times

[staska]

Good discussion.

I have attached some of mine more complex diagrams - calculating equivalent alpha striling engines. Which for our 300 / 600 K asks for 130-140 degree of phase and 0.7-0.8 volume ratios cold / hot. Found in Allan books, and more than one practical examples.

Cold pp - 153 degree of phase angle a lot of dead space and too big cold space. And wrong volume ratio - cold is bigger.

Hot pp (I did not changet names, but hot is cold and cold is hot and one need to divide resulting 1/ resulting volume ratio) - We are in place both with phase angle and volume ratios.

AG - alpha gamma by taking fatter rod we can make it so called alpha-gamma type, which will have only one hot cap, but diameters of piston will be a little bigger. But resultant geomytry is even better - slightly less addiotional dead space.

Will run sum nombers in mathcad to show power of each of two variants of similar alpha engines.

[staska]

Modeled numbers are below. Cold pp have wrong gas in wrong place, and need twice amount of hx sizes to have similar result compared to more compact hot pp.