60L drum medium temp Gamma build

[VincentG]

This will be my first portable large scale hot air engine build. It will be based on a common 15.8 us gallon/60l steel drum and use readily available parts and building techniques. I'll use what I've learned from my LTD engine testing and see if it translates to scale and higher temperatures. The plan for now is to finish the drum based displacer chamber and do some performance testing to determine how big the power cylinder should/can be.

The drum.

60l drum.jpg (217.17 KiB) Viewed 18966 times

As always the goal will be thermal isolation of all components. These 3/4 pipe nipples will be welded to the bottom of the drum and allow input to a heat exchanger located inside the drum. The only contact points with the drum will be the weld joints. Initially I plan to run a small engine exhaust though the exchanger to reclaim wasted exhaust heat.

60l drum heat pipes.jpg (255.75 KiB) Viewed 18966 times

The bottom of the drum will have around 3.5" of concrete poured in for stiffness and thermal separation. I'll wrap the pipe nipples in exhaust wrap and foil to serve as a thermal barrier to the concrete. The 1/4" bolts are there to secure the concrete to the drum.

60l drum bottom.jpg (257.41 KiB) Viewed 18966 times

I'm hopeful for very useful power out of this engine.

[VincentG]

Had some time in the shop to think about a lid retention system. These drums are available with v-band tops, which would work great except they may not be available for odd containers that people may use(I'm thinking large propane tank). More importantly, it might be a good idea to retain the lid with a spring mechanism so it may "blow off" in the case of an overpressure event.

Some quick math says the lid could be subject to over 600 pounds of force. These drums are thinner than I thought(yup easy to burn through with welder) so a good contact patch is needed for any head bolt brackets. I happened to have these laying around. Six automotive valve springs clamping down should provide enough(adjustable) tension.

I'm estimating this will end up with around 15 liters of swept volume, and nearly no dead space. At this point in design, I think a full displacer chamber can be built with parts and tools from just home depot and autozone.

Next up will be designing the top lid and a convenient way to activate the displacer through it. I'll save pouring concrete for the very last step to keep it light for now.

60l drum lid retainer 2.jpg (186.72 KiB) Viewed 18913 times

[attachment=1]60l drum lid retainer 1.jpg[/attachment .

[matt brown]

What specs do you have in mind? Do you have a copy of "An Introduction to LTD Stirling Engines" by James Senft? (I can't find my copy or I'd check it for ideas)

Offhand, I still favor the major volumes inverse the thermal ratio where DPvol/PPvol is inverse the thermal ratio. IOW if DPvol/PPvol=10 then thermal ratio=1.10 or in simple english, when

DPvol=1000cc with PPvol=100cc, then T high=330k and T low=300k, or
DPvol=1000cc with PPvol=200cc, then T high=360k and T low=300k, etc

The time honored safety valve for these low end DIY builds is a hot water pressure relief valve (they come in various values). I like the size !!!

Are you using any regenerator or going for more of an Otto scheme?

[VincentG]

Don't have the book. I'm going for a 300k-600k cycle but I want to optimize the chamber and see what it can actually do before I settle on a power piston displacement. Then I can make a adjustable power piston stroke to get in the sweet spot.

No plans for a regenerator. I'm not married to a particular cycle either. I'd like to see whatever can get the most pressure swing that resembles a square wave pattern as opposed to sinusoidal pressure fluctuations with no concern for "efficiency" as its normally considered. I think of efficiency in terms of utilizing the available temperature delta to get the most pressure swing.

[matt brown]

Vincent 60L.png (5.94 KiB) Viewed 18866 times

No regen, xlnt !!! My gamma study is bottled necked by 4 issues: (1) needs to be closed cycle (2) needs regen (3) low volume ratio/s (4) gas mass that goes thru regen is TWICE the gas mas that goes to PP (at least with my 300-600k reference thermal cycle).

The last point passes by under the radar, but is why comm'l SE have regen heat 5-6x ideal input heat...and this is why regen is so heavily studied. My long time view is that simply increasing the relative volume ratio reduces regen loss, but this is harder to achieve than at a casual glance.

I suspect Vincent is headed towards above graphic...a simple open cycle Otto where a suitable 2 stroke ICE is repurposed as PP and 'displacer' obscures heater during expansion and compression. My graphic shows displacer plate with upper and lower 'lip' but only the upper lip is needed (I was going to complicate this post regarding lower lip, but not worth it).

Anyway, the eagle eyed will notice that after expansion, the pressure differential (bottom 'hot' side vs top 'cold' side) will resist neutral force on typical 'displacer'. True, and this remains the challenge. This could be resolved by a rotary displacer in drastically different format or a bash valve could be added to piston head (piercing displacer at stg 4).

I'm always baffled by everyone's ICE love affair until they put on their ECE hat whereupon anything Otto is the evil empire...

[matt brown]

Otto bash valve.png (6.86 KiB) Viewed 18851 times

BTW the bash valve thru displacer that I suggested might appear exactly like this, but there's one major difference...this bash valve setup 'drains down' the reservoir since it's feeding working area, and will only work for a very short time. This is an xlnt example of where tracking only PVT values is inadequate and requires adding mass value to mix to see what's really happening. This should go without saying, but I'll add it...if this was possible, none of us would be jabbering about ECE !!!

The bash valve suggestion I made previously doesn't have this immediate issue, but still requires mass balance to sustain operation. The difference is very similar to intensive vs extensive. My previous crude graphic is not to scale, and there's still the issue of PP cylinder volume vs partial displacer cylinder volume (how about 'transfer volume' for this temporarily isolated volume). If the transfer and PP volumes are 'fixed' then the expansion ratio is inverse the compression ratio akin ICE. This means that the relatively massive heater volume of the displacer cylinder is sweet, but second fiddle to the transfer and PP volumes, since the efficiency of an Otto cycle is derived from the thermal ratio of the adiabatic compression process (not the thermal ratio of the entire cycle). Within same temperature extremes, any Otto cycle has more efficiency with a higher compression ratio. So, the gimmick is to strive for as high of compression ratio as possible (sound familiar) and allow enough 'overhead' for input (where overhead is the difference between temperature after compression and gas temperature after input vs source temperature).

[VincentG]

Good call Matt, and your graphics are getting good these days.

I don't have all the details worked out yet, but here's what I'm thinking for now.

The displacer will be primarily made from stacked layers of xps home insulation foam, with a heat resistant lower surface. The entirety of the inside of the drum will be lined with foam insulation and heat reflective foil of some type, such that the only other exposed metal surfaces will be the hot and cold heat exchangers. The exchangers will be obscured by the bowl shaped displacer as in your graphic.

The cycle will almost certainly be open. 15 liters is a lot of air to expand, so I'll need multiple power cylinders to get 7.5 liters of displacement. I'll keep that option open but first I'd like to explore one power cylinder(around 2 liters) with piston port exhaust whereby the rest of the expanding gas is directed to power turbine(to utilize mass flow) and then moves to heat my shop perhaps. Then after the hot exchanger is obscured, a one way valve ingests a fresh charge of air. Hopefully a tuned intake port will allow a charge density over atmospheric to start the next cycle.

As your graphic suggests, a cold exchanger may not be needed for open cycle if enough cooling results from the fresh air intake.

[matt brown]

My graphic shows displacer plate with upper and lower 'lip' but only the upper lip is needed (I was going to complicate this post regarding lower lip, but not worth it).

How ironic that I left lower lip on displacer only to have you complicate scheme LOL. My graphics are somewhere between Andy Warhol and Peter Max (limited by Excel and no time to learn Gimp). During my slow graphic process, I saw the Otto pitch and went for it due to its simplicity. Now, I'll have to rethink your scheme. Kinda amusing how two guys can look at the same thing and see two totally different things...

[VincentG]

Kinda amusing how two guys can look at the same thing and see two totally different things...

Ha, now I realize you were going for a hot reservoir scheme. I may give that a shot but I'm going for the heating element being totally shrouded when the displacer is down, allowing for chamber to fully cool down and draw in a full fresh charge before compression stroke. Or in the case of no pistons at all, the chamber will intake and exhaust its full capacity when heated and cooled.

[VincentG]

Today at work I had to service air brake chambers on an old international and it got me thinking. Why not use service brake chamber for a super simple piston design. Some quick math and the available torque potential is huge

These type 36 chambers are massive at around 2.5 liters of displacement each. Roughly 8" bore and 3" stroke. Once the return sping is removed they cycle very easily and only get better with break in. Heat may be a problem but the rubber diaphragms are very durable and cheap to replace, not to mention available everywhere and no lube requirements.

20230823_191026.jpg (57.48 KiB) Viewed 18746 times

Some modifications would be needed to allow a powered return stroke, as the diaphragms only push on a large steel washer and cannot pull it back. But that could be fixed by drilling a hole in the diaphragm and bolting it to the steel pushrod. With further modifications they could be made double acting as well.

Or if the return spring is left in they could be used as free piston generators.

[Tom Booth]

I've never seen the inside of one of those but from what you describe it sounds like you have a plan.

[Bumpkin]

I’ve thought about brake canisters for years since being an old trucker. Cheap easy durable. They seem overbuilt for atmospheric pressure engines, but they could take nearly X 10 reliably. Not my thing, but just saying…

Bumpkin

[VincentG]

Yea they seem like a perfect option for medium to low temp engines. Zero air leakage is a major benefit, and the ease of carrying spare diaphragms. I could easily see a trio of these type 36 air chambers powering a small car with steam engine like torque.

[Fool]

Hook it to the cold side and heat won't be a problem. You may need to enlarge the air inlet port for faster less restricted gas transfer.

[VincentG]

Yup, I was going to tap into the drum closer to the cold side. For sure the brake chamber ports seem on the small side, easy enough to weld on bigger npt bungs.