v8 Alpha stirling generator concept

[johnmaster]

I am still utterly confused as to why there aren't simple affordable designs being made and sold to provide a modest amount of power from waste heat. An ebay search turned up nothing special, lots of models, some crude imported small scale versions but nothing in the way of tried and true mid scale designs. Am I missing something obvious? I realize the amount of power you can extract is low for the size of the engine but if you use commonly available used parts and some recycling I can see it becoming worthwhile and not terribly expensive (maybe).

With that I had the idea to build an alpha out of a v8 automotive engine. Basically 4 alpha engines in parallel, heat one bank, cool the other. Some work would be required in cutting friction, perhaps polish the crankshaft for loose tolerances, remove camshaft, remove some cylinder rings for less drag, perhaps even releive the piston skirts. The heads would be removed and replaced with aluminum heat exchangers tailored for the purpose. 4 crossover tubes would connect the opposing cylinders.

The dc alternator and belt are already mounted with brackets, the starter motor could be used on a manual transmission flywheel and clutchplate to create a heavy flywheel and a starting mechanism.

So my questions are basically-
Can you see any obvious shortcomings that I might not have thought of? I am very handy so machining and fabrication come second nature to me.
Is 90 degree optimal, or even close?
If I can should I pressurize the crankcase and cylinders? What gas, Helium?

Thank you for any replies, would love to get some feedback on what I could expect to produce with a beast like this. Hoping to max the alternator (1000W)?

[Ian S C]

johnmaster, first problem, an IC engine is built too heavy, mostly designed for between 100 and 300hp, you are looking for just over 1hp.
Yes 90* is a good starting point for angles. For the bearings on a crankshaft for a motor like that,1" would be more than addequate.
The cold cylinder could do with better cooling, and it will require new cylinder heads.
The automotive pistons are no use, because you won't be using rings.
One member of the forum was converting a V twin compressor, but we never did hear about the result of that. Ian S C

[Geoff V]

Johnmaster

IanSC raises some valid points regarding bearing and piston ring friction, but there are several other fundamental problems.

Firstly, the cylinders on an ICE are rather too close together to permit coaxial heat exchangers, so it would require tubular exchangers where the tubes have to sit either side of the adjacent cylinder which limits the amount of heating area available.

Secondly, to produce even 250 watts per cylinder would require some pressurisation and the crankcase assembly is not really appropriate (unpressurised you would expect 0.1-0.3 watts per cc).

Thirdly, oil lubrication is required for plain bearings and preventing oil from entering the cyliders and fouling the heat exchangers and regenerator would be very difficult to achieve without excessive drag from an oil control ring, not to mention the risk of explosion if the engine is pressurised.

Furthermore, the extension required to the expansion piston would be unsupported except for the original piston skirt which might require excessive clearance to avoid rubbing.

Whereas I sympathise with your desire to use a ready made assembly, in light of the above you may well decide that there is less work involved starting with a clean sheet of paper.

IMO the most promising arrangement for this sort of power output is the JIm Dandy #6.

http://www.starspin.com/stirlings/jimd6.html

GeoffV

[johnmaster]

Thank you guys for your input, glad to have this forum to bounce ideas around. Friction is certainly a major consideration with turning something that can tolerate over 1000 hp (turbocharged) into a machine do a 1 hp job. Was thinking of perhaps polishing the journals to open the tolerances for the first go round and find some way to lower the oil pressure so it only trickles in. If it worked ok I could think about making a plate and pin crankshaft to replace the cast iron one and incorporate roller bearings to get rid of some of the parasitic drag from the oil pump and journal bearings. Probably could get away with 2 or 3 roller main bearings instead of 5 journal bearings for instance.

What would be different about the pistons other than taking all or most of the rings off? I was thinking of just reusing one of the oil control tension rings to see if it kept the majority of the oil out. i had planned on making piston extensions and bolting them to the tops of the pistons on both sides. Then looking for some finned air cooled 2 cycle cylinders to bolt on top of the aluminum plate cylinder heads. I could wrap the cold side in a liquid jacket and have a large recycled radiator to really expel the heat. The hot side could either be in the furnace exhaust stream or wrapped in a water jacket though the best temp I could probably get is 180-190 out of liquid before boiling issues arise. The furnace isn't built yet so I can modify it to suit the motor if direct exhaust is the best heat source.

The crankcase is modern and the intake is divorced from the valley cover so I think I should have no problem putting an atmosphere or 2 on it without major leakage. The flat plate cylinder heads would block the old pushrod path. I would use an inert gas to prevent pressurized explosion. Helium a good choice?

Just like most of my ideas it will turn out to be more work than it would be worth to do over again but there is only one way to find out right? Any other input would be helpful. Especially on piston clearances and crossover sizing, regenerator materials etc.

[johnmaster]

Also, if there is no crankcase lube, how are you oiling the cylinders? dry metal on dry metal sounds like a friction issue. What should clearances be on something like this?

[Geoff V]

Johnmaster

Dry running SE's use a modified PTFE such as Rulon or Crossflon for the piston bearer rings and the sealing rings, the rest of the bearings are usually low friction sealed ball or roller races.

Regarding your proposed temperatures, an alpha configuration with the same size bore and stroke will require the heater to be at least bright red say, 650C and the cooler around 50-70C, any lower temperature difference and the engine won't run. Depending on how good your heat exchanger are, you may also have to sleeve down the compression cylinders as it will probably need a displacement ratio of about 1.5:1.

Regarding pressurisation, the sump will need to be strengthened as it will see a force of about 1.5 tons at 2 barg and how do you plan to seal the crankshaft?

Sorry to appear so negative but these questions have to be addressed if you're going to avoid disappointment.

GeoffV

[johnmaster]

Thats why I am here, glad for the input and ability to bounce ideas off knowledgeable people before I start putting any money into this project. Don't worry about sounding negative or raining on my parade, better now than when I am 50 hours into this thing and find out I screwed up something simple that I could have learned just by asking.

There is going to be a limit to how much pressure i can put on the crankcase, it has simple lip seals on the crankshaft but the pan is pretty robust cast aluminum. I am more worried about the flat valley plate and timing cover. It will probably be a trial and error on if or how much pressure to add. figuring more like 1 barg (2 absolute) as a realistic expectation. The gaskets are silicone/steel and very strong looking-modern.

Tell me more about 1.5:1 ratio of hot and cold side pistons. Which side larger and why? The heat exchanger design is still up in the air but I was thinking about making piston extensions for both sides to greatly increase the surface area for heat exchange in the walls of each cylinder and to allow for lots of fins to move heat efficiently.

Making cylinder sleeves and new pistons might kill all ambition, otherwise so far I am still thinking it is do-able.
It was said "unpressurised you would expect 0.1-0.3 watts per cc" and .2 is 960w for a 4.8L so we are right on target even if unpressurised (assuming volume is total of all cylinders displacement and not just half). Is that net at the flywheel power?

[Ian S C]

With an Alpha motor the ratio between the cylinders is 1:1, with Gamma and Beta motors the ratio is near 1:1.5 power cyl: displacer cyl.
Geoff, I notice that Jim Symanski uses a !;! ratio for all his motors. His Gama #3 looks interesting at 500W shaft power. Ian S C

[Geoff V]

The question of displacement ratio is dictated by the ability of the heat exchangers to raise and lower the mean temperature of the working gas. If the mean WG temperature is too low, the WG will not have expanded enough to maintain a pressure difference across the piston and there will be no driving force (torque), this is regardless of whether it is a beta, gamma or an alpha. My last engine, an alpha, failed to start unless very hot (650c) and 3barg charge pressure, I reduced the bore of the compression cylinder even though it has slotted heat exchangers with 90 x 0.5 x 5mm slots, it now runs well and produces nearly 1watt per cc when pressurised with a displacement ratio of 1.56:1.

Unfortunately, most text books are misleading by advocating a DR of unity, this is probably due to the authors assuming all engines have the benefit of very sophisticated heat exchangers and a lot of charge pressure, in the world we operate in, as amateurs, nothing could be further from the truth.

John

I assume you are using a 4.7Ltr engine? Unfortunately an alpha, because of the phase angle, has a working volume some what less than the piston area x the stroke. When the displacer piston is at TDC the compression piston is half way up its stroke, so as the displacer descends there is displacement but no change in volume, hence the total working volume is less. We need to calculate the maximum and the minimum volumes (45deg either side of TDC and BDC) and take the difference to get the working volume. With the cylinders being close together and the heat exchangers being far from ideal and the probable need to sleeve the compression cylinders and reduce their volume you will have difficulty in realising your 1000watt target.

Your engine at a quarter volume, 1175cc, times 0.707 (sin45) equals 830cc times DR (1.5:1) gives 553cc working volume per cylinder pair, due to limited room for heat exchangers you'll be lucky to see 0.1watts/cc, so, 553 x 0.1 = 55.3watts x four cylinder pairs = 221watts total. To reach 1Kw total will require in excess of 5barg charge pressure, less if you use Helium, but with all the sealing issues you have, He will readily leak out and cost a fortune to keep replenishing.

GeoffV

[Geoff V]

Ian

I forgot to answer you question regarding the Jim Dandy#3.

With a working volume of 412cc and 500w at 4barg, this would equate to about 125w at atmospheric ( or a little more, due lower friction) which gives 0.303watts/cc, for a purpose built SE this sits nicely at the top end of my 'rule of thumb' of 0.1/0.3 watts per cc.

GeoffV

[Ian S C]

geoff, I imagine that Jim might be a bit optimistic when he quotes 500W at the shaft, and 250W + at the generator, also 60psi pressurisation. Interesting all the same. Ian S C

[johnmaster]

I have options for 4.8L, 5.3L and 6.0L (the LS family of GM truck engines) the 4.8L being the cheapest. It would be easier for me to make one side more or less efficient than to make sleeves and new pistons. For instance I could have longer or shorter piston extensions and exchangers to offset the difference required if need be. Is this the reason I see piston extensions on the hot side and usually a flat head on the cold side, are the cylinders then basically the same size because the hot side has so much more surface area than the cold side?

I could liquid cool the cold side with 0deg c water in the winter...I can't think of a more efficient cold side?

I am picturing manifolding the heat laterally so that it flows across the cylinders and not down the bank, this should make the spacing negligible, the cylinders would share heat and the cold side would be divorced and insulated on the other side of the block...

Thank you for the calc on displacement, I assumed I wouldn't be able to use the whole displacement but would have been nice if I could.

[Geoff V]

John

Extracting power from the Stirling cycle is very difficult, requiring sophisticated heat exchangers and lots and lots of pressure, starting with an engine which was designed as an ICE precludes the high pressures and limits the choice of heat exchangers.

In terms of effort and cost I'm not sure there is much to choose between a V8 conversion and a purpose built Stirling cycle engine, so it seems to me that you need to decide on your final destination. Is it that you want a V8 ICE that runs on the Stirling cycle regardless of power output, or do you want a Stirling cycle engine that produces 1000w?

I wish you well with your deliberations.

GeoffV

[johnmaster]

I just estimated in my head that from an engine this size that it would be neat to get it to make 1000w, I dont need that much power specifically, it just happens that is about the max I figure I could probably squeeze out of the alternator that comes with a used engine. I guess my real goal would be an engine that makes a useful enough amount of power to justify the time spent building it, something that could run 24/7 for extended periods of time and eventually be worthwhile installing a grid tie inverter to offset the energy costs in my home.

It doesn't have to be a recycled v8 but I figured for 200 I have the basics there and wouldn't need to reinvent to many wheels, the belt/pulleys/generator/crankshaft/flywheel/pistons/rods/oiling system/cylinders/starting system etc etc. In my mind it's more worthwhile to modify it to do the job with trial and error than to try to start from scratch and make or buy all the peices.

What I had in mind for the heat exchangers for either or both sides were something like this but with ducting to force all the heat over the fins

Here is the style of block I had in mind for the basis of the project

What features of a good heat exchanger do the best stirling engines have? I am over the idea of hydronically heating the hot side as you suggest that the temps are way to low for this. I would put the hot side right into the exhaust flue (rocket stoves are known to quickly disentegrate stove pipe so high temps should not be a problem).

Where is the best place to download the free e-book on stirling engines?

[johnmaster]

After more reading and use of the search button I am seeing what you are saying is the issue at hand, surface area and ability of the heat exchangers to get the heat rapidly into the working fluid as well as the ability for it to pull it out rapidly on the cold side... Ok, so I guess I am starting to see the light. Need to do more researching and learning I guess:)