LocostUSA.com

Ok, I apparently have some very good luck. Out of the blue, I located a cache of Corvair blueprints, full blueprints for every piece of the engine, fully detailed. Now, the Corvair is not an ideal engine, being 28" wide x 28" long, but does have some other advantages which would make it easy to produce.

For those who do not know, the Corvair was an air-cooled engine. I am making good headway, thanks to having the full blueprints, and have some clean files already. I'm going to reproduce it as/is, then work on making it water-cooled as an upgrade if that is ok with everyone... or keeping it air-cooled and taking advantage of the fact it would be sticking out the sides of the engine compartment, dunno yet. One thing I had thought of, and still researching, is that GM was working on a modular setup for the engine, allowing it to scale from 2 to 10 cylinders. A 4-cylinder engine would be just the right size for a Locost is my estimate.

As a general rule of thumb I don't like air cooled engines.

Conventional wisdom back in the day said you couldn't make reliable aircooled engines once you began to bump the HP. They said you needed the consistency the water cooling brought to the table, but you know air cooled motorcycle engines seem to have proven them wrong.

And can't you get some pretty reliable and reasonably high powered air cooled VW engines nowadays?

Is it just more precision work now?

The simplicity of an aircooled engine I think would be a plus to a Locost since we aren't really all that worried about heaters.

Well, the Corvair was available at up to 180 hp, and I've seen other air cooled engines in excess of 1000 hp. Air cooled does not necessarily mean no active cooling system is used. There are multiple solutions, from the corvairs top mounted fan to high speed ducted turbines. If you want a comparison, your average 100hp car outputs roughly 4,244 BTUs per minute, while my workstation computer here outputs roughly 4,024 BTU's per minute, a comparable amount, and my workstation is air-cooled.

I think your math is a bit off in 2 respects. While a 100 hp gas engine may deliver 4,244 btu's per minute, roughly 4-5 times that is the heat it must dissipate. That is assuming ~17-20% efficient engine.

2ndly 4,244 btu's per minute is about 75KW. That is 75,000 watts. on 120VAC that would draw 625 amps. That would be one heck of a workstation!

I think that would trip a breaker... At ConEd!

Nah. I once did some computer work in a lab where they had a meter on the wall. It went from 1-10 AC MegaWatts! They had a pretty hefty power line coming into the building...

They did call the power company when they turned their stuff on and off though...

Well, you never asked what my workstation was....

Next time you watch Jurrasic Park, pay attention to the red system that the girl is using to turn on the electric fence. I have one of those. And no, it is not 120V. A single processor in it is rated at 20 amp.

But that is getting off topic. It just happened to be the system I knew the heat output on, as it is listed on the panel on the back. Incredible machine, even if obsolete beyond measure today. A grand whopping 80 Mhz processor.

I checked and you are right. In the corvair engine, for example, it would be producing 24,964 BTU/minute.

In any case, the engines ran with 80% higher horsepower without issue.

I realize, I keep flipping around ideas, and as a result its hard to finish, so I am nailing down the design, and simplifying things.

Yes, I like Boxer engines, but due to space constraints in a Locost, boxers would be more time to design in order to fit right. Let's stick to an inline engine, just to get a working design done. Although a Wankel would be simpler to build, the sealing issues would be great.

I like the Crosley gear-driven overhead cam, so I think I will stick to that.

I've run across a tractor cylinder wet liner which is very inexpensive, 86mm ID x 171.5mm cylinder height, which I think would be a good baseline to work with. I've checked, and pistons are easy to come across.

Combustion chamber design is proving an interesting challenge, learning compression ratios, burn rates, etc. I know a lot of us out there have emissions standards to stick to, so have been studying different solutions used by various groups over the year, and hit a few ideas which I'll be incorporating.

The one real issue of the low-cost wet liner I found was that it would tend to push an Undersquare design, not ideal for a performance vehicle. I've found ones better for oversquare, but the cost does increase by a good amount.

Alternatively, the cost of motorcycle wet liners for Kawasaki 900cc engines are not much more than the tractor liners, but do suffer from smaller total size. I can increase the total displacement to ~250cc per cylinder with the right design, but this is still a small engine compared to the tractor cylinder options which would give a range of ~600-700cc depending on setup.

So, which would you all rather have, a larger cc with a lower rpm but higher low-end torque, or smaller cc with far higher rpm?

Porsche produced very reliable, simple, lightweight but powerful air cooled engines into the 90's but cold start-up emissions eventually forced them to tighter bore clearances and ring packaging which required water cooling.

If I built an engine tomorrow using existing parts I would maybe build a flat 8 with a 4 cylinder crankshaft using master and slave conrods as a radial engine uses.

The boxer design I was working on used just such an arrangement. I actually have that design mostly done, in both an air and water cooled arrangement, although I was aiming for a 4-cylinder due to space demands under the Locost hood.

Sometimes the obvious solution you cannot see. An 8 cylinder boxer case can be a 4 cylinder inline case easily. One design, two applications. I will admit to thinking of the smaller displacement liners in a 6 cylinder inline layout, but could do the bigger four instead.

If you want to make something that is better for a locost then what is available that is a pretty high bar these days. That said we do have different needs. Lots of horsepower per displacement is pretty useless as a primary goal for us, but it is good for advertising so that's a lot of what is available in the junk yards or donors. For us small physical size and light weight are the big primary factors. So I would expect pushrods and relatively large displacement. A typical airplane engine can be 6 liters and under 300 pounds...

Another approach would be to get something fun. So a small V8 of 2+ liters. Again pushrods for size.

I remember a Lotus designed engine for a light airplane from years ago. It had one piece cylinder and head to avoid head gaskets. That would help for boost and might be simpler. You could also use a combustion chamber in piston design with that like the Ford Kent engines. Pistons for those are easy to come by and would produce simple machining, one of the goals.

I think pollution requirements are a non starter. You could pass the tests, maybe. Those tests are to check the engine is sort of complying with the certification though. That certification is a many million dollar prospect, I would imagine.

Just for fun here's a picture of a downsized Cosworth V8 for an FSAE car. It's a small picture, sorry.

The manufacturers requirements to obtain smog certification aren't too bad in California despite what you may be told, they are however rather complicated (i.e. time consuming) and have limits such as;
Engine testing must take place over several days with duration specific engine off periods, the exhaust gases is bagged before testing, etc, etc.
It's the time element that's expensive, ~1k hours as I recall to meet every test requirement if you don't have to do any redesign. A redesign would require you to start at the beginning again. ......
I am unable to lay my hands on my C.A.R.B. issued handbook at the moment but I'll keep looking.
It goes into detail and even includes line drawings of an approved exhaust gas test system.