Working Lego 4 stroke Gas Engine + pics

[Aerolight]

lego engine 2 by michael waterfield, on Flickr

lego engine 1 by michael waterfield, on Flickr

lego piston head by michael waterfield, on Flickr

About 2 months ago I started building said engine, single cylinder 4x4 bore 5 stud stroke (no glue), for the ignition I used a system made for model engines with magnetic based timing and powered it via the Lego rechargeable battery / wire from a dead motor. The first incarnation used gate style valves but they where sticky and lost a lot of pressure, so I switched to using technic caped axles through lego system which is amazingly air tight and looks/works like a real engine (4 valve total arranged in a flat head configuration). The valves originally had variable timing via two diffs but I later removed it not due to the resistance but retiming the engine every time got tedious fast (the ignition magnet was driven by the same gear train as the valves (1/2 from the crank, 12 tooth too 24). Now for interesting part the running, the area behind the valves is in a open top box and its this I spray the fuel into. O don't use solvent of any kind, with gas type lighter fluid lego holds up with no sigh of damage after a few minutes of inconsistent firing, but I ran out and due to crippling stupidity/laziness I just grabbed the next most flammable thing I had - a solvent spay. After less than 10 seconds firing on the starter motor (two xl gear up x5 about, 42 8) the engine slowed and when I opened it up both the exit valves had their caps melted and there was some waviness to the piston top. After replacing the axles and getting some more lighter spray I changed the flywheel from a geared up Unimog wheel to a aluminium pully about 8 cm across and 2cm deep. With this configuration the engine runs for a few seconds (yes without the starter connected) a few seconds after fuel is added no matter what I try I cant get the engine to run without these gaps in ignition, It seams like there is too much fuel and after a few cycles the fuels concentration becomes low enough for combustion, this is where I believe the most improvement can be made and yes I mean a working lego carburettor, or at least something better that my box method. While running the engine sounds like some of the first ic engines made, probably due to the bad burn and merely non zero compression. . Now I imagine most of you are curious how I got useable pressure containment and no hang-ups with a lego cylinder/piston without glue, gaskets or any modification to the lego , well unfortunately that is hard to explain even though it is ldr friendly its a bit unorthodox, but you can see in the pics. But during building the cylinder I put the piston in and blew in the other end and the piston went 4 feet across the room consistently. I plan to make a video as soon as I get more than a few seconds of ignition out of it (first running lego engine hello youtube gold). O and finally I made it four stoke due to it being easier to run at low rpms and my ptsd from nitro cars has yet to leave me. Any advice/interest? (hopefully I added pics correctly)   

Edited August 4, 2019 by Aerolight
add pics

[deraven]

That does sound pretty interesting.  I'd like to see some photos and video of it in action.  I'm sure there's other interest here as well, but you'll probably get more traction to your topic if you edit the text a bit (so it's not just a wall of text that some folks might not read), and of course photos will catch peoples' eye, too.  Thanks for sharing, and welcome to EB!

[Mechbuilds]

I would love to see a lego creation with a metal gas powered engine that's heat wrapped to prevent damage to the lego parts around it. 

[Aerolight]

finally added picks via flicker hope you find the fact its not a pool of rainbow collared goo impressive 

[TeamThrifty]

This is great! I've got to see it running... video? Also details of the top-end, valves, how you're timing it etc I love working on engines, and i love lego.. this project is right up my street!

Also, i presume you chose the fuel based on its burn temp? I may have a 100 questions, sorry in advance..

 

[Aerolight]

The top end / head is built stud down so the roof of the cylinder is smooth (same build as top of piston), then the head and case are clamped together from the outside with technic. The valves are driven via a vertical swing bar with a wheel that contacts with a regular technic cam driven off a gear train just above the crank, valve timing is adjusted via removing intermediate gears and the ignition timing is adjusted by turning its magnet arm (red tape) (its driven via a white friction gear). As for fuel I just tried what I had and found that aside from the solvent, and a pool of lighter fluid, the temperature was completely reasonable. After running for a few minutes (with a lot of help from the starter) the lego was far from melting so for the moment my moneys on simple gas fire lighter. But the main problem with fuel is not temp but the range of air/fuel ratios where combustion can occur due to the lack of any real carburation, when I tried propane it would fire about once every 30 seconds opposed to the gas fire lighter which is very close to self sustaining. Although if the engine can self sustain I would think it would not run for long, a few minutes perhaps, but due to the exhaust valves being a point of weakness the engine will stop before a real damage is done to the block.

When I used the differential timing method (used on some GBCs) it shook its timing out and the timing became the same as an antilag cycle, the sound this made almost put me on the floor with laughter, imagine a little stationary engine hitting ak47 (potato gun) tune. When I make a video I will be sure to add this       "error". Thanks for the interest.

[Aerolight]

Update

I have got three new ways to regulate the air/fuel mixture now;

First is like a normal carburettor except the air is not drawn over liquid fuel but over gaseous fuel (due to lack of suction) and the vellum will be the point of adjustment.

Second is to have the inlet drawing from two separate places, one with air one with air and fuel, and using adjustable valves regulate the mixture.

Third is to use pneumatic pumps to supply the fuel and air instead of just relying on the suction from the engine. The pumps need surprisingly little energy to pump its the switch that sprays into the inlet that's probably the problem due to the rpm, however a mix of method 2 and 3 might be the best chance due to the increase, as slight as it is, of air/fuel flow and not requiring a pneumatic switch.

The best mix is only 5-10% fuel encase you are wondering.

Hopefully the next test will be in less than a week from now, I promised a friend I would only start it with him and he is away working this week. As I have said before I want to wait to upload a video until the engine can run for more than a few seconds, hopefully a improvement in the air/fuel mix is enough to make this happen. I am hopeful enough that I have built a chassis that the engine can sit in to drive around so stay tuned. 

 

 

[Ivorrr]

Still waiting for that video!

[Aerolight]

26th of august is the go date then video slightly after that - I need my uncle who's been away working for summer (hence bank holiday Monday)to helm the jug of water as the rest of my family sees this as a waste of water. Though while waiting I have made some improvements; there is now a prototype gasket between the head and the case and a workingish rpm gauge though it only really shows rotating and not rotating rather than anything helpful. But the best thing is a working "supercharger", I wanted a turbo for #spoollife but there is just too much pressure leakage so a ducted fan driven off the crank (into the carb)is the best I can do, there's not boost per say but rather more air flow so it may not help but its worth a shot. Not sure if its allowed but I took a modelling knife to the capped technic pins I'm using for valves and whittled them from a cross to a small rod (like a real valve) for better flow, actual porting!

[Aerolight]

Alas twas all for naught, I feel my time working on this has ended at least for now. 

After trying everything I could in regards to air fuel mixture (namely using easy start as is burns at the greatest range of concentrations) I finally got the engine to fire almost every cycle. This should have been more than enough but the engine just simply lacks compression, while a cylinder like this seals well enough for use with a vacuum cleaner (ominous)  the way a gas expansion produces force requires good compression above ease of movement; which this motor was built around as good compression with Lego is almost impossible perhaps even with glue.

As for the video I got my uncle to try and take one but the 'pop' sound was the only thing worth seeing (hearing) and our phones just would not pick up the sound over the valve gear so I cant even give you that (I know some of you would still like to see a video of it being turned over but without the pops it just falls flat) although now I wish I had at least taken a small clip with the head off just because after applying silicon lubricant to everything the piston was so smooth it was actually surprising and the method for the cylinder walls that enabled this was probably the best thing to come from this to be honest.

The engine as it is now is dead in the water, a fire started inside the airbox and I only saw it when smoke came out from the gaps. This warped practically all of the Lego in and around the airbox, not that much but enough to lock the valves when connected tightly. Its only a matter of time before the cylinder is affected and bricks any plans for something in the future.

So no more tests at least for now, I could use glue to get a good seal between Legos but I don't want to do that until I know I can fix the other problems (the piston seal/rings and fuel atomisation). I will be away from my Lego for a while now (yet again) though, but hopefully my next MOC will be somewhat more functional (perhaps a working Lego micrometre with a 20 number planet style digital display that's surprisingly accurate)

oh and don't try this without a complete disregard to your personal safety and that of your Lego. 

 

 

[Mechbuilds]

Interesting study. I'm surprised it even popped. Well done!