/diy/ - Do It Yourself

>>797663
Go on...

>>797663
I've always been intrigued by small displacement, high revving, inline engines.

Just look at this and say it isn't cool: https://www.youtube.com/watch?v=hkkPPTakc3s

So, I've always wanted to make, own, and especially USE a small engine, with the intention of it being used in a go kart.

Buuuuut, I've also wanted the consumables to be off the shelf, reliable parts, with dare I say it "go fast" parts available as well.

My ultimate goal is a water cooled four stroke, 12 valve, SOHC, 250cc I6 engine, with a turbocharger and blow through carburetor for said turbo.

But for now, I need to keep things simple; the plan above is for when I've got a full shop, dammit.

So: thus starts my endeavors into designing an aircooled 2 stroke, 200cc, inline six (I6) cylinder engine.

>>797663
I've always been intrigued by small displacement, high revving, inline engines.

Just look at this and say it isn't cool: https://www.youtube.com/watch?v=hkkPPTakc3s [Embed]

So, I've always wanted to make, own, and especially USE a small engine, with the intention of it being used in a go kart.

Buuuuut, I've also wanted the consumables to be off the shelf, reliable parts, with dare I say it "go fast" parts available as well.

My ultimate goal is a water cooled four stroke, 12 valve, SOHC, 250cc I6 engine, with a turbocharger and blow through carburetor for said turbo.

But for now, I need to keep things simple; the plan above is for when I've got a full shop, dammit.

So: thus starts my endeavors into designing an aircooled 2 stroke, 200cc, inline six (I6) cylinder engine.

>pic not related to anything in particular

>>797672
So, I'm highly basing this engine on the extremely popular and inexpensive 33cc/36mm bore 2-stroke engines used in chainsaws, weedeaters, motorized scooters, etc.

The parts I plan to use and design the engine around are these:

>36mm pistons with rings
>accompanying "head and cylinder combo" that 2 strokes have, like in OP pic
>spark plugs, duh: they thread into the head
>carburetors; they're inexpensive and have a lot of aftermarket support

I'm not exactly sure what I'm going to do for ignition, but I'll figure it out.

Is this a request for help or a blog entry tripcunt?

>>797678
So you're planning an I-6 with these components? Rods, caps, bearings, etc. etc... There's a lot of parts to source, even in a 2-stroke. Do you have the means to cast and machine your own block? Machine and balance a crankshaft?

You'll spend a fortune getting someone else to do it. Then making it work in a kart may not be the easiest thing in the world, depending on the characteristics of the finished motor...

>>797678
so, general block idea from the top.
Nothing more than a stretched 33cc block, basically.

>>797685
why the fuck are you signing your shitty image with your trip..?

>>797684
So you're planning an I-6 with these components?
wanting to design one, yes. Making it would be even better, but as I'm a full time student, would be a bit difficult to afford.

>rods
there are very small connecting rods from chainsaw engines the 29cc-35cc range that are two piece (I.E. can replace rod bearings, like pic related, Honda GXH50), and should work in this application with some custom bushings.
>cap
...do you mean cap and rotor, or main bearing cap?
Cap and rotor I'm not sure yet, and main bearing cap would have to be custom made.
>bearings
gonna have to be custom, unless I can find something else.


Yeah, there's a lot of parts to source, but I at least want to draw up a dimensionally correct plan so if someone wanted to, they could make one (that someone preferably being me, if funds and time allows). For having someone else do it: I'm at college majoring in Welding, and as such I have access to the machine shop and many friends who would be up for helping with this. I think I could have it covered if I simply had a plan.

And as for go karts, I know. This thing would make maybe six to eight horsepower, which is not much at all to be honest. I'd have better chances to just use it in a bike or something equally stupid...

Exceeeept, there is a tractor transmission called the Peerless 700-023. This a 4 speed (three forward, one reverse) manual transmission with clutch that can be mounted in an FR platform, that being engine in front and transmission in the rear, effectively turning the transmission into a transaxle. I could then fabricate a torque tube to get power from the engine to the trans, and then have a front engine, rear transmission car for dat 50/50 weight distribution.

And I could have a pony.

>>797682
request for critique and general advice. If I can find out what I need to work on, then I can find what off the shelf parts I can use, and once I find that out, I can start drafting blueprints.

>>797687
why not? It's fun.

>>797698
>>797684
>bearings
take a look at this article!
http://www.gasenginemagazine.com/gas-engines/babbitt-connecting-rod-bearings.aspx

>>797698
If you're thinking of mounting a 6-cylinder air-cooled motor longitudinally, you're going to have a bad time...

Have you thought about crankcase flow at all? You're going to have to. You'll probably just want to isolate each cylinder for simplicity's sake, which will cut in on friction losses.

There's a good reason most 2-strokes are 1-3 cylinders.

>>797663
>design an internal combustion engine
>build it myself
>out of off the shelf parts.

uhh, ok. you wont actually be designing anything if you just use off the shelf parts.

I dont think you understand how internal combustion engines work.
they need very tight tolerances to work. how will you get an off the shelf cylinder block?

>>797698
single cylinder 125cc honda motorbikes make 10HP.

I guess your engineering lab might have the tools to bore out a block. but a solid block of steel that size would cost a few hundred dollars. so you might need to cast it. but then you need to melt steel.

>>797678
>I'm not exactly sure what I'm going to do for ignition, but I'll figure it out.
unless you want to use a car distributor and coil and everything, youll probably have to get a custom 3 or 6 magnet flywheel and attach multiple coils around the flywheel to go to each cylinder and time them correctly

relevant
https://www.youtube.com/playlist?list=PLOkM2h167hjruSlogPv-U0xlOiEda42Jz
https://www.youtube.com/playlist?list=PLOkM2h167hjoHdkDu4dW1ujTyUSoU3BPE

>U0xlOiEda42Jz
>2Jz
it was meant to be op

>>797719
>but then you need to melt steel.
I have a foundry capable of melting mild steel, it's not a problem
>>797719
I'd be casting the block for sure
>>797718
I understand how engines work; my trip is based around the Datsun L24 engine.
And I WILL be designing an engine, since the block, crankshaft, main bearings, main bearing caps, and other things will have to be custom made. I will be designing those parts around the off the shelf parts.
>>797704
>crank case flow
Gonna have to think about that for sure

>>797922
Look at this for the block:
http://www.hotrod.com/how-to/engine/1412-how-to-build-your-own-sheetmetal-engine/
Might rethink casting, it's an expensive process.

You're probably going to have to fab up a water jacket anyways, anything aircooled longitudinal is going to have issues past three cilinders. Pre-war Nimbus motorcycles would get a super-hot rear cilinder for example.

Rethink the carbs, straight six with single carbs is going to be a PITA.

>>797672
Blowu through carbs on a turbo? Injectors are simpler and can make for bigger boost without knock.
Go DOHC if you're DIY'ing, better performance.

You haven't specified 2 stroke or 4 stroke yet, that's pretty damn critical.

>>797698
>front engine, rear transmission car for dat 50/50 weight distribution.
Doesn't work that way for karts.

Go look up automation, it's a fun game where you can play around with several engine prototypes before you start building.

>>797937
>Look at this for the block:
>http://www.hotrod.com/how-to/engine/1412-how-to-build-your-own-sheetmetal-engine/
>Might rethink casting, it's an expensive process.
If you've read the comment you linked to, you would have seen that I have a foundry.
>You're probably going to have to fab up a water jacket anyways
maybe, maybe not. I'll cross that bridge when I get to it. If I could case a block with an internal water jacket, I will. Still doing research on that concept
>Rethink the carbs, straight six with single carbs is going to be a PITA.
No, not really. Once you have them all working in tandem, they're pretty easy to work with.

>>797937
>Blowu through carbs on a turbo? Injectors are simpler and can make for bigger boost without knock.

any carburetor can be "blow through", I hope you realize this. And I'm not going with fuel injection, as that would require a head, valves, and THAT requires camshafts.
This also wouldn't have more than like 1-3 pounds of boost. Literally nothing in the large scheme of things. Also knock is not stopped by injectors at all, that has to do with compression ratio and your fuels octane rating.

>Go DOHC if you're DIY'ing, better performance.
You DO realize I would have to make a cylinder head, camshafts, valves, oil galleys in the head, timing gears, and a ton of other stuff, right? That was the whole point of a two stroke as a concept engine: no valves, no cams, no oil pumps, no nothing.


>You haven't specified 2 stroke or 4 stroke yet, that's pretty damn critical.
Read the damn thread, you dingus.

This thread is interesting to me. I'm a big car guy and recently I wrote the software for an engine management unit but shortly after lost my test engine in an unrelated incident. I've been toying with the idea of miniature engines for bench testing my ECU (it passes logic analyzer tests). I don't quite have your fab skills apparently so I was just going to go pick up an RC aircraft engine.

I'd offer to solve your ignition problems but my ECU is pretty pricey to build in small quantities. Pic related.

>>798140
that looks awesome to be honest; what exactly have you made there?

While I love the idea of EFI, I'm... well, to be honest, I know nothing about programming or electronics. There's a reason I own a 240Z: everything is mechanical.
I can sync six carburetors and draw a fairly detailed blueprint on paper, but I can't design or do SHIT with a computer. I own the student version of AutoCAD 2016, but have no idea how to use it. I downloaded it because I wanted to do this, but I don't know how to use it.

Buuuut as for your engine thing, why not go buy a 1NZ-FXE? The engine from the Prius. It's all aluminum, has four valves per cylinder, DOHC, VVT-i and uses the Attkison cycle, which would be interesting for you EFI wise.

They're also only like $110 from most wrecking yards, since nobody wants to do anything with them.

>>798149
>that looks awesome to be honest; what exactly have you made there?
ECU based off some pretty common hobby hardware. 4 ignition and 4 fuel channels capable of batch. It's shown real promise on pulse width accuracy. 2 Medium current GPIOs and 4 low current drives. Onboard wideband controller (swappable for 4 analog inputs in cases where one might want to run per cylinder WB) and onboard Cam and crank circuits for both VR and Hall. USB interface and SD card logging. It's basically everything I ever wanted in an ECU.

I get that. My junker car (and test engine) is a '60s bug. There's something nice about the old carbs and dizzy (although I'm not a points guy).

I didn't know the Prius engines were going that cheaply. I may have to hunt around before I pick up an engine. But I will say this for tiny, hobby engines, the test stand is easier to fab and they're closer to a clean slate than trying to fire up a modern behemoth as engine number one. The bug engine was a nice candidate because they put up with anything and keep running (besides low oil pressure lol).

>>798172
>The bug engine was a nice candidate because they put up with anything and keep running (besides low oil pressure lol).
go pickup another bug engine then: they're like $350 running on my local CL.

Now that I think of it, they would make ideal testbeds for designing EFI stuff.

>>797663
What if I told you.. there's already a way to splice together briggs engines?

what if I also told you, theres lots of race parts for said briggs engines?!

WHAT IF I TOLD YOU...
you're gonna need about $60,000 in machine equipment to do ANY of this?!!!?

>>797663
if OP made his engine bigger, how expensive/difficult would it be to get an engine number registered for it?

>>798203
>there's already a way to splice together briggs engines?
not reliably, at least that I know of.
>$60,000 in machine equipment
nope, not even close. I could easily do this with a $2,000 mill.

>>798206
>engine number registered for it
What is this engine number you speak of?

>>798184
>go pickup another bug engine then: they're like $350 running on my local CL.
For half that you can grab a brand new gas hobby engine. Hell you can get a two stroke glow for pocket money. I could get one of them to run sparking street gas I'd bet.

>>798215
'merrica doesnt have VIN or engine serial numbers?

>>798246
we have VIN's and engine serial numbers, but federally the engine serial number isn't required for anything. We don't register engines; that's almost as bad as registering guns.

For instance, in the state of Idaho: if I got this engine running and put it in a small go kart that had lights, a seat belt, brakes, a horn, a few other safety bits, and a VIN I made and attached to the frame, I could register it as a car. There's no need for an engine serial number in that.

Would I add an engine serial number to the block casting? Of course, it's a good idea and also fun. But legally required? Nope, not at all.

>>798246
Don't know about local clapistani laws, but my country removed engine serial numbers from registrations a long time ago.
Now you only have a VIN and the engine data is just displacement+power.

>>798249
>in the state of Idaho: if I got this engine running and put it in a small go kart that had lights, a seat belt, brakes, a horn, a few other safety bits, and a VIN I made and attached to the frame, I could register it as a car. There's no need for an engine serial number in that.

Damn, I envy your freedom. The amount of legislative bullshit a car needs to have here is enough to completly negate any kind of car manufacturing that isn't already a giant corporation.

And the only guns we don't register are pre 1885 designed black powder guns and air guns with less than 17 J muzzle energy.

>>798253
>Damn, I envy your freedom. The amount of legislative bullshit a car needs to have here is enough to completly negate any kind of car manufacturing that isn't already a giant corporation.
>And the only guns we don't register are pre 1885 designed black powder guns and air guns with less than 17 J muzzle energy.
I... I am so sorry, fellow /diy/er.

Even living here in California (also known as Kalifornistan), I can build and not register my own guns, and do anything I damn well please to a car if it's older than 1975. Any engine, anything (as long as it has the stock safety equipment like blinkers, seat belts, etc).

Hell, cars older than 1965 don't even have to have seat belts!

Also, we don't have car safety inspections here. The only "inspection" we have is a smog inspection to see if your car passes emissions regulations, and that's only for cars newer than 1975.

>>798249
wow, in most of the world its mandatory. the engine number gets registered with the VIN. you can change the engine number attached to the VIN though. its to stop chop shops.

>>797991
If you keep it aircooled, you're going to have to fab up some air ducts to get air to the rear cilinders. Keep in mind that you've got a 6 cilinder 2 stroke, instead of a 4 cilinder 4 stroke, so 50% longer, and twice as many combustions per revolution. They're gonna get hot really damn soon.

Getting 6 carbs to work in tandem is the problem.

Any carburator can be blow through, but they have to be modified for that.
Port or manifold fuel injection would not require valves on a two stroke.
If you're making just 3 psi, you're making a whole lot of exhaust restriction. The potential 20% power increase you could get will be offset by the exhaust restrictions you make. Keep in mind twostroke exhausts are a science in themselves, and I'd love to see a straight six exhaust.

Injectors get a better fuel stratification, and usually result in a better vaporasition. This will raise the engine's knock resistance, allowing for either a slightly higher CR or lower octane rating.

DOHC would be for the ''ultimate goal'' engine you specified in >>797672

>>798172
Tore Morten, is that you?

>>797704
Honda CBX

>>799018
Sir, the CBX was a 4-stroke.

>>799021
And thus even more complicated than a 2 stroke 6. The post says air cooling a i6 mounted transverse isnt possible. I can give 1 example that was.

>>799018
nice choice and cool engine, but that engine was mounted transversely, while I'd prefer a traditional longitudinal mounting in a chassis. Air cooling an engine with six cylinders with just air gets difficult, unless I were to get a body on the go kart with good air vents.

>>799021
literally no bearing on the subject of cooling, but thank you regardless.

>>797698
I'm almost positive those small 33cc engines have roller cranks for bottom end oiling. I think you'd be better off pressing together a multi piece crank ala kz750 2 stoke. Parts can be made individually and pressed together with rods in place. Will be a pain to balance, but a cast crank won't be much better.

I think the case will actually be fine, aside from the casting and machining tolerance required. The pressed crank will have main bearings in place. I think your main challenge will be ensuring that bottom end volume is minimum to maximize scavenging.

I'd run 6 carbs, one to each jug, and 6 expansion chambers. Suggest 60 deg clocking and fire 2 cylinders at a time to simplify ignition.

>>798942
>Tore Morten
Nope.

>>798140
Full blown ecu is probably overkill for this app unless he's going fuel injected. If cylinders are fired 2 at a time you could just set up a trigger wheel with 3 pickups, going to 3 scooter ignition modules and coils. Coils dual fire the cylinders. The scooter modules will take care of ignition advance vs rpm.

>>799074
Oh definitely. Heck, if you choose the right advance you can run an engine static. Need a coil that can handle balls to the walls dwell though.

>>799066
>I think the case will actually be fine, aside from the casting and machining tolerance required. The pressed crank will have main bearings in place.
except six? In a line? Seems a bit insane to expect it to have good torsional rigidity when spinning at 10,000RPM.

Maybe pressed together and then welded? I'd have a one time use crank then, though....

>>799127
Look at a kz750 crank. Torsional rigidity will be fine... This engine won t develop that much torque. And 10000 rpm operation isn't about stiffness, its about balance.

>>797698
Learning CAD would help a lot. That way you don# have to cast small fiddly parts, just get them CNC milled.

>>799127
What to me is insane is the idea you'll be able to pull this off at all. But who am I to judge? I'm just here to enable.

I was wrong about the KZ750 crank... what I meant was to refer you to roller cranks of pretty much any use. Here's one off a 4 cylinder motorcycle. It was welded up.

>>799188
>>799746
And here's one from an I8 Bugatti, showing that it's technically possible to make this for a 6.