|
Let's get something out of the way, here... the title of this thread is not being facetious in the slightest: https://www.youtube.com/watch?v=42BRqshbYA0 My coworker owns a 1970 Subaru 360 Sambar. If you're not familiar with the Subaru 360, it's a Kei-class car imported to the United States by Bricklin (the same guy who brought us the Yugo!) in the late 1960's. It has a 360cc two-stroke engine producing somewhere between nothing and almost nothing for power, and even with a curb weight of 998 pounds it can barely get out of its own way. It was quite seriously marketed on the basis of it being "cheap and ugly," and being labeled as "the most unsafe car in America" probably didn't help sales too much either, so they're definitely one of the more esoteric cars to own and work on in the United States. So of course we're loving with it!  If you haven't watched the video linked at the beginning of this post, do so now and it'll give you a pretty good idea of what we're dealing with here. The plan is to take the engine and wring as much performance as feasible from it by converting to fuel injection. I have had quite a bit of experience with Megasquirt EFI and the tricks and traps of full standalone tuning with no base maps, and it was not a terribly hard sell to my coworker to push him to convert to EFI after we had to push the car into a Wendy's parking lot and let it dry out after the engine got flooded trying to start it. The primary goal is achieving improved reliability for daily drivability, but I believe that doubling the stock horsepower of the engine should be easily possible without major rework of the engine (cooling will likely be a problem, but hey, we'll cross that bridge when we come to it). The most obvious place to start with "why is this car so drat slow?" is the carb. Let's take a look at the carb!  (not visible: the drip pan intended to collect all the fuel that pours out the front of the carb in normal operation. Seriously, that's a thing on this carb)  It can't be seen from this angle, but the final diameter of the port that mates up to the intake manifold is 28mm. The engine must be gasping for breath all the way to 5500 rpm. According to some very questionable internet sources, a twin carb model resulted in 36 hp versus the 25 hp of the single-carb model. The actual intake ports on the engine side are pretty big considering the displacement. I don't think there'd be much to gain from enlarging them, at least at this point in the project. Also it'd be very easy to get debris into the engine block, and they're not the easiest thing to source at junkyards if we wrecked one due to carelessness.  To replace the carb, we sourced a Kawasaki Ninja 300R throttle body set. The 300R is fuel injected from the factory and is probably pretty similar in terms of what we can expect for airflow through the engine, so it seemed a good match to the engine. Also, it looked like the intake spacing and the spacing of the throttle bodies was about right. Sure enough:  The spacing is drat perfect. This is gonna kick rear end. Based off of some kinda sketchy gasket patterns and a whole bunch of caliper measurements, we modeled and 3D printed intake runners to mate up to the throttle bodies.  We had to take about a centimeter off the runners to avoid crowding the mouths of the throttle bodies with the engine compartment hatch, but otherwise the mockup fit went very well. We'll probably send out the drawings to be machined sometime in the next week or so. For engine management we'll be running a Microsquirt ECU in Alpha-N mode, using rpm and throttle position for controlling fueling. Between the low 6.8:1 compression ratio, the engine being a two-stroke, and the individual throttle bodies, I highly doubt that manifold pressure will tell us anything useful. We're just doing fuel for now, and we'll work on spark after we get a rough tune in place. The stock points system sucks, but it should work well enough to get things running first. I figure we'll use a coil-on-plug setup in the end, if we can fit them on the engine, otherwise we'll get a couple coils and drive each cylinder separately. O2 sensor duties go to an Innovate Motorsports LC-2. I've owned a couple of their LC-1s before, and they were reliable, well designed, and cheap. The biggest complaint I had with them was that they didn't have any way to mount to anything, so you had to zip tie it onto something. They fixed that with the latest model, and I guess they also improved response time while they were at it. Kick rear end. The coworker doesn't have any welding equipment, and I had to leave all mine with my folks back in Massachusetts, so we've gotta get an exhaust shop to weld the bung for the O2 sensor onto the exhaust, but that shouldn't be a big deal once they get past the rust on the header pipe.  An unresolved issue is how we're going to handle fuel. I suggested getting an Aeromotive 13129 FPR and then routing the return back to the drain plug on the gas tank to avoid having to cut a new return line hole and getting it welded in. The 300R uses a returnless fuel system and it seemed like too much of a hassle to work out how best to control the fuel pump without making everything too complicated. Using a fuel pressure regulator with a built-in bypass should make that decision a little easier, at least. That's all the pictures and info I have for you all so far. I think it's gonna be a pretty kickass project, and shouldn't actually end up being too terribly expensive, considering. I shall update when we get a chance to wrench on the car again, possibly next weekend?
|
#
¿
Apr 3, 2015 09:00
|
|
|
|
| # ¿ May 10, 2024 02:37 |
|
|
I find it absolutely hilarious that the owner of the car posted about this project on the Subaru 360 Club forum and was greeted with so much haaaaate from pretty much everyone but the mod of the forum and one other guy who is doing a similar project. My favorite quote is "My 360s are not unreliable. When they have been gone through properly, they're more reliable than my '59 Thunderbird." High water mark right there! Alternatively, "I would rather trust the expertise of the original Japanese engineers than something cobbled together on my work bench at home." to which my coworker responded "And I would rather trust 45 years of automotive innovation, modern manufacturing tolerances, and an actual budget!"  W.r.t. the Lotus, I find it hilarious that here in Silicon Valley nobody seems to give a drat about it, but EVERYONE who sees the 360 comments on how awesome it is and how much they want one. I believe that after being converted to EFI, it'll be reliable and zippier enough to agree with their sentiment. 
|
|
#
¿
Apr 3, 2015 17:00
|
|
|
Seat Safety Switch posted:The fresh air vent and suicide doors are the coolest parts of the regular 360. Does the Sambar still have them? Sure does!
|
|
#
¿
Apr 4, 2015 00:13
|
|
|
It's got a mechanical injector pump that sprays into the crankcase.
|
|
#
¿
Apr 4, 2015 08:21
|
|
|
poo poo, it's really way past time I update, isn't it? Well when we last left off, we were mocking up how we were going to put the whole thing together. Then my coworker flew to Japan for a couple weeks and I got busy at work, and not a whole lot happened. MEDIOCRE. We got some parts in:  Cheapo inline Airtex pump, not so cheapo Aeromotive fuel pressure regulator, fuel pressure gauge, a bunch of AN-06 fittings and hose. Fuel systems are important, yo.  I got voluntold to install the fuel pump. I am glad that immediately beforehand I thought to put on goggles, because in the process of doing so it seems like I got drowned in all of the gasoline that has ever been refined in the history of mankind. My coworker forgot to remove the filler cap... it was under pressure in the tank.  I suppose it also didn't help that the nipple on the fuel tank was for 5/16" hose, while we were using 3/8" fuel line like normal people. Hooray for hose clamps! Of course the entire time that I was fighting with the hose clamp trying to get it tightened down, gasoline was pouring down my arm and both of us were drat near zonked out by the fumes afterwards.  The fuel pump was in, though! It is actually mounted in a pretty good location right before the engine bay. Should be fine. Some pictures of the feed and return lines:   As part of the ongoing process of pulling lovely old parts out of the car, the leaking factory oil tank was next.  This revealed the only rust-free spot on the car:  (It leaked a lot of oil. A loooot of oil.) Eventually we realized that we really had to just go ahead and get the ITB runners machined already or else nothing would ever get done. My coworker placed the order from ProtoLabs and we had them just a few days later. Sweet! I've used their services a few times now and they're pretty excellent to deal with if you need custom parts machined quickly.   I think it came out pretty well. Conveniently, K&N had an air filter that was the right size, runner diameter, and spacing for the throttle bodies.  That logo gives something like 15 extra horsepower all by itself, right? Also pictured there is the fuel fitting we pulled off of a factory fuel line for the Ninja when we realized that it really was not going to be a good idea to just jam the fuel line on there and clamp it. The line was cheap, though... something like 12 bucks on eBay. We were really tired of working in the pitch black, so the coworker bought a new absurdly bright light fixture for the garage... Old and busted:  New hotness:  It's still too dark in there at night, but it's a huge improvement over what it was. There are some more things to update on, but this post is getting absurdly long already, so I think I'll pause for now.
|
|
#
¿
May 29, 2015 18:16
|
|
|
Commodore_64 posted:How many amps can the charging system put out again? Those pumps can get kinda thirsty. Awesome proto part there, though. Are those SLS or something? The factory generator is about 500 W... before conversion losses. Let's just say that that is long gone. We've switched to an alternator from a John Deere tractor which more or less bolted right up. The runners are machined and bead blasted aluminum. Part of the reason for their kinda funky shape was to ensure they'd be machinable on a 3-axis CNC.
|
|
#
¿
May 30, 2015 01:23
|
|
|
It's full featured... lamp/field/sense. Because we're not using a couple of the relay slots in the fusebox we got at a junkyard, I might set the Megasquirt up to switch the alternator off while cranking. The cranking speed is much higher without that load on the engine, about 450 rpm versus 300-350 rpm with the alternator on.
|
|
#
¿
May 30, 2015 03:01
|
|
|
A teaser (until I have time to finish writing things up): https://www.youtube.com/watch?v=FKtzfF1vg1U
|
|
#
¿
Jun 3, 2015 08:25
|
|
|
mobby_6kl posted:Can you turbo this thing for double HP again? We'd have to make drastic modifications to the engine to keep it from blowing the intake charge right out the exhaust... the engine has no reed valves or any other form of port restriction. I think we're better off sticking to NA power with maybe some exhaust tuning cleverness. Ultimately, it's tuning the ignition timing that will make us power. Currently we're using the stock points system switching a GM HEI module to drive the coil. We'll convert to coil-near-plug once we have it generally sorted.
|
|
#
¿
Jun 3, 2015 21:30
|
|
|
Thankfully the oil pump's dynamic range is actually pretty limited. It's about three times as much oil per rev at full throttle as it is with the throttle fully closed, but it's still a fair whack. We will definitely take into consideration doing some premixing anyway, as I too have raised an eyebrow at the oiling system and its potential adequacy for full power runs. I will continue to disagree with your statement about carbs though, as I see no physical reasoning why a carbureted setup should be better suited to run a two-stroke motor than a properly tuned EFI system. Yeah, pressure pulses from the engine will oscillate in the carbs and add extra fuel, sure. Those pressure pulses won't magically behave differently with EFI, though, and they can be accounted for in determining the fueling table for the engine. The actual physical intake path is essentially identical to the carbureted setup, the exhaust path is exactly identical, and the dynamics of the engine should be largely unchanged. I'm not sure what you mean by "four-stroke induction" either, as the injectors are timed to fire at approximately the same time the carb would be feeding fuel into the intake charge. That too can be adjusted. I'm aware of the differences in ignition timing between two- and four-stroke applications, and spent the weekend reading about them. For now we are running off of the factory points system as a timing source for the ignition module, but we will switch to full spark control later on when we've got things settled as far as oiling goes and have a base tune. The factory distributor is an extremely basic design and I'm not sure it starts to retard after torque peak like it should on a two-stroke. From talking to a couple other folks who have converted their 360s to EFI, it appears that it has a nominal cranking advance of 6 degrees BTDC, ramps up to about 30 degrees BTDC at 3k rpm, and then flatlines around there. This is most definitely more advance than we want at higher rpm and load, and cutting that back might help free up some torque up at 5k+. The factory redline is only 6k rpm so... a lot of this is future work after we take care of prepping the engine for more. Thankfully there's a dyno shop not too far away in San Jose where we'll probably bring the car to tune things up when we're ready for that.
|
|
#
¿
Jun 4, 2015 07:37
|
|
|
We're in NorCal. The car was purchased in Oregon, because Portland, of course. The coworker travels up to Seattle pretty regularly for work though, and I'll probably be up there at some point too. Would be fun to meet some PacNWers.
|
|
#
¿
Jun 4, 2015 07:49
|
|
|
I haven't had a chance to post up the electrical wiring pictures and all that jazz, but I'll try and handle that this weekend. The extra horses will definitely make things run hotter. The factory fan is enormous and flows a huge amount of air through the squirrel cage layout they use, but we may end up strapping on some extra cooling fans to the bottom of the engine bay to try and force some more cooling air through there if necessary. Right now we're using exhaust air temp from the cooling routing as our "coolant" temperature for the purposes of engine warmup enrichment and it seems to do quite well. We'll probably end up jamming a K-type into the cylinder head at some point to get a better measure of CHT. Another person who has converted his 360 to EFI did that to keep tabs on everything. We're also probably going to get an EGT bung welded into the exhaust to measure exhaust temps for our tuning. That can wait until we're ready to start adjusting ignition timing, though.
|
|
#
¿
Jun 4, 2015 08:10
|
|
|
https://www.youtube.com/watch?v=CtV95U1Mat0 We have not yet doubled the juice. I suspect that 1. the throttle is not opening fully with the pedal planted and 2. ignition control will make a huge difference in performance.
|
|
#
¿
Jun 8, 2015 08:18
|
|
|
rscott posted:so will editing your videos so that horrific wind noise doesn't drown everything out jesus loving christ If you weren't hearing the horrific wind noise you'd be hearing the horrific volume of the engine or the horrifically bad commentary we have for the car's performance. DWI. Nerobro posted:you're going to find a TON of power in ignition timing. I can't wait for the results. I too cannot wait. We're ordering the coils this week to set it up for coil-near-plug. I grabbed pretty much all the 2-stroke books I could find. Tuning 2-strokes is a whole lot of voodoo, apparently. I have a feeling that dyno tuning is really the only surefire way to improve things, although it'll be pretty easy to tell if we're freeing up power on top speed runs! Slavvy posted:Does it have any kind of expansion chamber setup at all? Nope. I guess another 360 owner has made them and has them for sale, but they are ridiculously loud even with mufflers and I'm really not sure that that's a road worth looking down. The power band would be absurdly restrictive, even worse than it is now. Making reed valve cages would be a much better path forward for keeping the car "streetable."
|
|
#
¿
Jun 8, 2015 08:55
|
|
|
I'm the doofus copilot!
|
|
#
¿
Jun 8, 2015 17:37
|
|
|
So much downforce!
|
|
#
¿
Jun 9, 2015 02:19
|
|
|
Checking the throttle swing last night confirmed that "full" throttle there was only about 70-80% open. We're probably going to redesign a new throttle arm for the pedal which will give more travel.
|
|
#
¿
Jun 10, 2015 22:08
|
|
|
This is a car where the only factory switched 12V source in the engine compartment was the ignition coil positive line. There ain't any computer-reported anything beyond what we've programmed into the Megasquirt. The TPS is set to 0% at idle position and 100% at wide open. The trigger wheel is one of the sharp toothed small 36-1 wheels from DIYautotune.com using a Volvo hall effect sensor. Works great.
|
|
#
¿
Jun 10, 2015 23:55
|
|
|
|
| # ¿ May 10, 2024 02:37 |
|
|
What car are you 'squirting?
|
|
#
¿
Jun 11, 2015 00:23
|
|