Try this:
1. Clutch engages, variator does nothing until peak power rpm.
2. Electronically controlled linear actuator(of some kind) continuously adjusts spring tension to make variator stay at peak power rpm regardless of heat, belt stretch, etc. while it's climbing through the "gears".
3. Variator stays at peak power rpm indefinitely cause final ratio is higher than what can be reached with available power.
What I'm hoping to achieve anyway.
Completely off topic:
At 18000rpm with 145/90 timing the intake is closed for about 1.16ms during one revolution - that's a very small window of time for the intake charge to slow down, stop, then start accelerating in the opposite direction.
I think you know where I'm going with this...
Check out my YouTube channel! - 2STROKE STUFFING -
https://www.youtube.com/2STROKESTUFFING
Two strokes & rum!
Yes, there seems to be some sort of limit. Speedpros turbo twin (18,000 rpm) four stroke would never go over 16,000 rpm with his Ecotrons unit.
Ecotrons turn off the VE table of their four stroke EFI firmware so that only the Alpa-N table is active and sell that as their two stroke EFI version.
My single cylinder two stroke was naturally limited to less than 13,000 rpm so I never found out if that 16k limit applied to two strokes.
My opinion, for what it is worth:-
I found Ecotrons help dept unable to help very much with experimental two stroke tuning and is the main reason I moved onto another EFI system that I could modify for myself.
At 16k two stroke rpm you are running into serious “Time” issues as you have only 3,75mS to get everything done and just opening/closing the injector takes up most of this time without delivering any fuel.
Assuming “B” port or case injection. The good news is, that because there is always carry over fuel from one cycle to another and it takes six or so crank revolutions to completely consume the cloud from any particular crankcase injection event it is possible to get away with averaging the fuel required over two cycles.
KTM take advantage of this cloud averaging for their staged injection by alternating their “B” port injectors side to side and when they need more injector capacity to cover the reduced time available at higher rpm they fire both “B” port injectors together.
I think that due to cloud averaging it would be possible at 16,000 rpm to fire twice the fuel required and spread it over two cycles and still have pretty much the correct air/fuel ratio in the cylinder all the time.
Staged injection. Roughly. The volume of fuel required each cycle at Idle is not much different to the volume required at peak torque. Just at peak torque less fuel is wasted. You don’t need the bigger injector at peak torque because you need more fuel than at idle. You need the bigger injector so you can deliver the same amount of fuel in the reduced amount of time available at higher RPM. Staged injection allows you to use a small injector at idle and reduced throttle and engine loads for refined fuel tuning and a bigger injector to get the job done at higher engine loads and rpm.
Because Methanol is so forgiving of rich mixtures and your going to basically run at WOT and high rpm I would consider using an Arduino Nano and write my own fuel injection firmware. You don’t need a lot of the refinements found in regular EFI systems.
Nitro carries slightly more fuel than oxygen. So if a motor is already jetted to run correctly and then you add Nitro through another jet, like a solenoid power jet the air/fuel ratio automatically gets richer. Nitro could be switched in/out as desired.
The inputs required for a basic EFI are fairly simple.
TPS … throttle position.
RPM … revs.
IAT … inlet air temperature.
Baro … barometric pressure.
LCT… liquid coolant temperature.
TPS and RPM for the Alpha-N fueling table that is used to predict engine load (air being consumed).
IAT and Baro for Air density.
LCT for fuel enrichment. Both when cold starting and enriching again when the engine starts operating in the golden zone and controlling the water pump for consistent engine temperature.
The Arduino Nano with some MosFet add on boards could easily handle all of this.
A Nano can easily count in steps of 4 millionths of a second without tricky coding. Not that you will require very accurate injection positioning. As with cloud averaging you will completely inject for the first cycle and part of the second so accurate positioning becomes a bit moot. But if you have a choice, the end point of injection is more important than the beginning. Making the end point at BDC or just before the transfers close seems to be best. Unlike a four stroke you don't need a complicated multi tooth wheel because you don't need to know which cylinder is up or down and who is next for a squirt of juice.
The EFI algorithm for Methanol can be quite simple. Fuel load is a math calculation based on air density and fuel required for the air consumed at the engine load as predicted in the Alpha-N table.
A simple fuel injection setup like this could easily compensate for changes in air temperature and elevation making tuning at home for all out WOT runs at Bonneville easy. No mucking about with carbs and jetting when you get there as the EFI system will have automatically adjusted for changes in environmental conditions with the on board IAT and Baro measurements.
As an aside, instead of being able to just predict air flow (Alpha-N) for a good drag at WOT. For road racing, due to changes in the pipe resonance effect at low throttle settings. To get good throttle response you have to actually measure changes in the air flow through the motor to keep the air/fuel ratio correct when trailing the throttle through a corner.
One small point not to ignore - running Meth you need more than twice the fuel flow per injector on cycle.
So every part of the delivery system has to be sized to deliver this extra amount over petrol injection.
Ive got a thing thats unique and new.To prove it I'll have the last laugh on you.Cause instead of one head I got two.And you know two heads are better than one.
then add in more than about 15% nitro and fuel demand starts to increase quit alot. with just meth i had one pj basically fully open. with 35% nitro i tried four pj fully open and it was just barely starting to get to rich.
there wasnt much trouble getting full throttle sorted out. was just a matter of turning all the pj full open then close them little by little until it ran good. but closed throttle and idle was a different story. i grinded my own needle in increments until it was rich enough at idle but when the throttle shut after full throttle ,the rpm would run away for several seconds, althought it sounded scary, it never seized. . all the while i was fighting a flooding crankcase at idle so im not sure if richening the needle further would of helped the run on. perhaps it would of helped but the crankcase flooding would of been even worse im sure. i did find a simple solution to the run on, after throttle closes just wait a few seconds before pulling in the clutch. this prevented the piston from running away, while giving the crankcase time to refill with fuel. for a standard gearbox it worked great but that method likely wont work if you have a clutchless gear box that cant tolerate a load on the coast side of the gear dogs. which is going to be my main problem with nitro in this yami twin. how will i stop the run on, without back loading the trans. but i love a good challenge so we'll see what happens when i get to that point
i wanted to try a hilborn mechanical injector but that requires a large pump, driven by the crankshaft or some other meens. alot of money and time to figure it out. special made pulleys and brackets. in the end, would it even work. fuck it. i didnt even want to try. ill keep the shitter lectron instead
i hope adegnes gets his idea working good
Not sure if anyone has had a good look at this as an option or not. https://www.nanoefi.com
Sent from my iPhone using Tapatalk
forgot to mention fuel consumption of that particular engine [527cc single] was around 2.3L for 370meters with %35 mix. so yes the fuel system needs to be free flowing
Here is an example
http://mdac.com.au/
And another
https://speeduino.com/wiki/index.php/Speeduino
.
Of those three, the Speeduino is my pick because the firmware code is open source which means you can modify it to suit two stroke applications.
The other two are not as far as I can see open source and being aimed at four strokes you might have trouble adapting them to two stroke service.
If they tell you it is two stroke suitable the key is to ask "how do you measure air flow through the two stroke motor at part throttle?". Hint, it is not the same way as you would for a four stroke.
There is a big difference between nitro mixed into the methanol and nitro inducted separately to the methanol.
It might be worth trying the motor setup for methanol only and using a separate system for dropping nitro into the inlet when you need it. It could be as simple as a gravity feed system from a header tank with a valve that is opened by a re purposed auto lube oil pump cable at WOT. And if you want to go richer when the nitro is dropped in then mix a little extra meth with the nitro.
other than avoiding the crankcase flooding, what could be the benefits of introducing nitro seperate ? seems like it would be a bit more tricky to tune the full throttle aswell, since there would be a jet for the meth and perhaps multiple jets for nitro. and how do you know how much nitro your feeding the engine ? im going with frits old saying, kiss![]()
Yes Kiss and Simple Benefits ..... totally agree. Introducing the nitro separate makes tuning simpler.
After you have got the methanol air/fuel mixture right adding nitro separately is simple because the extra nitro makes the in cylinder mixture slightly richer and meth likes rich. You can just keep adding nitro separately to your hearts content without having to adjust the previously correct methanol jetting. Because the methanol was adjusted to use up all the normally inducted oxygen and the nitro brings its own fuel and sufficient oxygen to burn most of it.
First you get a good Methanol party going then Nitro is like the perfect surprise "life of the party" guest who brings their own booze and food with a little extra for the table.
The Hillborn and similar systems (Crower, there may be others) are pretty crude. Big pump as you say, and the fuel regulator is basically a variable restrictor on the tank return line. What doesn't go back to the tank goes into the engine. (the systems I have experience with, anyway - V8 speedway motors running methanol, and possibly "additives" if no-one is looking) WOT was the only parameter they cared about.
it's not a bad thing till you throw a KLR into the mix.
those cheap ass bitches can do anything with ductape.
(PostalDave on ADVrider)
There are currently 127 users browsing this thread. (0 members and 127 guests)
Bookmarks