Something about Variators although you didn't ask.

[@2STROKESTUFFING]

Get an exclusive 15% discount on Saily data plans! Use code 2strokestuffing at checkout. Download Saily app or go to saily.com/2strokestuffing

[@geemy9675]

the real question is why would you want a variator? you are throwing a bunch of variator tuning issues on top of the engine tuning, plus you are making less power at the wheel and making dyno testing way more complicated than it should be. just have a short ratio 6 speed gearbox, that will let you get to your top speed while staying in the power and during shifts. ball park the top speed you're aiming at and have a few sprockets for both best case scenario, and backup bigger sprocket if things don't go too well. you'll be able to adjust your shift points if carburation is impacted by altitude/temp and your power and moves. trust the rider skills and ability to adapt vs variator that will need constant adjustments. also land speed record is not an acceleration contest, you just need to get to your top speed before you run out of room or engine overheats/ breaks down. just need a beefy clutch to get enough slippage until you are into the powerband in 1st gear I mean I see where you are coming from. if you could make an extremely peaky engine with very high power but extremely narrow powerband, and you could tune a variator perfectly to stay within this powerband, you could see some gains that would only be achievable with a gearbox having a lot of speeds (like a fast & furious 20+ speed gearbox) but I think in reality gearbox will win. btw, single pulley variator will only get you so far as gear range is concerned, so you won't even have this advantage.

[@mikeruefenacht7354]

I like the idea of a gearbox. But 6 speed will not be enough. They used up to 21 gears in the 50cc gp bikes. Im not telling this is right but i thing 6 gears to go up to over 200kph will not be enough

[@heinzhaupthaar5590]

@@geemy9675 Variators aren't that hard to tune properly, most adolescents get that done well enough with no problem. As Mike said, six speed most certainly won't get you anywhere at all. If anything like that, it would rather have to be 3x6 speed or similar and I'm quite positive a simple variator, that doesn't even need to be tuned to anywhere near perfection, is way less of a hassle than an eighteen speed double gearbox would be. Don't forget - it's a tiny 50cc. Making it extremely peaky is not a choice, it's simply a prerequisite to make any meaningful power at all.

[@geemy9675]

@@heinzhaupthaar5590 six speed will get you just as fast as a variator you just need to gear it properly for top speed. even a peaky engine can go through the gears of a close ratio gearbox, which narrows down the problem to starting the bike with a tall 1st gear. this is the same problem wether you're using a gearbox or a variator with very tall gearing which also has a limited gear range. in both cases you will need to waste a bit of runway to get the bike moving to the power and in first gear/lowest ratio of the variator, which will require a lot of clutch action. the advantage of the manual clutch and gearbox vs CVT and automatic clutch is the rider controls the clutch. also the runway lost to get the bike into first gear isn't that long compared to the runway required to reach top speed (high speed > lot of ground covered per second and more time to accelerate at higher speeds. the two solutions are possible but the variator is more tuning work to do under stress and in hard conditions on the flat sands. variator setup needs: carburation + final gearing + variator weights + clutch biting point vs manual gearbox: carburation + final gearing and the rider controls the clutch to launch the bike + the shift points. if you miss your launch and stall the bike you can restart it and retry vs tuning the clutch. if you shift at too high rpm you just lose like a second vs variator tuned for too high rpm wont move the belt high enough inside the pulley=> you lose top speed if you short shift the manual gearbox and the bike bogs down, just downshift and try again, vs variator tuned for too low rpm will will stop accelerating before it reaches its potential top speed. teenagers tune their variators and usually make a compromise between acceleration and top speed. if you have been a teenager tuning a two stroke moped variator, you'll know a variator tuned for top speed doesn't provide the best acceleration and variator tuned for te best acceleration falls short in terms of top speed. highest performing stock 50cc two strokes with the peakiest engines like Aprilia rs50/Yamaha tzr 50/derbi gpr50 etc were all using gearboxes. Grand Prix racing requires to have a gear properly geared to exit each corner of a race in the power and. land speed racing is about being able to get from each gear to the next in a reasonable time. totally different job, although 50cc two stroke gp bikes from 60 to 80s have used anywhere from 4 to 9 speeds. I'm not aware of a 21 speed grand prix motorcycle, but if it did exist it was probably an outlier. BTW the 2018 bidalot 50cc land speed record was with a pretty much stock/mildly tuned bidalot rf50(wr) that has a pretty flat torque curve, and 6 speed gearbox. the run is pretty uneventful, you can see the rider simply redlining the engine and giving very gentle clutch bite to get moving without stalling, then it's just about rowing through the gears. kzbin.info/www/bejne/n6DZkqavjc-sgdk

[@1-3645]

@@mikeruefenacht7354 The more gears the more metal (weight) you have to throwe around..

[@adamadamhoney]

Also once you get repeatable pulls and data, then worry about gearing. I agree the close ratio box is probably more usable for initial testing at least. Weigh isn't huge for what he's doing, at this point

[@hygri]

I love your surrealist rambles though the metal meadows. Here for the journey.

[@heinzhaupthaar5590]

I think if you run into any problems with variators it might be well worth having a chat with some snowmobile folks - and if I'm not mistaken you might be in a pretty good region for that. Afaik their variator cvt systems are by far the most developed around if you exclude highly complex car transmissions and such. They use levers as well and get a really great deal of very fine tuneability out of them via different weights on different places of the lever, varying the lever itself, taking into consideration the changing geometry of an outward moving lever and whatnot. Might be in part due to the variator taking on the function of the clutch as well in modern snowmobiles, but in the end it doesn't really matter how they got there.

[@rogalandimportservice6747]

The sled guys also know how to turbocharge 2strokes 😅👍

[@heinzhaupthaar5590]

@rogalandimportservice6747 Yeah, I had the turbo conversation with Alex (that his name?) already a few years back, but I didn't think it through - 50cc is just way too small for even the smallest available turbo. Even if it would manage to make the turbo move some air, with a two stroke you need a certain drive pressure in the turbine side and I highly doubt a 50cc can manage enough exhaust gas for 15-30 psi. That's what makes turbo two strokes work in the first place after all, and it's the part of the equation all those naysayers going on and on about how it's impossible to turbo a two stroke fail to understand. I'm not sure if you even meant it that way though. Anyway, those turbo sleds sure are a sight to behold. 500hp 750-1000cc two strokes, cvt and incredible traction - that's just utterly insane.

[@rogalandimportservice6747]

Yeah I know the issues about small displacement, but I think the practical from the 2stroke turbo sleds could be usefull for alex with the rotrex blower. Also look at what fueltech is doing now with the big turbo promod engines with the e-gate, just dumping excess boost when needes. Exiting times with 5400++ hp at the hubs of doorslammer cars 😁​@heinzhaupthaar5590

[@TarenGarond]

@@heinzhaupthaar5590 The 50cc world speed record(Buddfab streamliner) was done with a turbocharger so it seems viable.

[@heinzhaupthaar5590]

@TarenGarond Interesting, didn't know about that one yet. I knew smaller turbos became available more widespread in the meantime, but considering just a couple of years ago we weren't able to find anything from a reputable manufacturer that would've worked for our 150cc project I'm still quite doubtful if there's anything off the shelf that would work. That it's generally a viable route wasn't in question anyhow, just the availability of an affordable unit that fits the application.

[@morilamorila2825]

I like your aproach to problems. Yes there is a viable solution. but if we walk on save well known ways, we can never discover new maybe better places

[@ariel2t]

I agree. About going in well know ways, but don't mean that you can't try new ideas. For example controlling the ratio of fuel for methanol and nitro, is a thing to be tested. But injection is a challenge in it self. So a safe way could be using two carbs in the same Engine. One with 100% methanol and the other with 75% methanol and 25% nitro. Setting jets in a carb is well know operation

[@gregorytripodi6917]

​@@ariel2tk

[@lesmaybury793]

I wondered how those things work, so many thanks for explaining that so well. I watched another video recently where riders were complaining about juddering and squealing sounds on take off. The guy had an ADV350 with just a few thousand miles on it. He striped the variator and it was full of dust off the belt. A good clean and it was as good as new.

[@colinhamer6506]

This might be a good time to mention a idea I had again. The basic idea is to take the idea of a variator and forget the centrifugal part for now and make it change ratios using a similar set-up to a manual car clutch having constant spring pressure on the variator plates and using a thrust bearing to move them with anything from a simple cable and lever to a automated hydraulic or motorised setup. There's also the option of adding a thrust bearing to a normal variator to give you the ability to have some manual control over it

[@MrJermbob]

Yeah hydraulic clutch throw out bearing behind it was always my idea. But honestly a dialed cvt doesnt need manipulating. This is just trying to reinvent the wheel kind of junk.

[@colinhamer6506]

@MrJermbob I agree when a cvt is diald in it in theory should be the best way getting to that point while playing around with crazy experimental ideas is one of the reasons behind what I'm thinking being able to manually play around with the gearing without the need to change parts. And having the ability to take over control when doing speed runs at Bonneville should increase the chances of getting the best results

[@adamadamhoney]

Could possibly control a pwm variable solenoid with an Arduino and MOSFET calculated of RPM and add validation against ground speed. Probably need an electric over hydraulic system to use this but would be low pressure and volume. Wouldn't be that hard to design and a adapt some valving to single act a piston variator control and still have a manual clutch for launch, specifying the launch range to keep the variator ratio static as a set lUnch rpm range

[@adamadamhoney]

More thoughts, could use gearbox oil for the hydraulic, will probably need an oil bathed area for a Torrington to actuate the piston on the aviator. Possibly modify the case to house and run an oil feed as a bleed from the hydraulics using engine oil and a larger pump for the circuit. I hope this makes some sense, I've had a few😅

[@colinhamer6506]

@@adamadamhoney 👍

[@mattyoung2880]

I ride dirt bikes in the summer and snowmobiles in the winter. Snowmobiles run a ctv clutch setup and I’ve learned lots from Aaen’s clutch tuning book over the years. Well worth the money to figure out the magical CVT clutch !!! 😊

[@CalvinWenzel-ri2pt]

Look up Barry Holtzman. He was working on a mechanical fuel injection system for 2 stroke snowmobile engines long ago. Was close to perfecting it , but never heard anything more. I think it used cylinder pressure to spray the fuel into the engine.

[@paulevans5822]

I'm here for the ludicrous ideas.

[@davidcoates4852]

I bought a scooter that someone had put heavier rollers in, the effect was the front pully would almost fully close before the clutch engaged. The symptom was like the engine was bogging down and the scooter would not move. It was like trying to move off in top gear, no torque. I didn't even buy new rollers, I just drilled the centres out to make them lighter. 🔧👍

[@jeffreybago7439]

What if you used a two-speed clutch setup? One clutch on motor with extra sprocket. Second clutch on your Jack shaft with one -way ratcheting sprocket. They use that setup on race bicycles with motors.

[@IFRYRCE]

I didn't ask, but I wanted to know anyways! I look forward to anything you make a video about. I have been an EFI evangelist forever though, I firmly support you just going with EFI. Megasquirt (or Microsquirt, specifically) is cheap, basically infinitely configurable, people have run 2 strokes on it for years. It's got a ton of spare input/output to control other systems on your bike as well, which I think you'd find handy. You could automate the launch lever for instance, or setup tables to compensate for fuel trim vs EGT which might prevent another piston seizure event. Additionally I think it's recommended that you use a mass airflow sensor to control the fueling, as you won't get a great signal from a MAP to tune it via speed density on a 2 stroke (or any single cylinder) with all that pulsing going on.

[@ferrumignis]

MAF sensors also have problems with reflected pulses bouncing back and forward through them, four stroke engines running long duration cams can be quite problematic. I suspect Alpha N is the only thing that's likely to be useable on an extremely peaky two stroke.

[@IFRYRCE]

@ferrumignis very true, but its minimized somewhat by putting it in front of the throttle body, especially if you can put a long intake pipe on it. I think MS also allows for a blend of alpha N and MAF or speed density as well based on RPM, although it's been a long time since I looked into that.

[@obi-wancannoli5858]

Look at snowmobile cvt systems. Very similar, but no extra clutch. Everything is handled by the belt. Simpler, very tuneable, can handle enormous power. And if we'll adjusted, will keep the engine at peak power rpm from engagement to maximum speed.

[@snocrushr]

A CVT or variator is only about 80% efficient even when it is tuned perfectly . Consider it has to work against spring pressure and and friction and where you have very little power to spare I think you would be much better off with a manual mechanically or electronic over mechanical operated variator. The reason they are on everything is that they are a cheap to manufacture automatic transmission. As for fuel injection how about 2 separate FI systems, one for low speed and one for high speed or one for gas and one for nitro. The big car drag racers use 2 and even 3 sets of injectors and control systems as well as a Nitros system sometimes.

[@geemy9675]

@@snocrushr agree gearbox is the way to go. whatever torque lost by not being exactly at peak torque when shifting means can be gained by having less parasitic losses. also gearbox means you can push each gear as far as possible so that after shifting you still have enough torque to go through the next gear. plus if you are ever stuck below the powerband and the bike cant rev, you always have the option to use the clutch. variator not perfectly tuned and paired a very peaky engin will leave you stuck with no alternative method. if you add the carburation issues that can happen with the specific conditions of flat sands you could be stuck with very sub par runs and need to retune the bike for each run when the bike with gearbox could just keep doing more runs and focusing on less parameters to tune. I think EFI could help a lot if you can premap a bunch of different conditions then the engine can run on point for every run and all you have to do is focus on gearing

[@jimdale9187]

@@geemy9675 Agreed, however, if you gear a little optimistically, and are very close ratio on your tallest gears, then you are done with gearing, and just run whatever gear the bike will pull. The real problem is tuning to the atmospheric conditions, which are only available at Bonneville. Agree with pre-mapped EFI, and then tweaking the parameters as needed.

[@geemy9675]

@jimdale9187 yeah especially if your not a whole professional team like Bidalot, you want to have as little tuning to do on the field. if it's carburated you probably can't get around carburation

[@springford9511]

"A CVT or variator is only about 80% efficient". Hmmm. A 14hp 125cc scooter then would have about 3hp going into the transmission. Say 2kW. I think it would get pretty hot. Does that happen?

[@jimdale9187]

@@springford9511 The CVT is completely covered by a protective shield, so Alex would only be aware of excessive heat if there was damage to the belt. Maybe bench testing could determine the magnitude of inefficiency.

[@BillWrites-t2e]

A single varaiator seems like more weight and moving parts than any benefit you may gain over a double varaiator and a fixed motor mount. My mod sled had stiffer springs in the secondary and lighter in the primary. The lighter primary gave it some revs before engaging and when it did she had bite because it was in the powerband at launch. The stiffer secondary made the clutch open slower keeping my gearing in the low end longer so I had peak power before I was going 100mph, this made my sled feel so rowdy with peak power around 60mph and a steady climb to 100 beyond that . I would imagine your looking for the opposite effect, I was set up for hill climbing and acceleration . I had a 1000cc arctic cat with about 200hp and it felt like a rocket after the clutch mods. Seems like for top speed you would want the primary to grab right away and change slowly to get you upto speed as you get into the powerband . I would imagine that you wouldnt want much change to take place on the pulleys until your in the powerband with a small engine . The secondary springs need to match the movement up front, running the setup and watching the clucths move is the only way I know how to tune these. Ratchet stap to a rafter and braaap braaap until they move together . I used pure guess work

[@Petrin_Paja]

Hi. I'm commenting for the first time even though I've been following your channel for how many years😅 as a tip, if you use EFI. use such a small nozzle in the first step that it is fully open at a certain load, and you will get an even fuel flow. and the second nozzle then starts working in addition to it later when the load is at its heaviest and gives more fuel. in other words, use two-stage injection control. really easy job.

[@simenthorvaldsen]

2 videos in the same week, this is our lucky day 😊

[@niemam5825]

Mount regular EFI system but... with two, separate fuel systems! One for gasoline (for idling and slower speeds) and second for CH3NO2 (for wide open throttle). Each system should have own pump, fuel pressure regulator, fuel lines and injector. Mount small gas tank (1L max.) with small fuel pump (from 50 ccm 4-stroke mopeds) and second, bigger tank for more spicy fuel. All on single, common throttle body. Everything you could control with regular Speeduino, Ecumaster or other standalone system. You elliminate all problems with too low or too high vacuum in carb and avoid fuel foaming!

[@tomboxyz5564]

With EFI the dual fuel would not really be needed, EFI gives a lot of flexibility and precision, it should be able to start and idle on nitromethane just fine, his engine can also just idle at like 3000 or 4000rpm and that wouldn't be an issue. One way to handle idle would be staged injection, 2 injectors, on idle one injector does all the work, as you get into higher load and higher RPM the 2nd injector starts to help, it's built into speeduino already.

[@niemam5825]

@tomboxyz5564 Yeah, but nitromethane is REALLY expensive fuel in whole Europe and you need special permisions from government to buy it, at pure form. It is expensive even as fuel for rc model engines. Nitromethane wouldn't be useful when warm-up. About 12 yrs. ago I was able to buy nitromethane in Europe but it cost me over $20 only for one liter. Now it's impossible. EFI system got one more safety advantage - there's no chance to runaway engine caused by glowing spark plug ground electrode.

[@tomboxyz5564]

@@niemam5825 since he most likely will use an electric coolant pump, the first stage of warming up can be done with a small electric heater (powered by an external battery, no need to carry that weight once it's warmed up) and EFI also allows for flex fuel, so it's possible to have 2 tanks, a switchover valve, 1 pump, 1 regulator, 1 or 2 injectors on the same line, that lets you start on gas, run on nitro and purge everything with gas afterwards very easily

[@niemam5825]

@@tomboxyz5564 Purging whole fuel system between gas and nitro could be a nightmare. Gas (teoretically) is non polar, but nitromethane it is. Second thing is - can we implant gas fuel system to nitromethane? I know that sillicone tubing works well with nitro, but quickly degrades in contact with gasoline. How with other plastics? Once upon time I forgot to pour "after run oil" to my rc engine... and I left it in a couple of months. After this time crankshaft won't even budge! Engine inside was heavily corroded. How with fuel pump and injectors? Third time - oil solubility in nitromethane. Ester molecules in synthetic oil are non polar, nitromethane is polar but it mixes with something non polar like diethyl ether. Maybe engine must have external lubrication system?

[@tomboxyz5564]

@@niemam5825 One of the reasons for purging is that a lot of the materials acutally do degrade in contact with nitro. If you keep the fuel lines between the valve, regulator and injectors short then the non-mixing is not an issue, you might have to rev it when switch over and it might run rough for a moment, but that's it. You can also go for 2 regulators and put the valve right before the injectors. This also kinda deals with lubrication, you might have less lubrication when switching over, but since that happens on low load and not at very high RPM, that's minimal wear. A seperate lubrication pump like on some factory engines would also work, there's electric lubrication pumps too, which could probably be controlled by EFI

[@LHJhamselv]

Hi. about the "efi" i was crunching some googled numbers, 50cc @ 2000rpm (asuming 100% cylinder fill) you are pumping 6m3 of air. Pure nitromethane needs 1,7kg of air pr kg fuel. equals apx 4liters of nitromethane pr hour. the formula is linier so @ 20000rpm you neede apx 40l of nitro pr hr... maybe its too bulkey but you can get nozzels from oil furnice from 0.2gph to 28gph @ 100psi(7bar) i don't know if you can get the spreed with one nozzle, but if you pump can deliver you can add a 2nd nozzle on a solonoide valve and ajust the speed/pressure until you get the deciered mixtur during the rev range. i think the controller could be relativly easy made with an arduino, but mind a solonoid makes "noise" when de-energiesed just like an ignition coil so use a fly back diode so you don't get interferance in the arduino circut. I'm not to familiar with nitro, so plz make you own calculations, after that maybe a fuel flow test, to dial in the pump idel speed an pump to rmp relations. just my 2cents. Chin up, love your family, and then crack on.

[@mikeseventhree3827]

Why not have one end of the spring not on a fixed point. Maybe controlled by the launch lever instead. That would give some control in the equation of roller force + spring tension + torque = rpm.

[@anzekranjc3312]

As far as I understand CVTs the problem isn't the clutch grabbing before the wanted RPM, it can grab way lower and it will accleerate to the set RPM like it would with a fixed gear ratio, then the RPM will stop climbing and it won't change because the rollers/weights don't produce enough force to overcome engine power. That only happens if you run a stock rear pulley on the torque drive, which has a drop-off in the slots that change the torque spring resistance on the belt throughout the belt travel. This can be fixed with installing a rear pulley which has linear slots. RPM changes throughout the belt travel can also happen if the front pulley (this mostly happens on variators themselves) isn't flat anymore and is worn out. Also the ramps in the CVT don't effect engine RPM as much, they have a bigger effect in how responsive the variator shifting is (when you're coasting and you blip the throttle for a second, how quick the RPM rises and lowers back down, the contra spring also has a pretty big effect on this but I won't go into that) and also has a direct impact on wheel speed acceleration throughout the belt travel. The steepness basically controlls how much force is required somewhere in the belt travel range to accelerate. For example a powerful modified 2-stroke which produces north of 10hp can use steeper, more linear ramps for quicker acceleration in the low speed range and in the high speed range the ramps don't get much steeper so the acceleration is as linear as possible. As for a 4-stroke scooter with an engine that isn't nearly as powerful as a 2-stroke, this one can benefit from ramps that start ascending very slowly and at the end shoot up. This makes it so that the engine doesn't need as much power to move the variator at slow speeds and can still accelerate at decent rate, but for the high speeds the acceleration bogs down. This is better because the engine wouldn't have enough power for much more top speed, but at least the acceleration at slow speeds is decent. Just my 2 cents, sorry for the lack of proper english and excuse me if anything I posted here is not 100% correct, I'm speaking this from my very limited experience.

[@Paulman50]

Rotec Industries in Australia make a carb just like you describe, like pissing into the wind, it also has a variable mixture that can be altered on the fly. It was designed for use in planes.

[@landroveraddict2457]

An electro-mechanical speed variator where the pulley position is driven by closed circuit steppers. Could be programed to any power curve you want and could use feedback from both engine and wheel RPM.

[@bradzartman761]

As I understand it, chain drive transmits approximately 98% of the power it receives, and even a toothed belt on gears doesn’t get that high. A shaft drive has the highest losses and drops below 80% if I remember correctly.

[@jimdale9187]

Would anyone like to design a Derailleur system for 420 motorcycle chain?

[@laburette]

A " normal " variator like the spx have swing point about it have extra heavy load ...this is why we add extra load mount at 90° ( périphériquly), to counter wheight the Force 😉🤜🛠🤛

[@teropiispala2576]

I was just about the write I'd make electric control, but then you said it. It has always been fascinating me, and I'm going to build one to my lathe.I'll probably take easy approach there by making two handed threaded rod to pull and push plates closer and further, applying the force through bearings. There's no hurry in this application, and it should work fine. For faster control with less force, something more clever would be needed. Maybe something other than centrifugal force to move the rollers between the plates

[@noenlunde7028]

Very good, now you are on the right track. The simpler, the less room for error 👍

[@HighDefBNG]

Great explanation! Thanks 2t S

[@SFearox]

You want a scooter to always be in the powerband from 0 to top speed. I had a yamaha aerox btw. Then a Honda NSR

[@lithium534]

I remember when I raced my scooter that there was an electric version that you could program positions for your variator so that the scooter would have gears. I'm sure that now a days you could program a controller to keep it at a selected rpm. 25 years ago arduinos did not exist.😅

[@teltwosheds2972]

Chuck a planetoid gearbox in the rear hub to give more variations to aid the primary variator 😊

[@zacharygillette7810]

Your exactly right, i have one of the top competitor 100cc 2t scooters. Cvts are ser up to constantly gain rpm, my scoot gains 1600-1800rpm from launch to 8th mile. Launch as 13000 and end near 14800. Cvts are what allow these little engines to do 6.8-7.2 8th miles. I know in the United States theres almost no scooters that will run with me, ive been 1.61sec 60ft 3.48sec 0-60mph 4.51sec 300ft 4.63sec 0-70mph 4.80sec 330ft. Havent made a full 150m or 201m pass yet.

[@oliverscorsim]

Launch leavers work but for land speed iv found in my limited experience that gears are the way. Now i cant articulate why but i can say its what u saw the most of. I have a bike with a short powerband up high and have found a bleed valve inline with a hydro clutch gets me off the line consistently then i flat shift from there. I think for you cvt is part of the solution but i think you need to keep it simple and add a 2speed transmission with a manual clutch after the cvt with two very different gears one for take off to 50 then a high gear for 50-140 the cvt should keep it in the groove but the gear spread should work out better and be much simpler while giving you control that i think youl appreciate. Then using a wet clutch wouldn't be very hard either dealing with the heat of high rpm engagement.

[@finnsk3]

Interesting. I would just use an actuator to swing the engine using an Arduino. Set a target rpm and adjust the actuator based on the delta. You would just need an actuator with enough torque and speed.

[@Mat-u3d]

id suggest a linear encoder and a stepper motor to control tension on the engine spring. should be controllable with Arduino and could even program the launch control into it.

[@OddsandEndsMachining]

Have you considered using a carb with an electronic power jet and adapting it for your need? You can control fuel quantity by pulsing the solenoid for the power jet.

[@Reiner_Markenfreund]

Flachschiebervergaser eignen sich gut als Luftmengenregler. Allerdings ist es nicht so einfach, die passende Kraftstoffmenge zu bestimmen. Betrachten Sie die Membranvergaser von Tillotson oder anderen Herstellern, die für räumlich frei drehbare Vergasermotoren verwendet werden. Wenn Sie den Flachschieber beibehalten und die Kraftstoffregelung der Membranvergaser hinzufügen(aus zwei Vergasern einen bauen), lösen Sie bereits viele Probleme. Modellflugzeuge werden oft mit einem Drucktank betrieben, der den Kraftstoffbehälter durch den Überdruck eines Zweitaktmotorsystems auf einen idealen, leistungsabhängigen Druck bringt. Dies erspart ein Pumpensystem und funktioniert bei richtiger Abstimmung perfekt.

[@noreverse1152]

AC compressor clutch, electrically controlled? I should not burn that easy and my guess is that it can handle the torque and rpm is not an issue. Unrelated, the centrifugal clutch with rollers has less hysteresis compared with the one with swing weights

[@kevinsellsit5584]

I think the AC clutch would require a significant amount of weight both from the battery needed to provide sufficient current and from the copper winding sufficient to hold the power, not to mention the iron itself.

[@tomboxyz5564]

Automotive AC compressors don't run at such high RPM and they are in the 1kW region, so not even remotely close to the power and torque of his engine. You also usually don't turn your AC on at redline

[@Reiner_Markenfreund]

@@kevinsellsit5584 In einem Frequenzbereich von 250 Hz bis 300 Hz ermöglicht ein Wechselstromgenerator und ein Wechselstrommotor mit umschaltbarer Phasenzahl und Stern-Dreieck-Umschaltung ein sehr effizientes und leichtes 16-Gang-Getriebe. Es wird lediglich ein rotierendes Polrad als Generator und ein weiteres als Elektromotor benötigt, während die Wicklungen unbewegt dazwischen angeordnet werden können. Mit Bipolar-Kondesatoren kann der ideasle Arbeitsbereich sogar perfekt optimiert werden, dicht bei COS 1.

[@1-3645]

A variator with a belt have a very high friction (power turning to heat) that will cost you a couple of horsepower, an alternative might be found in your Rotrex supercharger. If you look inside it you will find cogwheels without cogs (i dont remember what that kind of transmission is called), that metod to transfer power can also be used to make a variator ( i have driven a big lawnmower that used that principle to regulate its speed) it has low friction (after hours of driving that lawnmower the variator was barely luke warm).

[@1-3645]

Friction drive with perpendicular axes was what i was tinking about.

[@rootbeer666]

2 STROKE DIRECT INJECTION! (It's a thing!)

[@bandedsteels1739]

Maybe a wot button. Wide open throttle with a constant fuel pressure. Set the duty cycle on pressure return changeover solenoid. Don’t take the ball and spring check valve out, putter stronger spring in and run higher pressure and it will atomise better👍 get in touch for water/nitro meth injection advice👍

[@CrazyTony65]

I don't know how to accomplish this but, I think you need a lockup variator, either manual or, if you can figure out, automatic. This way you could hold it at a specific RPM until redline, they lockup automatic transmissions in drag racing, there has to be an advantage. The lockup in drag transmissions is when the engine is still climbing on torque/RPM, not right at peak.

[@Leroydv97]

You should use a pulsed injection. You can control the volume by the injector open time. Still don't really understand why you want a continuous spray.

[@hunnybunnysheavymetalmusic6542]

A hydrostatic transmission would resolve your issue... But you might would want to get one which is made with infinitely variable geometry on a slip ring. It is both lighter, and can handle more speed [RPM] and really good horsepower at such a low mass as you are working with.

[@henningsea6094]

Hi, would it be a possibility to somehow control the distance of the variator disk distance electronically with a servo or so?

[@turbohoolik]

One suggest for fuel system. Why not to use solenoid valve to adjust fuel pressure like diesel cars are doing in common rail system? It would be easy to control with arduino and you don't have to have to create something new.

[@murraymccann1417]

I don’t remember the numbers, but I know that cvt’s get less efficient as rpm goes up. Yamaha got around this on their 4 stroke sleds by putting a gear reduction on their clutches, so the clutch spins slower than the crank. 18000 seems really fast for a belt.

[@davidcoates4852]

I have seen belt pulleys with a two speed automatic planetary gear box inside them, that operate with a centrifugal clutch, used in industrial washing machines.

[@mikevincke1136]

I am a scooter racer. The variator switches depending on the engine speed and not depending on the torque.

[@eugenelang1596]

Mikuni burco 34 mm and Tillotson carbs are basically mechanical injection

[@neilbathe]

There is nothing I can contribute to the discussion, but WOW...

[@ASoftaaja]

Is it really the engine torque that is creating the opposing force to the rollers? I have thought that the opposing force comes only from the contra spring/"return" spring and engine torque should have no effect in the rpm setpoint. However, in a normal scooter CVT, in the rear vari there is usually a spiral helix thing which causes the sheave to rotate at the same time as it compresses. That is actually a torque sensor and as far as I know, that will create an effect that is dependent on engine torque. Basically the more torque the engine has, the spiral will automatically create more resistance against compressing the return spring which in turn will raise the RPM setpoint. This makes it possible to have the revs decrease when you let off the throttle. It is quite surprising that the variator actually shifts to higher gear when you let off the throttle. On 49ccscoot channel there should be a video about this. I actually think that the front vari ramp angles have no effect (ha...again the classic "no effect"). If the angle is shallow, it just means that the rollers will move "faster" over the ramp. But then the amount that the sheave moves will also be less. So the net effect is "no difference". If the ramp angles made a difference to the behavior, I suppose that tuning the variator would be far more difficult and we would see a large selection of different ramp plates. I have not seen ANY specifications of front vari ramp angles. But the shifting behavior does exist. I think it is mostly due to belt slippage and the angle of the rear helix. I have seen those helixes that have 2 different angles. When the angle changes, there will be a "shift". Then the belt may slip on takeoff which tends to cause an overrev at first and then it "shifts" lower.

[@2STROKESTUFFING]

I didn't really give this much thought until the runs at Bonneville. Down jetting for better AFR made rpm rise. Engine torque definitely has an effect. More torque makes the shift point higher with the same weight/spring combo. If it didn't matter a set weight/spring combo would give the same rpm in any engine(and variator tuning would be a breeze).

[@ASoftaaja]

@@2STROKESTUFFING I asked this from ChatGPT and it seems to agree that torque should not primarily effect the rpm. But it suggests that there is a secondary effect: "Torque Feedback and Shifting Dynamics: This torque feedback indirectly influences the shifting behavior because it resists or assists the movement of the pulleys. Under high torque conditions (e.g., during hard acceleration), the increased tension in the belt can delay the shifting of the primary pulley slightly, allowing the engine to stay at a higher RPM longer. Under lower torque (such as cruising or light acceleration), the primary pulley can more easily achieve its intended shift RPM, which results in smoother, earlier shifting to a higher ratio." The helix in the rear pulley definitely works like this but sounds like the same effect could be there also without it. The machine brain didn't talk about the helix at all. Maybe when the belt is under more tension from torque, it is harder for it to slip upwards on the front pulley. But primarily the movement of the variator should be rpm controlled and I don't think there is that "getting stuck" to wrong side of the torque curve that you mention.

[@2STROKESTUFFING]

Thanks for the input! Makes sense that torque doesn't affect shifting much too, as otherwise half throttle would lower rpm vs WOT, and it usually doesn't. Our Bonneville runs were constant WOT/hard acceleration, could be why torque had a huge impact. Jetting from about 900 to 1100f egt gave a 2000rpm rise.

[@simongroot7147]

If you plan to go and test the land speed bike in the USA in December or January, Colorado and Bonville will be really cold. Warmer than Norway but much colder than speed week. This could upset your variator tuning and any engine tuning.

[@mikestaples9044]

Free jazz! Ludicrous ideas please! I can find stuff that works anywhere. Love it

[@Curtislow2]

6:38 reminds me of the governor system on a steam engine.

[@enestekin2512]

the most of turkish moped (mobylettes and peugeots) racers use stiffer springs and presses exhaust with their feet and locks the engine its place while gaining speed, when it goes fast enough they relaese and gain 10 or 20 km/h. idk how it works but if it works it isnt stupid

[@stephencory2904]

I am in similar single person dev position ( albeit for eco product ), and i love this site....however i get confused by the approach. There seems to be a lot of jumping about e.g Bonnevaille records, most powerful 2-stroke, variator, dyno builds etc and now fuels system for another old bike/ engine. Even Elon Musk rapid dev/rapid failure doesn't go for so many different directions. I think a bit of focus might help ... eg. - Target Bonni 202x - Build Engine - but put it in frame same as Bonni ( surely hacking an old farme is more diffilcult than replicating the elegant design you had and will reduce flights back and forth to US /) , - dont waste time with old bike / hack - focus on engine molding - tune 1 variable at a time and records results - build fast/fail fast but kep track of what works/doesn't ... anyway you are still inspriational, let me know if you are coming to UK and i will buy you a beer.

[@jasonbirch1182]

Have you considered Bosch CIS K-JETRONIC or at least some parts? It's Constant Injection System One injector can probably handle your power needs. The fuel distributor can be plugged off on the circuits not used or you could probably machine off the rest you don't use to make it smaller. It weighs air with a mechanical flap but I would assume you can simplify since you are going for constant rpm.

[@MikeyAntonakakis]

Centripetal force = m*v^2/r. Axial force on the variator = centripetal force * sin(ramp angle), at least off the top of my head I think it's sine. The ramp's increasing angle is primarily to account for the larger radius as the variator actuates. If the variator was designed and tuned to hold a constant rpm throughout actuation (which would have a constant engine torque), the ramp angle needs to increase to maintain a constant axial force. Then deviations away from that ideally-increasing ramp angle can be made to account for the increasing spring tension, achieve a constantly-increasing rpm as the variator actuates, etc.

[@MikeyAntonakakis]

That said, I think you are overthinking single-variated setup. If you want the best acceleration, tune the variator to actuate at peak power, ideally keeping constant rpm through the entire actuation (more or less). Tune clutch to fully engage BEFORE variator begins to actuate. Ideally at peak torque rpm, which will always be lower than peak power rpm. This way, variator stays in lowest ratio until clutch is fully engaged, then you accelerate a little more in a fixed gear ratio up to peak power, then the variator keeps you at peak power until it is fully actuated, then you rev out until road load power (rolling resistance + aero drag + drivetrain friction) equals engine power. Perfect chain drive ratio would have that happen at exactly your peak power rpm for a land speed run, or just overgear a bit so the variator never quite fully actuates (assuming you tuned it for peak power). In practice on my own 70mph 50cc Peugeot, peak power is somewhere around 11k-12k. Clutch pulley is tuned for about 9000rpm. Variator keeps me around 11-11.5k throughout the entire rev range. Revs out to 13k. There's a video on my channel about float bowl foaming with onboard go-pro that demonstrates the clutch/variator tuning as I ride around.

[@MikeyAntonakakis]

For additional logic on why to tune clutch for peak torque and variator for peak power, during clutch slip you want to be in the lowest variator ratio to get the greatest rear wheel torque and therefore greatest acceleration; a slipping clutch transfers almost all the torque going into it, but not all the power (power = force * velocity so at a standstill you have 0hp coming out of the clutch, and all that power is turning into heat). Once the clutch is fully locked it transfers 100% of the power that goes into it - at that point, you want to get to peak power as soon as possible. With the variator still in lowest ratio, you'll easily accelerate between peak torque up to peak power, then to continue maximum acceleration, you want to variate at as close to peak power as possible, because at any given speed maximum acceleration will be achieved only by being at maximum power (thanks again to P=F*V). To reach the maximum speed that your maximum power will provide, you do NOT want to rev any higher than peak power RPM. By tuning this way, you only really need to make enough power/torque below your peak torque rpm to accelerate the pulleys and overcome the no-load friction, then you only need to make any significant power between your peak torque and peak power rpms. The engine could barely run outside of that maybe 3000rpm range and the bike will still rip (e.g. my Peugeot could barely accelerate the unloaded pulleys, would take like 3-4 seconds to rev from idle to 9000rpm, but absolutely rips off the line and goes quite fast).

[@MikeyAntonakakis]

Last one, I promise: having the clutch engage later, especially after the variator starts to actuate, is no good - if you still manage to accelerate, best case scenario you lose out on acceleration because you're not launching in the lowest ratio. A worse case is the variator finishes variating before the clutch fully engages, wasting a ton of power as heat - like trying to launch a car and accelerate in 6th gear - yeah, with a sensitive clutch foot you might be able to do it without stalling but you will definitely not be achieving peak performance.

[@TerrorTubbie666]

8:33 I watched some scooterdragracing last week. Listening to the sound, they all seemed to stay on the same rpm from start to finish. BUT that was filmed from the startingline. So shouldn't Doppler-effect make the sound of a scooter moving away from you AND accelerating go down if rpm stayed the same? So I guess these dragracescooters do climb a little bit in rpm during a race, exactly enough to compensate for Doppler-effect.

[@Fumacenta2tempos]

Should i tune the variator to run on peak tork or power?

[@adamadamhoney]

Floating above intake trumpets I believe is the best position ie older F1 cars and a lot of motorsport design? Probably need a plenum but that also can be factored into any boost which I'm not sure if that's been addressed before?

[@Mhoeflich1]

Could a varatio working use pneumatic pressue to counter the belt? this could be achived by transducers meaxuring torch and the air pressure to the cyclinder controlling the belt length.

[@HulluJanne]

I bet that SPX style is much more linear, or has a specific behavior that can't be altered much. When comparing to the scooter variator, the slope of the roller ramps determines much of the behavior. Back in the early 2000's I bought a Malossi Multivar 2000 for my Gilera DNA and the common forum posts all said that it will not gain or change anything if not going for bigger power or 70cc, and even then it could be a disappointment because Piaggio's variator was said to be highly optimized and many didn't see any benefit in changing it out. I didn't exactly believe the talks and bought one. Right after I installed it I noticed that even with the same amount of weights the RPM's went noticeably higher immediately after takeoff and then came down a bit. It really took of from a stand still. The original was way more linear and I'd say even sloppy in comparison. I'd guess the benefit of the SPX style may be quicker response and possibly lower weight. It also looks like cheaper to manufacture...?

[@mikemelbrooks]

I would have thought an Arduino hooked up to the tensioning spring via a relay would be able to make sure the engine stays at or near peak torque/ peak power?

[@Fumacenta2tempos]

What should i do when the belt dont travel all the front variator long but goes all the way in the rear one?

[@danowens7145]

I think using direct injection of methanol and nitro. And use a carburetor for regular port fueling.

[@panikrev175]

Is it out of scope to suggest a camshaft-driven injection system? Would it work? - like the kind you’d see in old diesels? Run it rich and as the air volume increases with throttle it’ll reach a more-stoichiometric ratio near WOT, without variable cam timing? Then you can fine tune with ignition timing, would that work?

[@billybroughton5335]

This might be a dumb Idea but maybe you could use a regular carb and somehow pressureise the bowl so it acts like an injector 🤷 you could throttle it by changing the pressure.

[@Tunerfreak99]

Perhaps you could use a direct injector from a direct injected gasoline car. They can be fine tuned and handle very high pressures

[@jonhoyles714]

maybe some efi map of power curve coverted into some electronic clutch to grab at set limit in software not sure how you would do that but worth a go :)

[@renedejong6863]

and there is some thing else ..... you can set the spring tension from the rear clutch ,,,,, so they grip ,,,, much later , than original that are set too grip at 1000 rpm ,,,, ..... also the newer , rear vario's have a removeble cage ,,,, were in the pins slide ,,, they have a other slope , working on the(rear vario) spring tension ...... but race-track scooters ,,,, dont have that top-speed you are looking for.... (150 +) .... i think a 2 vario system would give you more range to fidle with .......

[@jammi__]

Still sticking with carbs is pointless. Just go with EFI, there's plenty of explorations there, and there's nothing you'll do better using carbs, and you don't really have an option if you want to do boost and such, and trying to reach anything close to optimal operation in situations where the parameters change constantly while the engine runs. If you want to experiment, try variable volume intake things applied to variable volume exhaust, electronically controlled, of course..

[@TheCunneen]

Just put injectors in a throttle body!! You shouldn't be scared of EFI its much simpler once you get your head around it. It takes the guesswork out of using a carb

[@rosso4122]

I didn't ask because I didn't know to ask. This is very interesting, thank you!

[@MrTamiyacowboy]

if we could somehow control the torque so its only at max with max rpm it would work. i get where you are coming from 2SS. i loved messing around with my variator on my lil 50cc , i ended up with a mix of weights i was running 3x 3.5g and 3x 4.5g . the lower weight rollers gave me a nice quick pullaway but would max out then the higher weight rollers would take over. i was able to maintain around 9k rpm throught out. i am wondering if it could be something you may look into tuning the weights for your need. if you go this route you need to stager the weight placement in the variator ie 3.5g then a 4.5g so you keep the weights balanced around the push plate .

[@blainemacdonald6929]

snowmobiles have mastered belt drive cvts mine engages at 5000 and sits at 8100 perfectly until it runs out of final drive ratio

[@FJB752]

That is a good point, CVT is probably NOT the best for land speed 5 mile runs. A snowmobile is the LARGEST user of CVT and now SXS vehicles. A 1000 hp snowmobile just went 202mph in 1000 ft on ice. CVT works good for short fast Excelleration. I think a gearbox is the way to go. Pretty soon you will have build something to see how fast you CAN go. Beat the record then try to beat your own record.

[@MarkDaleADV]

Burgman 650 has an electronically controlled cvt, maybe parts could be utilised?

[@adrianzarate31]

try to check yamaha mio aerox turbo electronic variator pulley

[@Kowalamaster]

What if you took the ramped variator with the weights, and made a ramp into a basin (dip) to match the Power curve rather than just a incline curve? so it would grab still go up to the power, and as it crests the torque curve, it falls in the valley / basin and holds there or at the edge /end of the Variator casing. Granted I do not know much about variators though. Like my brain says that it would go up to the crest with the torque but then dip but be held at peak power as it produces a little bit less torque so the inverse valley holds the weights there? No idea if it would be possible to make with the flapper weighted one, but the ramped centrifugal one I presume may work? I do not know if I have explained it right, Though I am sure people can tell me.

[@doncairns9275]

Keep ploughing your own furrow. It's because of people asking questions like "what happens if we" that we have heart surgery, nylon stockings, and cheese. Keep it up. You're an undiscovered hero. 👍

[@rapturedreamvision7205]

Would a rotary valve give you the fuel pulses you want with a carburetor?

[@snakezdewiggle6084]

🔵You could make 👉Two Variators👈 in series.

[@Mybeardog]

Thanks again 👍

[@richardlincoln8438]

Personally, i would be pleased to see a reliable engine producing results on Your dyno before You turn it into another Rube Goldberg device Alex. It's entertaining to watch. Best Wishes to You, Your Family and Friends.

[@alexanderbonardi4514]

Why not adapt an automotive injector? If you choose from a direct injection model with appropriate CC you will have the ability to generate a good spray pattern (wherever you install it) using high pressure fuel (usually somewhere in the 100bar region, but will work great with as low as 20bar). The injector coils will work with 12v and a PWM waveform ( you could adjust this to engine load)...just saying

[@hanssimonen2677]

Whats the brand for the "pully" Variator?

[@arjanschild3552]

Instead of a tensioner spring why don't you use an actuator that you control by an arduino and a self defied algorithm?

[@jamest.5001]

So the clutch would need to grab above pulling down the engine allowing it to climb back to the power peak. Why not just use gears? Use two 5speed gear sets. With a 2:1 chain drive between them skipping between them shifting, 10 gears. It should keep vrhe enging close to between peak tq and peak power! (Possibly, its going to take more figring than i did.. i would much rather use gears!!)

[@Voting-does-nothing]

Put a turbo on it already mate ❤️👍🇬🇧

[@MarcoPolloII]

Will a magnetic clutch work?

[@Reiner_Markenfreund]

Ja, wenn der Stator eine eisenlose Zwischenwicklung zwischen zwei Polradteilen aufweist, ist das sogar besonders vorteilhaft. Der Stator könnte auch aus Silberrohr, Kupferrohr oder Aluminiumrohr gewickelt sein, das im Inneren flüssiges CO2 als Kühlmittel verwendet; der entstehende Gasdruck könnte über eine Turbine dem Antrieb wieder zugeführt werden. Ein Wirkungsgrad nahe 97% ist dadurch möglich.

[@linusschollin3389]

Har lite prylar liggandes, spjälhus och spridare från 15hk Mercury om det kan vara något?

[@MaNNeRz91]

Can't you do a crankcase injection system and then just control the air flow btw?

[@mociczyczki]

imagine cvt l” discs „ change radious electronicallly like normal sprocket with chain be some hydraulic etc system even if there be hole on disc circle betweetetth which catch chain it be still work just fine to some rsnge of rsdious change beside there not need to be holes as some internal other teeth csme out as rsdious wiill increaseing asf

[@jodymorris1072]

Check out the fuel injection on the Honda Trail 125?? 🎉

[@BricoleurTV]

What’s wrong with gears?

[@ukaszszynkowski8960]

👍🏻

[@landroveraddict2457]

Beaten paths are for beaten men. Keep going.

[@amirbasssystem]

❤