Need help, everyone...  By  upfartoolate (Read 1,737 times)

What kind of engine are you building? I have only heard of Desmodromic valves on Ducatis?

I think I'll call it the Kymcati FrankenYager. LOL

It's a Honda GY6 derivative, a Kymco SJ40, apparently. It's factory-overbored to 62 mm, with the usual 57.8mm stroke (174.5cc). Fuel injected, liquid cooled. Rev limiter is 9200 RPM, I'll be extending that to 11,000 RPM after I get an offset conrod made, and the rotating assembly professionally balanced. The piston will be WPC treated over its entire surface (to toughen it and reduce stress risers), then ceramic coated on the crown, and WS2 coated on the skirt. The increased piston weight due to the ceramic will be partly offset with a titanium wrist pin (also WPC and WS2 treated).

The main purpose of the build is to get frictional losses as low as possible, and to get mechanical efficiency as high as possible. So an offset connecting rod will help increase mechanical efficiency during the power stroke, the Desmo valve actuation will get rid of having to compress those valve springs, hybrid ceramic MicroBlue bearings will reduce friction, WPC treatment and WS2 coating of rings and cylinder will reduce friction while increasing ring seal, gapless Total Seal top ring will reduce blowby, electrical system mods to reduce electrical waste (ground shunt voltage regulator, chiefly), etc., etc.

I'm even going so far as to put mag-drive electric coolant pumps on it. The microcontroller will vary coolant temperature based upon engine load... hotter at light load for more efficient operation, cooler at heavy load to prevent heat problems. The pumps have the ability to circulate all the coolant in the system every 2 seconds at full speed with both pumps on, so it'll be able to tightly regulate temperature. It'll monitor head, exhaust and coolant temperature and kick on both pumps and the cooling fan in event of an overheat on any of those, so the engine's protected from melting down. The microcontroller and pumps will take ~22 watts with both pumps at full tilt (the engine is tiny, so the pumps are tiny), which replaces the OEM pump which can take upwards of 1/4 HP at WOT.

Combine that with micro-polished and WS2 treated rear gears (7.15:1 to replace the OEM 8.408:1 gears), hybrid ceramic bearings in the wheels and rear gears, and a sprag clutch in the rear rim, and this bike should be able to get up to speed and coast for quite a while.

After the bike's as mechanically efficient as I can make it, I'll design an aerodynamic body for it along the lines of the Akira bike, but more fully enclosed.

At 11,000 RPM with the new rear gears and aerodynamic body, the bike should be able to reach 112 MPH. It'll rarely see that speed, most likely just once for testing (that's insanely fast for a scooter)... the point of the bike is to try to reach 150 MPG at highway speeds.

Eventually, if I can find a toroidal IVT (infinitely variable transmission) transmission, I'll swap that in to replace the CVT. That'll allow me to get up to speed, and keep the engine in its most efficient RPM range for high-fuel-efficiency cruising. The IVT will be controlled via a twist-grip on the left handlebar. A side benefit is that since the IVT has "geared neutral", when the bike is stopped and the engine is running, the bike can't roll backwards when stopped on a hill, even with no brakes applied. Due to the sprag clutch, it'll be able to roll forward, though, but that's not so much a problem as rolling backwards. My legs can hold the bike from rolling forward, whereas my feet just drag if it rolls backward. So that little annoyance will be eliminated. I hate sitting at long lights on steep uphills, having to keep the brake on the whole time. I like to be able to take my hands off the handlebars and relax a bit. Another side benefit of the IVT is that I can tailor the performance on-the-fly... if I want to get the engine screaming and take off fast, I can. If I want to keep the engine in its most efficient RPM range and take off slowly, I can. All with just a twist of the wrist.

All the hybric ceramic bearings have been delivered, the new micro-polished and WS2 treated rear gears have been delivered. I'm just waiting on one needle bearing to be delivered, and I can swap in the rear gears, and the hybrid ceramic bearings on the rear gears and wheels. There are no hybrid ceramic or ceramic counterparts for needle bearings, so I had to use conventional needle bearings.

I didn't want to use the same cheapie OEM needle bearing (it's got a plastic roller cage, and is only rated to 11,000 RPM, whereas in the bike, it can go up to 13,000! No wonder it fails so quickly.), so I found an Italian site that sells the same size needle bearing for Moto Guzzi and Aprilia bikes. It should be a higher quality bearing. It's rated to 20,000 RPM.

So no one knows of a metal fabricator / CAD design house that can create a Desmodromic valve actuation system? Anyone? Anyone?

Just an update... the new rear gears are worn in, and I've put 8.5 grams of tungsten disulfide in the engine oil and 1.75 grams in the gear oil to reduce friction. I'm now running a Pulstar HE1HT9 spark plug which has completely cured the cold rough idle, and on the 12th of this month I installed a Wolverine 9.0 120 volt 125 watt 3" diameter block heater on the underside of the engine, along with a Westek TE06WHB 2 Outlet Digital Timer. The timer has battery backup so it doesn't lose the date and time even when not plugged in for months, and it's got 20 programmable timer events. I've got it set up to turn on an hour before I leave home to go to work, and an hour before I leave work to go home.

With the changes thus far, I've increased fuel efficiency from around 65 MPG (historical average over ~10,000 miles) to 91.172 MPG on the last fuel-up. I expect it'll get further fuel efficiency bumps with each additional modification I make... reaching 150 MPG might be easier than I expected, but I'm going to see how high I can push it.

I also found a guy who's invented an IVT that'd be perfect for scooters like ours... he's named Tom Troester, and his IVT is called the Telam IVT. It's small, light, low frictional loss, can handle loads of torque, and is a geared IVT so there's no slippage. I just have to convince him to custom-build one for me... that's what his company does, so hopefully he'll take on the project. I'll mount the IVT on the engine shaft, and transfer the power to the rear gears via a flat cogged belt so there's very little friction. The IVT gear ratio will be changed via a twist-grip on the left handlebar. I'm not sure how much Troester's IVT costs, I've put an inquiry in to his company and am waiting for a reply. Whatever it costs, me want, me get.

I've also bought a Pertronix FlameThrower HV 60,000 volt ignition coil and MagneCor R-100 CN 10 mm spark plug wire. I got the extra-thick spark plug wire for the additional insulation, since I'll be experimenting with even higher voltage plasma discharge ignition later on. I've shipped the coil, wire and a spark plug to my electronics guy, who is designing an isolator (what the industry calls an "ignitor") to isolate the coil's back-EMF from the ECU and tachometer, while still allowing the tachometer to work. The isolator will be something I'll eventually offer for sale, since it lets us run pretty much any coil we want without worrying about blowing out our ECU, and with certainty that our tachometers will continue working as they did with the stock coil.

And my electronics guy informs me the micro-controller for the electric coolant pumps is nearly complete. I should have a bike-installable version within a couple months for road testing. That's another thing I'll be offering for sale, since the only other company offering electric coolant pumps and controllers can't PWM-control pumps as small as our bikes take... their controller just turns the pumps on and off (and once it's off, how does it know when to turn it back on, if there's no coolant flow?!). My microcontroller will control up to two pumps (5 amps each) and two fans (10 amps each), has a fast pulsed-warmup mode (pulses the pumps so the water around the cylinder is warmer than the bulk coolant in the rest of the system), a "ride it like you stole it" mode (if the bike's not fully warmed up, and you jump on and ride, it'll exit pulsed-warmup mode and vary pump speed according to throttle position), operational ramp mode (after the bike is warmed up, it varies pump speed to keep coolant at the correct temperature), an emergency overheat mode (it monitors cylinder head, exhaust and coolant temperatures... if any of those reach their user-configurable limits, both pumps kick on 100% and both fans turn on until the condition is cleared), and an optional run-on mode (the system can run-on for a user-configurable time after the bike is shut down to prevent heat soak). The other controllers only have 6 temperature setpoints and don't monitor head or exhaust temps, whereas with mine the temperature setpoints can be changed in increments of 1 degree F. It's configurable by plugging the microcontroller into a computer via USB, and has a dashboard readout to tell you pump speed (I opted for a simple LED bar graph to show pump speed for this iteration, since when riding you don't have a lot of time to process numerical readouts, but it could have numerical readouts in later versions). My microcontroller can control 2-wire, 3-wire and 4-wire pump motors, monitors pump PWM feedback (on 3-wire and 4-wire motors), and bumps up PWM duty cycle if it senses the pump has stalled. If it still can't get the pump to run, it'll start the other pump... the pumps are usually in a lead / lag configuration, and the microcontroller keeps track of pump run time, switching which pump is lead and which is lag to even out run times between the pumps.

One of the really cool things that can be done with my microcontroller is to vary coolant temperature based upon engine speed... if you're just putting around town at low RPMs, it jacks up coolant temperature by a user-configurable amount to make the engine more efficient. If you really get on the throttle, it'll lower the coolant temperature setpoint to prevent pre-ignition. Because it's got two pumps and two fans, and because the bulk coolant exiting the radiator(s) will always be cooler than the temperature setpoint for the coolant exiting the engine, it will be able to nearly instantly drop the temperature of the coolant around the cylinder by kicking the pumps on high.

Might be easier to get the entire cylinder and head, and just figure out how to fit the cylinder to the block. I'll research what cylinder bores Ducati used. My bike is 62 mm bore. If I use a sleeved mm Ducati cylinder and head with a 8mm thick ceramic sleeve, I could keep my 62 mm bore, get ceramic heat shield on the cylinder wall, and have Desmo. The piston face, head and exposed portions of the valves will also be ceramic coated. The engine will have WS2 in it to minimize friction, so the ceramic sleeve should last quite a while.

One of the reasons I want Desmo valve actuation is that rather than going for a high RPM screamer, this bike is a high-MPG project bike. So we want the engine turning as slowly as possible at any given speed (which is also why I'm trying to locate an IVT, which would let me tune engine speed at any given road speed so that the engine's turning slow but the throttle's open more to reduce pumping losses).

en.wikipedia.org/wiki/Desmodromic_valve#Disadvantages
"With conventional cams, stress is highest at full lift, when turning at zero speed (engine cranking), and diminishes with increasing speed as inertial force of the valve counters spring pressure, while a desmodromic cam has essentially no load at zero speed (in the absence of springs), its load being entirely inertial, and therefore increasing with speed."

So with conventional valvetrains, the slower you turn the engine, the more valvetrain stress there is. Vice versa with Desmo... the slower your engine turns, the less stress (and thus the less power required to move the valves).

The whole goal is to remove as many parasitic power drains as possible to increase efficiency. This is the tack I've been taking for the first stage in the project, and thus far the MPG has risen from an average of ~65 MPG over ~10,000 miles to 91.172 MPG on the last fuel-up, an all-time high.