Building a race motor...

[Som]

Well, I just bought a gen 1 motor. The race bike's motor still works, this one is going to be a side project for the next 6-8 months or so. My goal is to get as much power out of this engine as possible, within a rougly $2-3k price range.

Some questions...

What are the known limiting factors for this motor to run at 12-13k RPM? Bear in mind, at this point, I'm considering all the options regardless of price... so that means if custom camshafts are needed for smoother lobes to prevent lift, I want to know. The only real things that I wouldn't want to *replace* would be head/block. Modifications are expected... porting the head to increase the flow. I imagine there's no playing with downdraft angles (aluminum head, right? pus it looks like there's no room on the head to get any more vertical anyway).

Stiffer valve springs... lighter valves...

Could oil circulation be a problem for the higher RPMs? What about water circulation... would the higher RPMs cause cavitation at the edges of the impeller blades?

Custom timing chains for the higher RPM?

Crank I'd get lightened/balanced. Lighter rods. Pistons... lighter. Not sure if I'll work on maximizing piston travel by customizing the shape of the head... probably not, but could be an option. Cylinder walls lined.

What are the WSMC rules about bore? Someone was telling me that the rules allow for 1mm overbore, even if it takes you over the technical cc limit. Is this true for all 3 SV classes? 1mm bore out on this motor I think brings the cc to 660 or so? I noticed Wiseco only offers standard, 2mm, and 3mm pistons...

I may make my own ignition timing module just to be able to fine tune the timing. Not sure how useful this would be yet, though...

Flatslide carbs... don't know how much room is on the head, or if it would even be strong enough to take another hole... even though it'll likely be a quick dead end I want to look into adding a direct injection port. I'm sure there's a hundred reasons why it wouldn't work, but I'm still gonna look into it. If I end up making my own brain box anyway, what's one more thing to control while I'm at it. :)

This is a learning project for me... so I'm not really looking for responses like "you could pay someone half the price for a better engine". I'll give myself a passing grade just for getting the motor to run when I rebuild it.

Thanks for any input!

- Som

[Tom]

If you want to keep your bike going, your garage might need to look like Zman's from a few winters ago:

http://i137.photobucket.com/albums/q228/twowheeltom/1434337-motors_2.jpg

because you'll be building one of these:

http://upload.wikimedia.org/wikipedia/commons/thumb/7/73/M-67Grenade.jpg/371px-M-67Grenade.jpg

:D

[Kurt'sSV]

Sounds like a hell of a project.

Doesn't your bike still have a stock front end? Build a GSX-R front end for your bike, first.

Looking back I wish I would have bought the GSX-R front end/build before I bought the flats. Horsepower's worthless w/o control.

[leaftye]

He probably just wants to build a strong motor if he's going to build one at all. That's how I'd see it. If I need to spend time and money on a motor, I might as well give a little more to get more power out of it.

[Som]

Yeah, this is more of a learning project than it is an "improve my lap times" project. I've been wanting to build a car motor for years to get practical experience with a lot of the things I've read about... figured a twin would be a much cheaper experience, plus I could put the engine to use afterwards -- assuming it performs.

There are plenty of things I'm going to look at... hell, I'm even going to look into driving the valves with solenoids and ditching the cams. Doesn't mean I'll do it, but I'm going to look into the technical feasibility. If I can't find solenoids small enough to fit, strong enough to push the valves, fast enough to push them at high RPMs, precise enough to allow for close interference tolerance, then I'll stop considering it. :)

$3k isn't necessarily my limit, though it'll probably take me longer to build if it costs more than that.

- Som

[Angel]

Yeah WSMC rules allow for a 1mm overbore. I think bigger 2 and 3 mm over are used by racers in other series that allow 650 twins to go up in size (I think CCS goes up to 700cc)

[svsrevlis]

Som, Are you an engineer or something? Sounds like a heck of a project but I think it would be great for you to keep us posted as you move forward. Especially if you can actually manage to ditch the cams. Does anybody actually do that on bike engines now?

You going to be at WSMC this weekend? I will be there on Sunday to race and get a few more novice points. #655

[odie_trackie]

As a controls engineer with experience with electrical actuators, I think you are going to have a dickens of a time finding anything that can overcome your valve springs and that doesn't even take in the dynamic forces of opening it fast enough for 12,000 RPM.

The current required to activate a solenoid that fast would probably require wires so thick, your engine would gain a lot of weight. Not to mention all the other electrical hardware to generate that kind of electrical power.

Of course an interesting idea is a Desmo-electric valves. No valve spring (except a light one to keep the valve closed when the engine isn't powered, just the solenoid keeping it open and closed. Thought the control system would have to run REALLY fast.

At 12kRPM, it takes the crank .005s (5ms) to make one revolution. Since you would need to control the valves down to 1 degree of crank position that means .005/360 = .0000138s at least.

in order to avoid aliasing in your data acquisition, you will need to sample at 2.5 that = .000006s ->180kHz.
Good luck finding a controller that can do that. Most controllers only work down the the 1ms = 1 kHz range.

If the electrical cams were possible, I would imagine you would see that in MotoGP already or at least hear rumors about it.

I wonder if they make lightened cam sprockets. That might save some power.

[Darth Lefty]

Hey there's an idea, make it desmo :)

[Tom]

Hey there's an idea, make it desmo :)

that's like going from :monkey: to :D and then back to :monkey:

[Som]

As a controls engineer with experience with electrical actuators, I think you are going to have a dickens of a time finding anything that can overcome your valve springs and that doesn't even take in the dynamic forces of opening it fast enough for 12,000 RPM.

The current required to activate a solenoid that fast would probably require wires so thick, your engine would gain a lot of weight. Not to mention all the other electrical hardware to generate that kind of electrical power.

Of course an interesting idea is a Desmo-electric valves. No valve spring (except a light one to keep the valve closed when the engine isn't powered, just the solenoid keeping it open and closed. Thought the control system would have to run REALLY fast.

At 12kRPM, it takes the crank .005s (5ms) to make one revolution. Since you would need to control the valves down to 1 degree of crank position that means .005/360 = .0000138s at least.

in order to avoid aliasing in your data acquisition, you will need to sample at 2.5 that = .000006s ->180kHz.
Good luck finding a controller that can do that. Most controllers only work down the the 1ms = 1 kHz range.

If the electrical cams were possible, I would imagine you would see that in MotoGP already or at least hear rumors about it.

I wonder if they make lightened cam sprockets. That might save some power.

To be honest I never seriously considered keeping the valve spring for any high rev application. If I was going to try multiple attempts at this I would have considered a low-RPM version that only pushed the valves (and kept the spring) and then a high-RPM version that might use 2 pneumatic solenoid valves with push/pull capabilities.

Then again, I'm not sure how big pneumatic solenoids are or how much works is involved in getting them to work, but that's the whole point of the exercise. If I had a surefire idea of getting the motor built with solenoids, I probably wouldn't be posting about it. Haha... :)

As for the controller... 1kHz seems pretty damn slow for an ADC. I can't imagine there aren't specialized controllers for faster acquisition out there... again, all part of the learning process. :)

Thanks for the input!

Driving up to Willows Thursday night... there til Sunday. #768. We usually pit down by turn 1 near the end of the middle paddock aisle.

- Som

[since413]

I've tried it all , give me a call. With a lightened crank, ported heads, total loss ignition, flat slides, the whole enchilada, you can see 95hp that will last 1 season. But you better be a tuner. Or run 80hp on a bike that will last forever, ride hard, win anyway. (Kurt Whittington's race motor is in one of my team bikes still chugging away untouched except for some bits).

[Som]

I'll hit you up this weekend. :)

One other thing I might look into during this whole process is making an inexpensive engine dyno. I'm going to want something to help me gauge whether I'm moving in the right direction or not. Maybe match the chain up to a weighted tire or something... not quite sure what the most accurate way to do it would be, but it would be something I'd want to look into since I won't want to have to put the whole motor together just to get it on a chassis dyno.

- Som

[odie_trackie]

There are DACs that run at 1kHz and up, but they are bench type units and they are real expensive. An automotive rated DAC that can run at 1kHz is probably not that hard to find, but you have to have a controller that can work down there. I think a good Bosch unit that BWM, Audi et all uses would probably work but those are expensive.

I like the budget engine dyno idea, though. The dynos I have seen are usually based upon pumping some fluid around and correlating the pressure to torque. I think it's a matter of finding a pump that has a well documented relationship between pressure and input torque. Then it's just a matter of recording rpm and pressure and doing some math.

Of course you have to come up with a fluid recirculation scheme to deal with the heat in the fluid (which can affect your calculations).

I think you could use automotive cooling system parts (radiator, over flow tank, etc) since they are relatively cheap and have temp, pressure sensor built in.

I'm sure there are other designs, too. Just need to do a google search.

Good luck in your engine build.

[TIFOSO]

So you have the technical ability to build a special race motor but you can't keep your aprilia clean?.... wtf Som!? really, wtf!? :grin:

[Som]

Hahahaha! My Aprilia is clean!! Okay okay, right now it's a bit dusty, but I just cleaned it a couple weeks ago!!

:)

- Som