95Racer Development

A few people asked for more details on the auxilliary starter we built after having a lot of difficulty using rear wheel roller type starters. For one, the slipper clutch does exactly what it’s supposed to do. It slips! A lot! Because it’s a wet clutch, it’s not easily modified to accept a lock-out pin like Ducati used to use on their WSBK bikes. Even with the standard clutch (non slipper) installed, it was still very hard to start with the rear wheel either slipping on the rollers or just not getting the cranking speed high enough to where the engine would ‘catch’.

So, instead, we decided to build a fairly rudimentary prototype to test the concept using all the standard starter system parts, i.e. the starter, the reduction gearing, and the sprague clutch. It also uses the original bike’s battery. Funily, it took us by surprise how easy the bike is to start now. The crank based starter easily deals with the high compression engine and fires up after cranking for only a fraction of a second. This is great new and we are now building a second, more refined version that provides for an on-board battery mount and encloses all the gears for safety. We also need to change the end-of-crank engagement mechanism, most likely to spline type, and finish it off with ‘top hat’ that slightly recesses the spinning parts.

Anyway, here is a picture of the initial prototype. I did warn you that it was rudimentary.

And here is a link to a short video where you can see it in action. As you’ll see, it is a truly one man operation and the starter is very portable.

Just in time for Christmas, we finished our own special Christmas surprise. Feast your eyes on our 95Racer in its full glory. Well, that’s glory at least in our eyes.

2007 was very much a year of development mainly focused on getting the most from the engine. This also required a complete redesign of the airbox system as the stock one is woefully inadequate as you pursue serious gains in horsepower. Some related work then provided further opportunities to reduce weight.

The initial goal and the reason for the 95 moniker was to get 95 rear wheel horsepower in a 250cc GP sized, 4-stroke engined motorcycle weighing 95kg has very nearly been accomplished. Unsurprisingly perhaps, everything hinges around the engine. We are now only a couple horsepower short of our goal and are confident that we can at least reach it by spring 2008. The weight goal is still a little further off and the obstacle is again the engine. Much work has gone into lightening it (and everything else) but it is still overweight in our view. We are not too far off the 95 kg but further reduction will get seriously expensive. It is likely that the weight will largely remain as is for at least 2008. Still, the performance of the 95Racer is already pretty spectacular and we don’t this as a serious limitation.

Now, without further ado, here are a few of our favourite pictures. Let us know what you think. All pictures are courtesy of Cope Images, www.copeimages.com

Well, it seems my updates are like the proverbial buses that only come in groups with lengthy gaps in between. It must be the chilly weather we are now having that keeps me from spending too much time in a cold garage.

In any event, I am pleased to have finally completed the required carbon fibre work a couple of weeks ago. I have to admit that had I known how much work this would require, I’d not had started it most likely. Still, I learnt lots and not least have a much greater appreciation for the people who do this professionally, and why they charge what they do for one off work.

The main features of the new airbox configuration are:

• much larger volume (~15 litres)
• true ram-air system with air passing directly through the head stock
• monitoring of airbox pressure (and vacuum)
• accommodation for ECU, PowerCommander, and exhaust gas sensor amplifier
• tunable bellmouths for shape and length
• reduction in weight
• semi-structural front air scoop to carry dash-logger and upper fairing

As usual, most changes have plenty of ramifications. As such, this change required modifications to the frame, mainly in the headstock area, and a new tank. Relocating the ECU also required modifications to the wiring harness. as we were changing the headstock, we also incorporated a means to change the yoke offset. This then required a new way to mount the steering damper that now lives just above the bottom yoke. Its mounting is adjustable to account for the range of available yoke offsets.

So, what started as a means to make the most of the tuned engine with a proper airbox, turned into an almost year long project. The outcome, however, is very positive and track testing will commence in January. Let’s see which bit falls of first!

This picture shows the lower part of the airbox in place and one type of the carbon fibre bell mouths. You can also see where the ECU is now located at the upper front part of the airbox.

Here you can see the entire airbox installed with the tank still off. The top is simply joined to the bottom with black racers tape (duct tape). The bottom self-locates at the front around the air scoop and mounts to the throttle bodies via the stock mounting plate which we retained.

And lastly, a couple of pictures with the tank in place (now painted as you can see. The cover is held on with a quarter turn fastener. It certainly makes it easy to connect a laptop for data downloading and engine mapping.

You earn bonus points for knowning which other racing bike, achieving fame in the mid 90’s, had the top of the airbox exposed through the tank. I am still kicking myself for not buying one when they still could be had for only 15,000-20,000 UK Pounds about 8 years ago!

One of the items we identified pretty early on was that we wanted to shorten the left side of the engine. This is where the starter gears, flywheel, and generator live. To save weight, we wanted to eliminate the starter and its associated gears and so went a little bit further.

The tuned engine now has a shorter crank with a new, one-off flywheel and a light weight generator. This allowed us to shorten the left side engine cover considerably. You can see the difference in the following picture with the stock one on the right.

Here is a picture with the new flywheel installed on the shortened crank. The remaining flywheel mass is now located much closer to the centre of the engine.

The hole in the middle of the modified cover is for access with the external starter.

The next set of pictures show the difference between the stock flwheel and the replacement we are now using. It has about a third of the inertia and also weighs a lot less. We also make one that has even less inertia but it makes the engine less happy.

So from left to right:

• This is the lightest version of the flywheels we’ve tried so far. It has probably just slightly too little inertia, although it may work well at certain tracks. We will track test it in 2008 more thoroughly. The black band around the centre are the new rotor magnets imbedded in carbon fibre epoxy. It has just under a quarter of the inertia compared to the stock one.
• This is a modified stock version which has the recess for the sprague clutch machined off. It is 600 grams lighter and has 70% of the inertia compared to the stock one.
• The left most one is the stock one. It weighs 2.2 kg and has an inertia of 3.25 kg.dm2. The starter gear and sprague clutch has been removed in this shot.

You can also see the difference in the generator stator. The replacement produces about 200W which is still plenty to keep the battery topped up and run the ECU, fuel injection, fuel pump, and dash logger.

Eliminating the starter system and running a smaller generator also allows us to run a much smaller battery. The picture below shows the 3 different batteries that we’ve used.

From right to left:

• stock battery, 4140 g wet weight
• what we used last year, still running an on-board starter and standard generator, 2660 g
• what we are using now with the light weight generator and no on-board starter, 890 g

That means just in battery weight we saved over 3 kg. Quite a difference! All in pursuit of more power and lighter weight.

My available time over the last two week has been taken up with producing some more carbon fibre components. I am pleased with the progress achieved in terms of the process itself and learnt a tremendous amount since the beginning. After reading more than a fair share, I’ve started using pre-preg carbon fibre as opposed to wet-lay and vacuum bagging. The resultant parts are considerably better quality and the process is actually easier since you don’t have to bother with wet epoxy and its related clean-up. Although, it does require storing the pre-preg material in a freezer (good thing there’s a full size freezer chest in the garage), and a well controlled curing oven, which we built ourselves. After some research, I decided to try Sprint 85 as the pre-preg material which is made by Gurit. And, the early results are very encouraging. It is a two layer carbon fibre material with a layer of epoxy in between. This makes it easy to lay-up dry and it then takes 10 hours in the curing oven at 80 deg. Celsius with carefully controlled temperature ramps. Vaccum bagging is still required, of course.

In any event, the first semi-structural piece I’ve attempted is the front air intake conveyor. It feeds fresh air straight through the headstock into the airbox. It also serves as a front fairing mount and as a mount for the AIM Sports dash. It is, in effect, a front subframe and ram-air tube in one. Total weight is 350 grams. Last year’s version made from 7020 aluminium weighed 1.6 kg and had no air channel.

These pictures show the detail of the air conveyor/front subframe mounted to the frame, and with the fairing in place.

In the side view, you can see the carbon conveyor extending fore and aft of the headstock. The large yellow foam piece is the male mold for the airbox (to be made next)

The yellow part visible at the end of the conveyor (inside) is the airbox. It’s hard to imagine a more direct path for ram-air. I don’t have a huge expectation of any additional horsepower from the effects of ram-air. This subject is much studied, yet the results seem to vary widely. Our airbox will have a pressure sensor and the airbox pressure will be connected to the AIM dash logger. So, at least we’ll have some data if we decide to develop this further. Based on my limited knowledge on this subject, it seems that one thing you don’t want is large peaks of negative pressure, i.e. an airbox vacuum.

A somewhat gratuitous picture of the reasonable close-up from the right side.

The next few pictures show the new tank made from aluminium and made by the same fabricator who did all the metal work on the rest of the bike, including frame, swinging arm, subframe. Of course, he made me build a full-scale model in MDF first! I didn’t expect to have to do any wood work in building this bike.

The large yellow ‘ish foam is the male mould for the airbox. It forms part of the top of the tank in order to get as much volume as possible. At the front just behind the headstock, it has a cover that hides the ECU and a couple of other electronic boxes. That makes the electronics easily accessible without having to remove the tank and they are now well protected in case of a nasty crash.

I am really pleased with the outcome of the tank both its shape and how it works together with the airbox. All of the tank’s volume is located vertically above the gearbox. If necessary, additional volume could be added by extending it under the seat unit. As it is, it holds about 10 litres, which is plenty for club racing.

And lastly, a frontal view with the top fairing in place. You can see the v-shaped air intake that I grafted onto the fairing. This slips inside the front of the air conveyor and thereby constitutes the front fairing mount. You can also notice how well that Kawasaki 650cc twin engine “hides” inside the fairing. In frontal profile and area, it doesn’t look any different to the Honda RS250, which is of course what the goal was.