Drivetrain

With the Yamaha R1 engine sitting roughly in position, I turned to looking at the drivetrain. Packaging is tight as indicated earlier. I have ten inches from the back of the engine to the axle centerline. To maintain some idea of the process, I'll walk through my design's development chronologically. I tried some ideas that I found wouldn't pan out before I narrowed in on my current path.

Speaking with Don Preston in town about differentials and gearbox options, it became clear that most of the custom work that a race shop could provide would break my budget. I also called a race shop in California about the gearbox they made for the Craft/Murray Rocket (Gordon Murray of McLaren F1 fame, www.weismann.net/rocket.html, www.weismann.net/supercar.html). That gearbox was also too pricey. I considered adapting a Honda Civic transmission, but it doesn't meet my packaging targets.

Two viable plans emerged from the investigations though. First was to build the car without a differential. For track driving it shouldn't matter as much. Around town, it could make steering tough. Plan B will be to machine a piece-together differential case for an existing differential third member.

To test my concept, I first made a couple prototypes:

Second, I calculated my needed reduction ratio and laid out a differential, reduction, and optional shifter. It appears that I can shrink the layout a bit further and move the engine aft an inch or two. The diff and halfshafts will be 94 or later Miata parts unless I find another suitable alternative donor.

Third, I looked into the work required to build a solid-linked independent rear. I realized quickly that even if I used existing halfshafts, I'd still need a splined center shaft to connect the sides. If I use typical Miata or Honda halfshafts, I'd need a hollow tube with internal broached splines. That option seems more challenging than just building a case for the differential and gear reduction. I'm told that the Festiva and some other compacts have the femal splines on the halfshaft, but I'd still need to cut the male splines on a center section, and for the machining effort, I'd rather still have a differential.

At this point I'll interupt and look at another set of parts that I researched in parallel with the drivetrain work. Suspension uprights and brake systems are a complex set of machined parts. They are complex in geometry and proper operation requires a great deal of attention to detail. To make it short and sweet, making or contracting out custom brakes and uprights is beyond my budget and my immediate time constraints. I need to find existing parts that can be adapted to my needs, and they need to be cheap.

Here, my two research projects merge. If I obtain my uprights/hubs/brakes and my diff/halfshafts from a single donor source, the integration problem becomes relatively simple. I quickly turned to the part source that I know very well, the 1990-2005 MX-5 Miata. The suspension is double-wishbone all around. The parts are made for one of the lightest mass-market sportscars on the road. The differential can be open or torsen style with the same halfshafts so I can interchange third member type as desired. In addition, I upgraded my'92 Miata to larger '99 model rotors so I have a set of four '92 rotors and a set of front calipers. To make things even sweeter, I have a set of spare Miata wheels with tires. It's an easy solution.

The downside is that Miata parts are heavy for a car of this size class. I'm estimating 80 pounds per corner with the coilover. That brings the corners up to nearly the weight of an entire 600cc racing bike, or a third of the entire vehicle weight. I figure though that I have two years to get the car on the road in prototype trim or the project faces significant risk of dying. There are several months worth of part time design and construction involved in making the parts myself that I can otherwise borrow from the Miata.

So, I ordered a Miata open differential on ebay. I picked one at a much better price than I could get for a torsen. In fact, the difference is enough to pay for all four uprights. I'm shopping around now for the best price on the suspension parts and halfshafts. I'm also trying to decide how soon I should get these parts in hand versus what other parts or tools might delay the project first.

The diff looked good and I was looking forward to building a mock-up of my case. I needed to talk gears and sprockets with suppliers though to really solidify the internal/external layout. The two big questions are:

1. Will I chain drive the differential out in the open or build an enclosed case for weather protection with internal gear reduction?

2. Contingent on question one, will I try to make the reduction unit shiftable for an integrated reversing gear?

I could be dense enough to try building the shifting box. I plan though to make it easy to lock the thing in first if I do try the shifting trick. That way, if the shifter doesn't work well, I can lock it in normal driving direction. The shifter though should be easier than the typical arrangment because the bike shifter in the engine case is already handling all my basic shifting needs. All I need is to put the bike shifter in neutral, push the shifter over for forward/reverse, lock it in place, and go. The input and output shafts won't need any synchro because I can stop both of those shafts before I switch direction. I really see cost as the main deciding factor that remains.

Simplicity Prevails

Given a few weeks though to actually get into gearbox design, reality began to set in. I'd need a gear with exceptionally coarse teeth , maybe three or four per inch, roughly two inches wide to carry 150 horsepower worth of torque into the rear wheels. That's heavy and hard to come by but not very shocking if you think about it. The crown gear in the Miata diff has coarse teeth about two inches wide. It's chain drive then if I want to keep it light and simple. I'm still developing my plan for exactly how to design the chain drive and keep the diff lubricated but structural concerns have sidetracked me know that I've decided on my approach. Reverse is back to being a significant challenge, but that electric reversing motor is back now at the top of my list.

Miata Hardware In-Hand

Move forward a few months and I've purchased the major remaining power transmission gear. My halfshafts are sitting with the differential and the rear hubs with their uprights are here. The pieces that I lack are the chain sprocket for the differential and any hardware appropriate for reverse.

I will be framing in the differential soon with plywwod to simulate the aluminum frame that will be built later. Getting the structure to hold the bearings nicely on either side of the diff while making a large detour to get around the sprocket will take some thought.

Sub-Optimal But Easy

The devil is in the details and some decisions have to be made because the alternatives are too expensive. There is a cottage industry that supplies sealed-case differentials and chain-drive gearboxes for cars like mine. These differential assemblies cost more though than all the brake, upright, halfshaft, and differential hardware that I have purchased for the car. It's fair to point out that I do not have a case for my differential yet, but I am planning to assemble my case from plate aluminum sourced from a local scrap yard.

Where I'm going is that there is no clean way to bolt a chain drive sprocket directly to the Miata differential and still keep the whole unit inside a sealed case for lubrication. With significant effort, I could seal up to the R1 transmission's spocket drive shaft and keep the whole chain system inside a case. Getting the accuracy required for such an effort though would take a significant engineering effort and I already have a great deal on my plate.

The solution will be to make the differential case a simple unit with seals around the halfshaft inner stubs like the Miata originally had. The ugly part is that I will un-bolt the half shafts at the inboard flange where the halfshafts can be separated as seen above, and I will add an adapter plate to bolt up to a 60 tooth Yamaha R1 sprocket to be chain driven from the engine drive sprocket. his means that I will be driving the left half-shaft and hoping the differential plays along nicely. I anticipate that there may be some akward braking behavior because the right rear wheel will be more likely to lock up than the left when the car is in gear. It's this though or add a great deal of work to the design of the differential case. I'm holding onto a little hope that another engineer here might take the challenge and design the fully sealed system.