29 Apr 2008

A surprising number of people wrote to say that I made the car too short in wheelbase. I’m not sure why they think that since I haven’t supplied any numbers. I think it’s due to the wide track, making the car look short. Anyhow, it’s a 100″ wheelbase and a 60″ track. Guess I need to create a Specifications table somewhere…

About the torque tube, it doesn’t seem very popular, plus I may have miscalculated the stiffness. Regardless, the shifter’s got to mount somewhere, as does the e-brake lever. Then there’s all the control cables, wires, and coolant lines. If the tube is deemed useless, where does all this stuff go?

27 Apr 2008

This week was spent working on the Google Sketchup chassis drawing. Google Sketch is simplistic and has bugs but looks like it’ll do fine. The chassis renderings are far from done but they give some idea of the direction I’m headed.

The SCCA rule book led to the overall layout; in fact, it’s pretty much an SCCA rollcage with some wheels attached! SCCA rules state that for a car less than 1500lbs, 1.375″ x 0.080″ DOM tubing can be used. Great – if this size existed in the real world. Oh I’m sure that it does, somewhere, but I’ve been to about a dozen metal supplier sites and no one has it. Since the car’s supposed to be easy to build it means no hard-to-find tubing or impossible-to-locate businesses to bend it. There’s a fine line between engineering perfection and actually building something. (One of my mechanical engineer buddies is always recommending some ultimate material or part to me. I found out the hard way that while his ideas are great, what he recommends is typically expensive, impossible to find – or both – but I digress.) Unless I can find a source of the smaller tube – and someone with the proper tubing bender – it looks like it’ll have to use 1.5″ throughout.

This brings up the question of wall thickness. In terms of stiffness, 1.5″ x 0.065 is better than 1.375″ x 0.080, and it’s lighter, but for a roll cage this may be a bad idea. Roll cage tubing may be exposed to large sideways point loads – think of a car rolling over and a tube striking a rock mid-span. Thin-wall tubing may display the “Pepsi-can syndrome”: high strength in compression, tension, and torsion, but collapsing like paper if the wall is dented even slightly.

Why all the angst about roll-cage tubing and SCCA rules? Because I don’t want anyone getting hurt due to me cutting corners to save a few pounds. If you want to go to thinner wall material, that’s your decision, but I won’t recommend it. If someone says that they don’t need a roll-cage because “that’s for racing, I’m only driving on the street”, I think it’s doubly important to have one. Meeting up with a car weighing 2-4 times as much makes for a really bad day; an accident is far more likely to happen on public streets than during a trackday event. Granted the speeds will be lower on the street but it’s more likely that there’ll be a collision, instead than going off the race track and rolling the car. Plus, unlike many Locosts, Midlana’s roll-cage is the chassis.

Some may look at it and say that there’s too many tubes. I haven’t yet spent time seeing if I can delete some of them, but even now it appears that removing even one will weaken the overall structure. The windscreen is present because most states demand one, so may as well use roll cage tubing for the frame (allowing easy mounting of side wind-deflectors.) If it has a windscreen, may as well connect the front hoop to the main hoop, too. The tubes on the roof are in an “X” to allow getting in and out easily. I haven’t counted, but I think Midlana has far less tubes than a Locost chassis.

Then there’s that big square tube on the floor, what’s with that? That’s a torque tube, adding – get this – 5000ft-lb/deg of torsional rigidity to the chassis (I’m double-checking this… I may have the value wrong). This is way more than any Locost and it’s only possible because the engine’s not in the way. It connects the front suspension sub-frame to the main hoop, where triangulation extends it back to the rear suspension. The center tube also serves to run coolant lines, mounting the shifter and emergency brake handle, plus throttle, emergency brake lines, and shifter cables to the rear. This frees up the side tubes to be used primarily for side-intrusion protection – and styling – instead of torsional control.

Okay, about the pictures. Like I said, it’s far from complete; the nose cone’s left out because it’s really hard to create in Sketchup. The chassis is detailed on one side only hence the lack of tubes on the far side. Also, there’s a number of different ways the side tubes can be run which really change the look. I fully rendered one that looks pretty good but would like comments on all of them. Driving the design are the two curved side-panels: the small ones inside the front wheels exhaust radiator air into the low pressure area. The large ones lead into the engine compartment for the oil cooler, intercooler, engine inlet, and general engine compartment cooling. I’m also counting on the curved panels to differentiate the car from looking like a dune buggy. The rear treatment I like, looking a lot like a 1930’s hot rod. Louvers can be added to break up the curved panel, and to help cooling (though they’re about $5 each!). The two boxes behind where the seats will be are the gas tanks; there’ll be a liquid-tight panel between the tanks and the seats. The torque tube necessitated splitting the tank into two units.

The last pictures show what it looks like without windscreen – quite striking I think. However, the reality is that most states require a windscreen, and being buffeted by wind can get tiring, never mind getting hit in the head by pebbles or large June bugs! As said before, the builder is going to be fully involved in the decision making process, no cookie-cutter plans here. Let me know what you guys think, as unlike with Kimini, it’s important to get input. If everyone dislikes the same thing, changes will be made. Overall I’m pretty happy with the layout, which overcomes nearly all of the shortcomings that I feel Locosts has. However, I don’t like that the driver’s eye still has a line-of-site view with the front tire contact patch. This means sand or rocks kicked up have a straight shot at said driver’s eyes – I may offer side panels to prevent that. Anyhow, let me know, good, bad, or otherwise.

20 Apr 2008

Spent the day teaching myself Google Sketchup. It’s a sad state of affairs that, for me, Sketchup is easier, faster, and more intuitive than Alibre Design. Of course, Sketchup is far more limited but nothing that’ll prevent its use here. The biggest issue is that I can’t close out any panel I want; it’s unable to produce a true 3D rendering of the finished design. I also can’t do profiles or sweeps, necessary to render the Seven nose. However, these limitations aren’t a big deal since the drawings are only for producing the chassis. The book will have plenty of pictures of the actual car so full-featured 3D CAD renderings aren’t really necessary.

I’ve been warned that what I’m attempting to do – design a mid-engine “Seven” – may look like a dune buggy. In my defense, the Seven’s only the starting point, Midlana’s already making her own path. While it uses a Seven’s fenders and nose, that’s it. I don’t think it’ll end up looking any more like a Seven than a dune buggy – it is what it is. I think that the side treatment of the chassis will be unique, but as said, I can’t render it fully to show you what I’m planning. It’s still mostly wire-frame so once it’s a bit more finished I’ll post it for comments; just don’t expect full-blown CAD because that’s not what Sketchup is.

15 Apr 2008

Sorry for the lack of updates; there just hasn’t been anything exciting going on. More parts are showing up, including a box of Honda hardware needed to bolt on the engine mounts and the transmission to the engine. That’s needed when the drivetrain goes onto the chassis table for fit-up. Which means the table needs to be built. Which means the car needs to be out of the garage. On that note, the seats showed up! They’ll be infinitely helpful in figuring out the actual size of the passenger compartment. Most seats, for whatever reason, have really poor technical data supplied by the manufacturers. Plus, they’re such a goofy 3D shape that it’s hard to properly model them, so having the physical parts really helps.

Picked up electrical connectors for the Honda ECU. This is much cheaper than buying an entire dash harness and tossing out 90% of it. It does remind me that, as with Kimini, electrical work will take months and second only to composite work, is about the least enjoyable. Speaking of electrical, I may have a line on a Hondata KPro ECU; hardly required now but a nice line item to cross off.

Design work continues, including the book. There’s a real advantage of writing a book in parallel with the actual design because I write everything down before I forget. I’m also pushing forward with deciding chassis tube placement, which is more about passenger protection than torsional rigidity. Torsion resistance is being handled separately.

8 Apr 2008

At the Del Mar Goodguys Hot Rod show, there were two very interested buyers regarding Kimini. One is local, the other from the east coast, and a third I met at a red light on the way home. If any of these pan-out it would be good timing because the shocks showed up this week and the seats are due in about a week. When they arrive I’m going to be very motivated to get going. If the sale falls through I’m appealing to my parents to let me store it at their place. Dad doesn’t seem too excited even though they have tons of space. I’m pitching the idea that I’ll visit more if Kimini is there…