build diary

April 13, 2006:
I seem to have conquered my fear of drilling holes in my fibreglass.
This is for an amazingly bright "you have no oil pressure!" light.
entry 1197
Time to work on the new windshield.
This one will probably be too low of course, but I reserve the right to make that error.
entry 1198

April 14, 2006:
A frustrating day.
I stuck all the various pieces back on the car so I could bring it in to work this morning. Paint the grille surround black, fit the Dzus fasteners, new foam tape for the bonnet, bolt on the front fenders, install the wheels, install new fuel injector connectors and tape up the wiring - you know, the usual. I also cut the new windshield template. I jumped into the car to bring it down to the glass shop and...nothing. It cranked enthusiastically (good little battery, that Odyssey!) but there was no fuel whatsoever. We pulled the ECU out and found a problem with the chip, so I changed that and loaded in my special programming. Finally, a few hours later (I have a very, very understanding boss) Tore off down the street to see if an upholstery shop could make a little tonneau cover and of course I'm a little late to get all this done before I leave next week. Then outside the upholstery shop, same deal. Nothing. I borrowed a screwdriver, popped the top off the ECU case and wiggled what we thought might be the problem and the car fired up. Argh. When I got back, it wouldn't start. Later in the day, going to pick up the glass (the Glass Brokerage once again came through with a same-day windshield for $40!) it fired right up. I'll be into the ECU again tonight. Not exactly what I wanted to do of course. I have a few other things I was planning on doing.

A little treatise on my new oil pressure light. I used to use a gauge for the oil pressure. The problem is that it's constantly moving (of course) and I'm afraid I'll miss it if a problem shows up. So I elected to go with a light that's right beside my rev gauge. This way, if it lights up I'll know to kill the engine immediately. Ideally I'll have both. Someday I'll revisit instrumentation and do it then. I did have to plug in a sender from a 1995 Miata to make this work but otherwise it was the same wiring.

The new vs the old.
Well, the template for the new anyhow. It's not a major change but I think it will really affect the look of the car.
entry 1200
The new oil pressure light.
It's the brightest light I could find at Radio Shack and is nice and visible.
entry 1201
A gratuitous shot of the car.
entry 1202

April 16, 2006:
Well, yesterday wasn't much better when it came to frustration.
I started by installing a new ECU and tidied up the wiring a bit. I even installed the ECU case on rubber isolators in case part of my problems were related to vibration (in the Seven? impossible!). I then turned to the windshield.

One of the things I did with this latest piece of glass was make it symmetrical. My previous versions were not. The bending of the channel went quite easily after all my hard-earned lessons last time. The plywood was 0.8mm thicker this time which actually made things a little more difficult. The nice thing about a symmetrical top is that instead of having to set up my bending tool twice, I simply turned the piece of channel around and did both sides that way. A bit of Scotchbrite on the aluminum and it looks really good. I hopped into the Seven to run down to work because I wanted to touch up some welding on the side support. That went well - and the Seven refused to start. It was the same symptoms as before except that I never managed to get it running again. I spent a long time on the phone with my friend and boss Bill to no avail. The current thinking is that something in my original ECU is damaging the processors in the chips. I'll try a new chip in this new ECU and see what happens. Argh.

Today was set aside to do some work on my trailer. It's been having some cracking problems where a loose fender once vibrated badly. I've been patching it up but it was time to fix that for good. Some hefty angle iron and some fun with the plasma cutter and it's much better. I wouldn't tow a 3000 lb vehicle on it, but 1300? No problem.

Once that was done I stuck the windshield on the side supports and went to bolt it up. Uhoh. My previous glass wasn't symmetrical because my scuttle isn't symmetrical. I also put the side supports a little too high and it was difficult to get the windshield level as it teeter-tottered on a high point in the middle of the scuttle. Some choice words, some oversize holes and it's done. The flat top makes a level mounting crucial. So how does it look? Fantastic. The rounded top has bugged me since I first put the windshield on the car. The new design, about 1.5" lower in the middle, looks much more aggressive and is still out of my sight line. It was a little frustrating but I'm done.

And that concludes the last full day I have to work on the Seven before the Texas meet. Mark Rivera reports that his car is running well. With a 12 psi turbo on the engine, it should really scoot. Chris Fiaccone's bike-engined car will also make mine look slow. I'm going to look like the sad also-ran!
entry 1203

April 17, 2006:
Exorcised!
Jeremy figured it out. With one of the ECUs attached to the simulator, he started twiddling dials. He poked one and the injectors woke up. What was it? The throttle position sensor (TPS) signal. I hadn't played with it because I "knew" it couldn't be the problem. Above a certain value, it was preventing the injectors from firing. With the wiring for the TPS removed, there was a phantom signal that was changing with heat. With a cold ECU, it would sometimes be low enough to let the injectors fire. As things warmed up, it would rise to the point where the car wouldn't get fuel. See, the ECU has a setup where if you crank on wide open throttle, it won't fire the injectors to help clear a flooded condition. This is what was being triggered. I solved the phantom signal problem by installing a jumper between the TPS signal input and the reference voltage coming out of the ECU. No more heat-related variance and a nice consistent start. The odd thing is that the TPS has been removed from the car for months. I trimmed some wires back when cleaning up the engine bay and that's when the signal must have started to creep. Very odd. I must admit that if your car is going to develop this problem, it should always develop it at the Flyin' Miata shop where there is more gathered Miata wiring knowledge and diagnostic tools than anywhere else on the planet. A big thanks to Jeremy for finding the trigger for the problem and an even bigger thanks to Bill who spent hours with me on trying to chase this down.

So, with two more evenings before I leave, I have only a few things to do. First is to bend up the new hood. The flat, louvred sheet arrived today. I may chicken out and get a local fabrication shop to help me as I'm scared to screw it up. I'll also cut a new stone shield for the driver's side, as the new one on the passenger's side makes it look battered. Pack the trailer, gather the tools (the garage is a little, umm, chaotic) and I'm pretty much ready to go. I hope so anyhow.
entry 1204

ECU hassles are just what I don't need at this point.
entry 1205
The new rubber isolators for the ECU.
In case anyone doesn't accuse me of being "locost", they're old bicycle inner tubes!
entry 1206
A bad picture of the new windshield.
With the grille in place and the filter sticking out through the hood, it's one mean looking little car right now.
entry 1207

April 18, 2006:
The new hood is finished.
It took a few hours - I don't want to figure just how many - but it worked far better than the original one. I sucked it up and went at it by myself. Instead of 0.050" 5052 alloy, this one is 0.060" 3003 and it's much easier to work. It doesn't exactly fall in to place but I was able to shape it nicely. The 5052 needed to be over-bent and then sprung back while the 3003 is soft enough to simply push it into place. A comparison of the fit of the hoods shows a massive improvement over my original and I'm very happy with it. I used a Scotchbrite wheel to do some final shaping and the finish is smooth and even. Even better is the fact that the louvres are not embarassingly backwards. They're also higher quality than my last one. A win overall!

One thing I did change that made the fabrication easier had to do with the lower edge. Instead of folding it under, I cut it off level with the frame rail. I then riveted a piece of alumimum channel along the inner edge to act as a support. This channel was left over from my original windshield fabrication - it was the stuff that I couldn't bend without kinking. Now I don't want it to bend so it's perfect. I'm not crazy about the rivets along the edge of the hood but I'll live with that.

So, what's left before I leave? I'd like to replace that stone shield, I want to ensure the fuel pump is clean and I have to pack. And that's about it! Wow. I do have to thank my girlfriend Janel for keeping me alive during this thrash. If it wasn't for her dropping by the shop with food I'd probably have starved to death.
entry 1208

April 19, 2006:
As the old hood looks on, the new one prepares for surgery.
See that engine hiding behind? That's waiting for a new project...
entry 1209
According to Ron Champion, all you do now is just use the scuttle and nose and bend the aluminum around.
Umm, it's not quite as simple as that sounds.
entry 1210
After some fairly lightweight work (well, by comparison to my first time around anyhow), here's how the hood fits without anything holding it down.
Whoops, something seems to be missing.
entry 1211
Now with a hole for the intake system and latches.
entry 1212
Louvres, now pointing the correct way.
entry 1213

April 20, 2006:
After a fairly easy night of packing and relaxing (not a completely even distribution of the two, but better than might be expected), I dropped the Seven off at a local upholstery shop to have a tonneau cover made for the rear.
Was this a good idea? Good question. I figured the cover would make a big difference, closing off the last mechanical area of the car. It was certainly a risk, leaving the car in the hands of a stranger only hours before my departure. It didn't help when they seemed a little puzzled when I showed up with the car at 8 am. So it was a moment of doubt as I left the shop with my car parked inside.

A few hours later, I got a phone call that the work was done. I jumped in the Jeep - once again, my girlfriend Janel has allowed me to use her precious Grand Cherokee as a beast of burden - and headed over with the trailer. To my great relief, the tonneau looked great. It was exactly what I wanted, closing off the rear nicely without impeding access to my fuel tank and giving the Seven a proper boot.

So, car strapped down and Jeep full of items such as two spare sets of tires, a set of cornerweight scales, an air tank, jack, tools, spare nuts and bolts and even a small Shop Vac, I took off for Texas.

That last sentence does overplay the sense of occasion that remained a few hours later. While I had beautiful weather for my trip over the high Red Mountain Pass, New Mexico proved to be long, flat and dull. The northern section of the state is not a rich area and I felt a little conspicuous. It's not the sort of place you build a little bright orange sports car for the hell of it. More entertaining was the fact that I saw many signs that announced my altitude as 7000'. Always 7000 and always fairly randomly. At one point, I decended a fairly major slope shortly after a 7000' sign and was presented with one that said 6800'. Okay, someone at the New Mexico DOT is playing games.

I stayed with my friends Sam and Carol Sharp in Alberquerque after rolling in fairly late. It might have added a bit of time to my trip, but it's always more pleasant to stay with someone after a long day on the road.
entry 1214

April 21, 2006:
Ugh, long long drive.
Sam and I ran around a little in the morning to have a big breakfast and then hunt down a plug for my O2 sensor bung. I like to remove the sensor for photos to make the car look a little better and I'd left my plug in Grand Junction. Whoops. So a relatively late start but at least I missed the traffic.

Drive drive drive. Boooring. The Jeep thought so as well as it decided to liven things up a bit. Some bad gas in Texas meant I had to stop at a WalMart and buy some miracle fuel injector cleaner and a bottle of octane boost just for fun. Miracle cure it was, the Jeep decided to let all the cylinders wake up and we carried on. A late night and I arrived at Curtis Wiseman's house after managing to completely confuse myself in his neighbourhood.
entry 1215

April 22, 2006:
Ahh, the Se7ens festival.
Turns out the festival is really a NASA race weekend with a Se7ens parking area. Fair enough, it made for some very interesting scenery running around - two Porsche Carrera GTs, for example - and a real track bias amongst participants. I was given a hearty welcome by the organisers and unloaded the car. Shortly after, Mark Rivera showed up with his heroic build that went from daily driver Miata to running Locost in about 4 months. He quickly set to work installing seats, side panels, fenders and a number of other pieces that some consider important to a functional car. Since I knew his car should be significantly faster than mine, I decided to emphasise my finished state by polishing. And polishing. And polishing. Mark suggested that perhaps I should spend some time ensuring the car was running well and I scoffed. After all, it's a Miata engine and it was running beautifully at home.

Chris Fiaccone showed up shortly after with his car. While he hasn't been quite as crazed as Mark over the last few months, he did step up the pace to have his motorcycle engined car ready. And it is a stunner. Painted flat black with some really nice touches on the rear lights and around the exhaust, it looked like a mutant hot rod and showed some beautiful workmanship. Another car faster than mine, so I polished some more.

A fourth Locost showed up as well. Jon's car was built partially as a father/son project, partially to let Jon learn metalworking skills and partly for the Grassroots Motorsports $2005 challenge. It's amazing. Jon built the nose cone and all other body parts by hand. It looks more like a Lotus 6 than a Lotus 7, but the amount of work involved rocks everyone back on their feet. I knew that I couldn't out-polish this one so I stopped and decided to go for a spin.

As has become tradition, the Seven decided to foul up a couple of plugs to celebrate the arrival at a new track. It also was running very rich as I discovered on a short run. Why? Well, thanks to my super-cool throttle bodies it runs in a fairly narrow range of vacuum. And that range had just shifted drastically enough that my highway cruise at sea level was similar to my wide open throttle at altitude. I took off to do some tuning and try to settle the car. I also discovered that the synchronisation between the two pairs of throttle bodies had developed a problem and my idle was now about 1300 rpm if I was lucky. At home it's closer to 850. It's as if there is a big vacuum leak but I can't find it anywhere. I had to dump a lot of extra fuel in the idle zones in the computer and pull a huge amount out of the cruise. Hmm, maybe I did spend too long polishing.

Or maybe not. I started talking to Howard Cole from Kit Car magazine and he did a photo shoot of the car. Hopefully he'll give a review of the book and accompany it with some photos of the real thing.

So, other than the poor running and stinky black smoke in the pits, how was it? Well, the car was very well received. The most common comment was that it looks far better in person than in my website, and the color in particular blew people away. I guess my photos don't show that it's metallic. This is probably because the car is usually dirty and photographed in a dim garage. I've also been working to improve the presentation recently. As always, the Birkin/Caterham/Ultralite owners were extremely welcoming.
entry 1216

April 23, 2006:
Tuning troubles.
Argh. The Seven was running better after I plugged in the laptop and made some major adjustments to the setup, but it still wouldn't idle happily. I played with the synchronisation of the throttle bodies without much luck as the problem seemed to be heat related. When cold, the car would behave just as it does at home. As it warmed up, the throttles would start to stick a little and the exhaust would belch black smoke as I blipped the throttle. Is it a synchronisation problem or something else? Given ambient temperatures around 90F, heat was quite possibly a factor. Above 4500 rpm or so, it runs strong. Below that there's a flat spot. I've been back to the computer a number of times to try and sort it out. If I'd kept my stock intake manifold, this wouldn't be an issue at all. It's very embarassing, especially when Larry Webster from Car and Driver is hanging about and I can't get out from underhood. Even he said the car looked much better in person than in the book.

I did manage to get out on track at one point. Brian from World Class Motorsports took me out in one of their Ultralites. The Motorsport Ranch is a fairly quick track but one that has a lot of very odd corners. The most impressive things about the car were the brakes and the unchecked acceleration above 100 mph. Where my car is running out of steam, the S2000 engine goes into VTEC mode and leaps forward. A very compelling ride.

We took the group of four Locosts out for a few little photo shoots and to let Larry drive them. He commented on the hot footwell in my car (true, especially in the 90F Texas heat) and on the poor idle. He was happy with the way the car handled the bumpy back road we were using for the test drive. It was a good test of compliance, that's for sure. The new suspension did a great job of sucking it up despite the high spring rates. When Chris Fiaccone had the chance to drive the car, he was really impressed with how well it braked and how well it rode. "It feels like a real car" was his comment. He also felt it made his car feel unfinished, but then again I've had an extra couple of years to tweak mine! The brakes were what really seemed to get his attention. It appears my work here has been worthwhile.

While Mark had no problems with his manic turbo Locost, Chris was having some teething problems. He'd been battling with them yesterday but I'd been too busy to pay much attention. I was able to donate a couple of fasteners to the cause and help him find a major running problem that had the car immobilised. Once that was sorted out, he was able to take to the road to the sound of cheering from the spectators who had watched his long battle. Unfortunately, he discovered that the car got very unstable on deceleration. We spent some time looking at the front suspension to try and find this and it wasn't until late in the day that someone discovered play in the rear suspension. It turns out that one of his wheel bearings was completely shot so the left rear wheel was flopping around. A local Spec Miata race shop was willing to help us out with a new rear hub but they'd just used up their last spare. It's a real shame that Chris will have to miss out on the drag strip session tomorrow but I think he'll be just as happy to go home and sleep for about a week! I'm glad I got a chance to see his car though, it is a beauty. I even had the chance to get behind the wheel and try it out. It's almost 400 lbs lighter than my Seven - yes, that's right - and it feels as if it's going to be a total riot when it's all done. The car is a rocket and a real looker.

Tomorrow we head to the drag strip for some instrumented testing. My car will likely end up in the middle of the three as it's the lightest and has a bit more power than Jon's gorgeous low-budget build, and of course I can't touch Mark's car for horsepower. My car's still not happy at around 4000 rpm but hopefully that won't be an issue in this sort of testing. I guess I'll know soon enough - the laptop will be busy!

An interesting side note - neither Chris nor Mark had ever seen a Seven before, never mind driven one! They decided to build their cars based on other reasons. I've been a Seven enthusiast for a long, long time. I guess it does show that if you're looking for a cheap sports car (ooh, subtle plug there!), the template is a difficult one to ignore.
entry 1217

<< | show individual entries | >>