31 Mar 2013

Moved the license plate straight down about 12″ – looks much better and no longer blocks the engine compartment vent.

Took the wife for a ride as first-time passenger. She got real animated when I got on it… I haven’t a clue what she was saying due to all the car noises, but there was a bunch of arm-waving 🙂

Drove around a bit myself later on and got on it a couple times. That was the first time the shifter had been “used in anger”, and later I was backing out the car after mounting the Go-Pro and snap, the shifter broke! If that had happened anywhere but home it would have almost certainly required a tow – what’s that saying about a race being ended by a $0.25 part? Anyhow, “before” and “after” shots are shown. The failure occurred at the thread root of the long shoulder bolt which is why the bolt is missing its threads. After looking at it, the obvious fix was to substitute a single piece of bar stock to extend straight through, replacing both bolts and removing all bending loads on the threads. It’s done and installed but by then the light was gone. It was also noticed that a couple of the screws had backed off during what few drives there have been, a consequence of not using thread-locking compound – fixed. And people wonder “what’s the point of these shakedown drives?…”

One thing I haven’t tried yet is a hard standing start, partly for fear of breaking something, partly because there’s been too much traffic, and there’s not yet having mastered the straight-cut gears. I can’t seem to shift cleanly from first to second smoothly even using the clutch so I need to figure that out. However, the desire to know what its 0-60 time is and to record it is strong… It’ll also answer the big question regarding how good the traction is, a common problem in high-power front-engine Locosts. I purposely chose a low final drive to help avoid it and between that and the 67% rear weight bias, it’s promising.

Oh, and the Miata vent assembly (whatever it does) was leaking a bit after hard acceleration with the tank full, due to not being sealed adequately to the top of the tank. The vent assembly was removed, sealant added, and reinstalled. Time will tell.

29 Mar 2013

Stop-nuts were put on the brake and clutch master cylinder shafts, removing one variable from what could be causing the shifting clutch engagement point. After thinking it through, it seems unlikely that it’s “the clutch needs to wear-in.” That would make it engage further from the floor, not closer. A leak in the system would be… well, a leak, but the system shows no sign of that. A buddy suggested that maybe the clutch master cylinder itself is defective, with fluid leaking past the piston. Maybe, but where’s it leaking to? I’d expect to see external evidence… unless it’s getting by the piston and returning to the reservoir instead of doing its part to push the clutch in… hmmm. That could be it but we’ll wait and see if the check-nut fixes the issue – I hope so but am suspicious.

Recalibrated the fuel level sensor. The tank was drained by disconnecting the fuel return line and pumping the gas into a container until bubbles appeared. That’s a far more realistic version of “empty” than what CAD calculates. With empty established, readings were taken (with the engine running to stabilize the voltage) as one gallon at a time was added. What’s unclear is exactly how far the tank can be filled since the fuel filler pipe and safety-reversion float from the Miata were used. The bottom of the filler assembly dictates the amount of fuel that can be added before the fuel vent has to deal with it. I got to 12 gallons and the tank started making burping noises so things were stopped there, counting on the math equation to work if it’s filled further. I’m not sure what will happen if I over-fill it such that there’s standing pressure on both the fuel level sensor and Miata vent assembly (not sure what it does.) I don’t want to be at the gas station and have it spit back out the filler, or suddenly see gas dripping under the car…

I made the judgment call to route returning fuel from the engine to the main accumulator section in the tank. That seemed to make more sense then having the tank do its part with its mouse doors to get the fuel near the pickup, only to have the fuel returned elsewhere. This way it keeps that section as full as possible, but I may be over-thinking it; perhaps the fuel has something else up its sleeve and won’t follow my imagined rules. The point of mentioning this is that even though “empty” was established, the section from which the fuel is drawn will be kept more full, so even when the tank indicates empty there should be time to high-tail it to a gas station.

26 Mar 2013

Corrected the wastegate plumbing and took it for a short drive… Sheez, yeah, it’s pretty quick, even with the reduced boost! Some quick math indicates roughly 325 hp, which will be fine for the initial learning curve, and a good idea for taking it easy on the engine at track events.

24 Mar 2013

Shakedown/test drives have started. A lot of people have been bugging me about why I bother, just drive it instead of these wimpy baby steps. The thing is, driving a freshly-built car is kind of taking up running, you don’t just sign up for a marathon, you build up to it. The day started with a bunch of laps around the block, then venturing  out further, finding a few issues along the way:

1. The clutch anomaly is still present. It may be due to not having a check-nut on the clutch – my goof – or perhaps the clutch has to break in. Regardless, twice it had to be adjusted and while I’m not too worried, yet, the Concern Meter is up off its zero peg…

2. The fuel level sensor was calibrated using a 12V power supply, but a fully-charged battery is more like 14V, making the readings wrong by a fair amount at the low end. The fuel tank will be pumped out and one gallon at a time added, recording sensor output each time. That equation will be entered into the dash and then there won’t be any guessing how much it’s really holding – I don’t like inaccurate dash readings.

3. Part way through one drive, the oil temperature and pressure suddenly shot up, with oil pressure indicating about 420 lbs… sure. Thinking it through, since both sensors are located on the dual remote oil filter housing, the suspicion was that the aluminum housing had become ungrounded from the chassis (necessary for the one-wire sensors to generate readings.) Sure enough, the filter block didn’t have a reliable ground so some powdercoat was ground off and a split-lock nut added under a mounting bolt – fixed.

4. The alarm kept getting flagged by oil pressure dropping below the 20 psi threshold. The problem is that the alarms are very simple-minded; they’re just a value below which a warning is flagged. The problem of course is that how low oil pressure gets very much depends on both engine rpm and oil temperature, both of which the dash knows, but the software doesn’t allow for more complex alarm conditions. Oh well.

5. The engine really wanted to run but it was being limited by the ECU. The setup is in place to have boost-by-gear, which means literally that; each gear can have its own boost level, sort of a poor-man’s traction-control. For now it’s disabled, expecting that boost would be limited by the springs inside the wastegates, but it isn’t. It may be because the solenoid is normally-closed; perhaps it’s trapped air above the diaphragm which is keeping boost pressure from opening the wastegate valve. I unscrewed the top hose from the wastegates so that the air vents directly to the atmosphere but didn’t have time to try it. The point of this long explanation is that the ECU cut off fuel at 11 psi (I set it there as protection against me overdoing things… good thing.) The result was that at around 4000 rpm, the ECU shut off fuel – hard – which was like kicking the engine in the nuts… not glamorous for seeing how well the engine would pull, with the party being over before it even started. Given that the engine’s built for 8300 rpm, it’s not even half way there!

That said, I wanted a wild engine and I got it, kind of like having a crazy redhead for a girlfriend, one with tats, piercings and owning a sportbike – or so I imagine. Between the boost and straight-cut gears, she’s going to take some getting used to.

I went around a few corners at maybe 30 mph and it’s very clear that Midlana corners really well; it’ll be interesting to see how things go at higher speed 🙂  For the above reason regarding the engine, I never really get on it from a stand-still since it would only barely get going before having the fuel pulled out. Still, the one time I did, it didn’t spin the tires. That’s very promising and confirms that having much of the weight over the rear-driven wheels provides great traction. It’s one reason why I’m going to run without boost-by-gear for now and see how it goes.

So why no video? It’s boring! Between all the above issues there just isn’t anything interesting. That, and driving down our local Highway 101 wasn’t any fun as there was a lot of bumper-to-bumper traffic. Not much to see except license plates at eye-level…

21 Mar 2013

And so it is done. The car now has insurance and is 100% street-legal!

There’s no rush though, as various dash parameters still need adjusting; don’t want to head out and have wrong values being reported. Also tried adding a thermal blanket over the exhaust manifold and… ugh, note to self, don’t do this in a garage. The amount of smelly smoke that off-gasses is pretty bad; I’ll be hearing about that for a while. Also, I just wrapped the blanket around the header loosely and it looks pretty crummy. Anyhow, now the build moves from construction to testing! Baby steps.

18 Mar 2013

Received the die-cut decals – awesome service from http://www.fullintensitygrafx.com. They’re one of the few places that handles small orders and also have a great color selection. Regular readers will know that I don’t complement companies unless they really deserve it – they do – and I didn’t screw up the installation.

Small applause for American Modern Insurance Group approving coverage of Midlana. However, I’m withholding celebrating until I hear from Grundy, which I consider a more established company (never mind also being 40% cheaper!) but either way the car should have insurance by the weekend!

17 Mar 2013

Decided to corner-weigh the car instead of fussing with the rear plate placement. Weight of the complete car with six gallons of fuel is 1616 lbs, or about 1580 empty – it is what it is. Initially the cross-weight was way off at around 55% but now it’s right on, and the car is right at 67% rear weight bias – it’ll be interesting to see how it handles.

The dash variables are all working… they’re all reading “something.” There’s a number of parameters which aren’t scaled right, rpm for one. Don’t know why that’s off since it was working fine at one time. Need to plug in the laptop to the Hondata ECU to see what the real rpm (and MAP) are to figure out the error ratios. Fuel level is low today; yesterday it indicated about six gallons and today it’s reading four, which may be due to the sensor output varying with battery voltage… that’ll be dealt with later.

Sigh… the rear license plate. Where it is in the picture is the obvious place for it, except that it’s blocking a big portion of the exhaust vent. For that reason it needs to move but I’m not sure where. Directly above the exhaust? In the center of the curved section? Or move it up and forward, attaching it to the main roll hoop cross tube behind the passenger seat? I don’t even want to think about the front plate… for now I’m going to run without. Putting it on the front grill will cut incoming airflow to the radiator by about 80%. About the only other places it can go is either on an A-arm (ugh) or make a frame that sticks out from a headlight mount.

16 Mar 2013

Much of the day was consumed by relearning and reconfiguring the dash and data logger. Everything looks promising though there’s a few loose ends, such as it not allowing a low fuel alarm at two gallons… why? The GPS appears to be working; the little light says it’s locked in, and altitude is accurate, but both latitude and longitude are totally wrong, yet stable. I haven’t figured out if it’s a units issue, a reference problem, or what.

Re-bled the clutch which improved things; I hope that the anomaly from before is resolved.

Disassembled the leaking coolant adaptor and slathered on sealant over the O-rings that weren’t doing their job.

Tomorrow will be for checking out parameters with the engine running, moving the license plate down so that it’s not blocking the rear engine vent, and moving stuff around in the garage to finally let my wife park her car inside… she’s been more than patient for a very long time.

15 Mar 2013

Applied to Grundy for insurance. What’s interesting is how a signed application, pictures, and check must be sent, even before they’ve approved the policy; ‘spose it’s to save time if approved. Haven’t heard anything from the first company, American Modern Insurance Group, which leads me to wonder if their so-called “lock-out period” for the earthquake continues, or if they have no intention of contacting me, or they’re just clueless. Less and less impressed as more time passes. Let’s hope that Grundy comes through so I don’t have deal with AMIG… It’ll be interesting to see if Grundy pulls the earthquake card or if that’s something that only certain companies exploit.

Received notification that the decals shipped… dang fast turn-around time of about 18 hrs!

Received the longer Dzus fasteners so they were swapped in as required.

Received the entry pass for the Del Mar hot rod show. I suspect that Midlana will likely be about the only car there without a big chrome-plated V8.<br><br>

Have to get reacquainted with the Race Technology digital dash and data logger; after 3 years I’ve pretty much forgotten how to configure both. The various screens need to be configured to something useful so priorities must be chosen about what variables go where. MAP and oil pressure are both critical since both can kill the engine within seconds, though it’s also unlikely that I’d notice in time to get off the gas. The good thing is that alarms can be set so that’ll be done for sure. No more quick drives down the street and having no idea what oil pressure is… Hopefully RT’s products have improved. First impressions were… less than impressive due to so many bugs.

14 Mar 2013

Suspicious of the earthquake excuse below, I applied to another insurance company and we’ll let competition decide which to go with. Actually, between the earthquake claim and them getting the odd idea that “drives along the beach” meant driving in the sand(!), I’m not feeling a lot of love for them. Frankly, I’m kind of hoping that the new company will offer a policy first…

Ordered die-cut decals for the sides of the car, advertising “midlana.com.”