Didn’t feel like messing with the car so instead took the wife out to brunch and a movie.
Late in the afternoon the car was taken out to further test the possible heat/disengagement link. As it warmed up, sure enough, the ability to get into reverse became more difficult. Reverse is always the most difficult, followed by first to a lesser degree, followed by the other gears even less so. By the time I pulled into the garage it was like yesterday, unable to put it into reverse without being “rude.” Interestingly, if it was put into first and then reverse, it went right in, as though it either stopped the gear cluster from spinning, or at least matches the speeds. Put it back into neutral and let out the clutch, then push it back in, and once again it can’t be put into reverse. It’s not an issue of waiting for the gears to spin-down either, as I could wait 30 seconds on the clutch and it still won’t go.
I wonder why reverse is the problem? I’m beginning to suspect that the pilot bushing may be binding on the transmission input shaft, dragging it along by some amount when it gets hot and expands. That’s just my pet theory at the moment since I don’t know what I’m talking about. I’ve sent inquiries to both the engine builder and clutch manufacturer to see what they say, but as it is, it seems like the engine’s going to come out – I just don’t know what to look for once it’s apart.
It wasn’t all bad today; the new brakes were bedded-in and boy do they work! Like with Kimini, I knew it was good when I got motion-sickness from it stopping harder than expected; didn’t barf but did get nauseated. After recovering I tried a few hard stops to see which end locked up first and it was hard to tell, so the balance is pretty close as-is. It’ll be fine-tuned later.
Things just get more and more interesting. As it was left last night, the engine was warm when the car exhibited the partial gear-release problem. I gave up, mostly because I don’t understand what’s going on – another way of saying that I was fed up with it, and old enough to know what happens if I push things further.
So today I got in, started it up, and it slid right into any gear I cared to try, even after setting the clutch stop back to where it wasn’t working yesterday. So why would the clutch have the problem only when the engine’s warm – or more specifically, after it’s heat-soaked? I confirmed that the slave cylinder isn’t leaking, not even a little, and it’s already been bled several times. So I know more than yesterday, but still don’t know what to do in order to correct it. If a light bulb hasn’t gone on by Monday I’ll call the clutch manufacturer.
First, the good news. The spherical bearings were installed without drama. On public streets it’s impossible to say whether the handling is better, but I’m confident it will be on-track.
The big brakes are in and the combination of larger master cylinders and more aggressive brake pads made a huge difference, never mind the disc size. Even thought the cylinders are larger the overall ratio isn’t much different, with braking effort being very reasonable. Due to having finished the installation yesterday, the “test drives” amounted to driving between red lights along with all the other commuter traffic 🙁
Now the bad news; the clutch issue is back – or it never left. It had been working fine for the last 200 miles or so and I guess I thought it was gone. After yesterday’s test drive the car was parked in the driveway while the garage was cleaned, then pulled in; the clutch worked perfectly well at that time. Several hours later I wanted to drive to the gas station but trying to back the car out showed that “it” is back; the clutch doesn’t fully disengage the transmission. That is, if I push in the clutch and slowly push it into gear, the gears are still spinning and grind. However, it’s right on the edge of being fully-disengaged because the gears aren’t spinning very fast and it’s not hard to put it in gear anyway.
I don’t understand… how can it be working when the car’s shut off then not after several hours of sitting? It’s already been bled several times, and pumping the pedal made no difference. I’m going to call the clutch manufacturer and talk to them about it, about how much disengagement (at the clutch slave cylinder) is needed for full disengagement. There’s something that keeps bugging me – I don’t know if there’s a pilot bearing in the crankshaft… would a lack of one cause this? The engine builder installed the clutch but couldn’t remember whether he installed a bearing, though getting the clutch discs lined up would be near impossible without a bearing being there. Still, it was ultimately my responsibility to confirm whether there was a bearing in-place and I didn’t check. That said, when the transmission was installed, the splines lined up fine so who knows.
The concern – in addition to facing the potential huge job of pulling the engine – is that I don’t know where the problem is. If the engine’s pulled out and the transaxle separated, there’s a very good chance that I’ll look at all the parts, shrug my shoulders, and think, “it all looks fine, now what?” Now what indeed – taking it apart when I know what’s wrong is one thing, but taking it apart and not knowing what to look for is something different. I’ve got a couple questions posted on the Honda forum but so far there have been no replies.
Been busy. A test drive confirmed that the cooling issue has been solved once and for all. The engine now warms up quickly and then stays right at 82-83 deg C. Nice… and the cooling fan now does its thing as well.
The header was wrapped, for better or for worse. The turbo can use the heat and the engine compartment can do without it, so it’s a fair trade. We’ll see how the 321 stainless holds up.
A thicker and far more pliant Remflex exhaust manifold gasket replaced the stock part. There were leak marks on the stock gasket so the upgrade was a good idea.
The big brake parts arrived and they sure are… big! Even though they’re 38% larger (11.75″ versus 10″) they’re lighter than the stock parts. Got them nearly installed but forgot the -3 AN hose adaptors and new master cylinders. A spreadsheet program was used to calculate the new brake master cylinder sizes and it really underscored the advantage of the rear weight bias and larger rear tires: at 1 G, all four brakes are doing the same amount of work, and for this reason the same vented rotor is used at all four corners.
Lastly is the issue of the rear uprights and how the stock Miata rubber bushings are shifting, allowing the A-arms to directly contact the upright – not good. In the interest of time, a spherical bearing solution was ordered, though I have mixed feelings about them having to be welded in. We’ll see how that goes or whether an alternate approach is needed. OTOH, how long should a large spherical bearing last in that application… probably forever. Hopefully by next weekend the car will be ready for more outings.
Oh, and be sure to start visiting the Midlana forum at http://www.midlana.com/forum/; as things shift from construction to driving and getting the book out, I’m starting to be more active on there.
Modified the Honda thermostat housing: epoxied in an adaptor for one return, used an existing inlet for the second, and threading the housing for the third. Seeing the two thermostat assemblies side by side, it’s no wonder that the aftermarket unit can’t work installed in the same position. I learned later that the aftermarket part can be installed in one of two different positions: replacing the entire Honda assembly, or replacing only the thermostat cover on the Honda assembly. I’d incorrectly used it in place of the Honda assembly, hence the problems. Have to get a bit more hose for the AN fitting, but I’m confident that the cooling issues has been solved.
In other news, the rear uprights contain a rubber bushing (a factory Mazda part that I planned to do away with as time permits) which has shifted such that the uprights are wearing against the upper A-arms. That’s no good and has to be fixed, either with more sturdy bushings, or better yet, replacing it with spherical bearings. Parts are on the way.
Header wrap arrived but that’s for later this week. Speaking of the header, a buddy, (thanks, Raul!) pointed out that factory exhaust manifold gaskets won’t seal well when feeding a turbo due to the back-pressure. He recommended a gasket by Remflex, which has a very good reputation for sealing and staying sealed. A gasket is on the way.
Didn’t care for the manually-set idle; if it was set high enough to keep the engine running cold, it runs too fast once warmed up. The automatic idle valve was reinstalled (and since the vacuum leaks were fixed) it now idles perfect – that’s the good news.
The rest of the day was spent messing with the coolant system. A coolant-bypass hose was added from the head to the water pump to better replicate the factory setup, but after putting everything back together, I realized all I’d done was effectively short out the radiator. Since the work was already done a test drive was run and the findings were interesting. With the bypass hose the car <em>still</em> runs too cool, though warmer than before. Unfortunately I’d changed two things at once so didn’t know what was doing what: adding the bypass hose, and installing a new thermostat with only three holes drilled in the housing. With both mods coolant temperature was up to 70 deg C (158 deg F)… better, but still too low. As an experiment I pinched off the bypass hose with a C-clamp, and the coolant temp actually got cooler – huh?
Fed up with the aftermarket housing, the stock Honda unit was dug up and things got a bit clearer after seeing how it’s made. Seems that Honda uses a two-stage thermostat, one that gates coolant locally within the block and head until the thermostat opens. Once warmed up, that local circuit (which the aftermarket unit does not have) closes and the engine then gets coolant from the radiator. The radiator and ducting have more than proven their effectiveness so it’s up to me to manage that capacity. I can either fabricate a thermostat housing or adapt the Honda part. The reason for the aftermarket housing (having assumed that the designer knew what he was doing…) was because three things were needed to be at the inlet of the water pump: the cooling fan switch, the return line from the turbo’s water-cooled jacket, and the header tank. The factory part had only two inlets which is why it was sat aside.
The question is whether the Honda unit can be adapted to work, but since it’s “free” it’ll be modified to see how it works. There’s two issues: one inlet is smooth-bore, expecting a hard line with an O-ring, and the other is that I need three inlets but have only two. I don’t know what type of plastic the housing is made from; will it stick to adhesives, or will drilling and threading cause it to crack (or worse, work for months and <em>then</em> crack?) In order to use the smooth-bore port, the ID will either have to be threaded with a fitting or plugged. I just don’t want it to fail far from home… Maybe an aluminum manifold that’s a press-fit, glued into the smooth port so the plastic housing doesn’t have to be drilled or tapped.
In other news, header wrap is on the way, along with some big bad-ass brake parts – no more messing about. I’m on-call next weekend and have to stick around home, but the above gives plenty to do. There’s a reason why it’s called “shake-down drives” or “R&D.” I’ll probably start taking it apart – again – this week, spreading the task over several days to lessen the sting of doing it for the third time…
The drive to Cars and Coffee was informative. The drive up was at an average speed of 65-70 mph, along with a couple of spurts to see how the car pulled. At 65 mph in 6th gear it’s running at 2500 rpm, just below turbo spool-up. The 110 mile round-trip consumed 3.5 gallons which works out to 31.4 mpg – which is impressive given its poor aerodynamics – but of course I’d like to think that some of that improvement is due to the various mods Midlana makes to the “Lotus Seven” shape.
The other good news is that the cooling system works… too well. You may remember the fiasco concerning the aftermarket thermostat housing, where the thermostat senses cold coolant instead of hot (and no, it cannot be flipped around.) It’s not just me; other users are equally annoyed with the unit and leave the thermostat out in order for it to work. I wanted a thermostat, so holes were drilled around the edge in order to let hot coolant circulate, only it appears I drilled too many. At steady speed on the freeway the coolant never rises above 53 deg C (127 deg F) with an ambient temperature of about 17 C. That’s crazy cold for an engine; both mileage and engine performance should improve by running it warmer. It means either modifying another thermostat with fewer holes, or blocking off part of the radiator. I vote for less holes since that way the thermostat is doing its job and radiator cooling capacity isn’t thrown away when needed, like when stuck in traffic. As a separate issue, Hondata, the Honda ECU company, states that “… The fan should be switched based on the water temperature on the return from the radiator – not by the temperature of coolant exiting the cylinder head…” I didn’t know that and wired it so that only the ECU (and I) can enable the fan. Adding a thermal switch means draining the system and welding a fitting onto the exit of the radiator, so it makes sense to do both mods at the same time. It also solves another oversight, that I didn’t put a coolant drain at a low point in the system.
Lastly, the actual drive up and back wasn’t all that much fun – let me rephrase that: I have new-found respect for our British friends driving small windy open-top cars on cold wet days, see the last picture. It provides fuel for thinking about wind wings, doors, tops, and leather bomber jackets.
About Cars and Coffee, Midlana was very well received. My thanks to Max and John for saving me a good spot (they weren’t kidding, getting there after 6 am means parking in the back.) I’ll let the pictures do the talking and you’ll see that sure enough, anything and everything shows up 🙂 Note the cop at the exit making sure to squash any throttle “exuberance.”
Earlier this week the automatic idle control was removed in an effort to get the car to idle more consistently. While that was being done, what was causing the mischief was found, probably. The throttlebody required an adapter, which came with its own gasket, and my mistake was assuming the vendor knew what they were doing. Turns out that there were two vacuum leaks due to the gasket being a bit too “universal”, with several cutouts intruding into areas where they didn’t belong. Since the automatic valve had already been removed at that point it was left off. If the manually-set idle (now adjusted the old fashioned way, with a screw) does not work well, the automatic valve will be reinstalled.
On a whim it was decided to go to the sort-of-local Cars and Coffee event held every Saturday morning – early! Apparently it’s become more popular in the last five years and I’m told that arriving after about 6:30 am is bad news for getting a spot in the main lot. I let my buddy, Max (simi-pro photographer) know that I’ll be heading up so we might get some good shots suitable for a book cover! Ugh, getting up at 5 am during the week is rough enough, but for some reason, doing it on the weekend seems worse, but I’ll live. The cool thing is that by 9am everyone’s heading home, which leaves the rest of the day for other stuff. Sorry for the late notice, but I want to stretch Midlana’s legs a bit, and with the event being 60 miles from here it’ll do just that.
For Sunday, part of me wants to head out into the mountains and desert – where cell coverage is spotty. Depending how tomorrow goes, we’ll see – driving out that way is a much bigger deal than freeway driving, with hundreds of turns and chances for more “things” to happen. If I wimp-out there’s a small car show in a rich area of town, or do I set all driving aside and work on the book instead?
Ah yes, there’s nothing like the odd collection of: high heels, tattoos, short shorts, gray hair, piercings, butt cleavage, push-up bras and mustaches – which can only mean that the Del Mar Goodguy’s Hot rod show has rolled back into town. Oh, and there were a bunch of cars there, too…
Midlana got a lot of attention, more so than Kimini. Part of it is me wising up on how to best present it. On Saturday, the mad rush in the morning put us near the back of the lot which didn’t help. However, I figured that if I parked nose-in and removed the engine cover, that should draw people over, and it worked pretty well, with 2-6 people always checking it out. However, a few were heard to say “that’s a kit”, like it was thus somehow beneath them to give it any further attention – okay. For Sunday a sign was added to point out that it wasn’t a kit, and that a book containing plans is forthcoming. Plus, because Sunday was less crowded in the morning we snagged an end slot near the main drag. That, and the turbocharger were pure magic for drawing attention. Of course, there were a few references to the dreaded “sandrail” and “dune buggy.” I worked long and hard to avoid that, but since they were few I guess I’ll live 🙂
Then there’s the four pictures below of the twin V8, quad supercharged car. The belts running to the superchargers were really skinny, so it’s unclear just how much work they were actually doing. But the comment most frequently heard in the crowd was, “How does he see where he’s going?” Good question.
Also at the car show was an autocross. To be honest I expected it to be a real circus; surely all those pretty 800-hp cars would lose it in the first corner, only that hardly ever happened. Autocross tends to be about 90% driver and 10% car and there were some surprisingly fast cars out there with two things in common. One, they were driven by seasoned autocrossers and two, allot of these “old” cars actually had complete Corvette drivetrains <em>and</em> suspension under them. To me that makes them a lot more like “rebodied Corvettes”, but whatever. (One in particular was virtually a NASCAR “pickup truck.” A side effect of the autocross was me being asked about 137 times why I wasn’t running my car in the event. I explained that Midlana only had about 70 miles on it and I was still getting used to it, and that I wanted to “break in the car, not break it.”
I’d been hearing what sounded like an exhaust leak from one cylinder and started looking around, and found it near the center of the header exhaust manifold. After removing the heat wrap for a better look, the center bolt on the flange was found loose so it was tightened up, so we’ll see if that fixes it. Since the flange was milled flat it seems odd for it to be leaking… a defective (new) exhaust gasket maybe? The back-pressure caused by the turbo can’t be helping things, but it’s still not acceptable. The header wrap is already a big disappointment for what it cost, already trying to disintegrate, with the aluminum becoming unbonded from the cloth and flaking off, and the cloth itself is getting brittle. I’m considering having the header and turbo down-pipe coated by the same place that did the turbocharger turbine housing, but of course that mean having the car side-lined for several weeks. Plus, once the header is coated it means that if it cracks it can’t be welded without grinding the coating off… Another approach is to fabricate a thin stainless heat shield that surrounds the header tubes – which would be far cheaper, using air as the insulator. We’ll see.
Somewhat related, the brakes were bedded in a couple days ago which involves half a dozen hard stops. The pedal’s a little soft so there may still be some air in the system, but that’s a non-issue. The bothersome thing is that after the sixth hard stop there was smoke coming off the front pads, and that seems like evidence that they’re going to be marginal when run hard. My brother suggested running the first trackday at California Speedway, where they also run NASCAR. For open trackday events, they run part of the main oval, then drop down into the in-field in order to add a bit more “road racing” to balance out the “NASCAR” left-turn-only thing. There, my brother and his then 240-hp managed to blue his brake discs, which are larger than mine, getting up to about 155 mph+. With at least another 100 hp and being the same weight, it seems highly likely that I’m going to cook the stock Miata brakes. I really hate to spend $$$ for the event only to have the pedal go soft… never mind the safety aspect of fast-approaching concrete walls… There are a couple big brake kits for the Miata which will be investigated. I figure that I can sell off my brake system to someone building a less demanding which well help offset the upgrade cost, somewhat…
Regarding the book, it’s a bit of a Catch-22, where if I’m driving the car I’m not working on the book, and vice-versa. I’m pushing to get it up to date so the only thing lacking is results from trackday events. After that it’s on to final review and getting it out!
Put on the race pipe just for fun, and while it’s louder it’s not as bad as expected since the turbo does a decent job as a muffler. Sure can hear the turbo whistle, but unfortunately the muffler’s going back on to avoid the wrong kind of attention. After dark, stopped on a level street and adjusted the headlights, getting some funny looks… whatever.