Destination SSM One Miata's Journey through Hell

4Apr/14Off

Update

Posted by David de Regt

I took much of last year off, made some poor suspension choices, and didn't pay much attention to the car.  I got my butt kicked in most events I went to, and wasn't very happy with the car or my driving.  In the end, though, it gave me what I needed -- most of a year off.  I flew around the country and codrove with the Woottons a few times, autocrossed a bunch of random cars, and had some fun anyway.  I also noticed that, despite lots of attention to the car everywhere I posted it for sale, no one even sniffed around at a real offer, so I know that, someday when I do actually sell, parting it out is the way to go.

1530326_619095470648_962636239_n[1]For this year, I'm going to make a little more of an effort again.  We moved into a new house last year and didn't have a good way to deal with the towing situation, so this year my dad sold his giant trailer and I bought a tiny little trailer, perfect for a Miata.  Most importantly of all, I can back the trailer, with the car on it, into my garage, and leave it there, ready for an event at any time, with the truck parked in front of it.  I'm hoping that I'll find a lot more motivation to autocross when it means spending 5 minutes in the morning and 10 in the evening loading/unloading instead of 2+ hours on each side getting the truck/trailer to/from my parents'.

I decided to attack some of my root annoyances with the car as well this year.  At the end of last season, I realized that my traction control system was a little flaky.  While it often helped, it often cut in when it had no reason to.  I spent some time debugging it at the first event this year and learned that it is, in fact, cutting fuel even with the system completely disabled -- when I bypass the unit and plug the injectors directly into the ECU, it runs perfectly.  I've been working with RaceLogic the past couple weeks trying to narrow down what the problem actually is.  So far, we think it's the adjuster unit, but it will require further diagnosis.  Until I get that sorted out, the unit is staying fully disconnected, forcing me to learn how to modulate a gas pedal -- something I'm not very good at, apparently.

intake[1]Another big annoyance was the general tune of the car.  In addition to feeling like I didn't have enough power, there was always a bunch of hiccups in the tune (above and beyond the RaceLogic-induced ones).  Throttle pump (tip-in) was always sketchy, leading to unpredictable power levels on corner exit.  At 2012 nationals we noticed that underhood temps were crazy, and put a hole in the hood for the intake.  Since then, I'd switched back to a more traditional intake, but pulling air from the driver's side headlight area.  I also upgraded to Flyin Miata's big flex fuel kit, which is a pretty awesome setup, and upgraded to ID1000 injectors in preparation for E85.

miatadyno_20140302[1]Last spring, I changed from an AEM to a Hydra Nemesis for an ECU, and immediately was able to improve the drivability of the car quite a bit.  However, it was still far off what I wanted/needed it to be for consistent driving.  I gave the car to Kris Osheim of KO Racing early this year and he spent a few weeks road tuning it to help with drivability and low end torque.  The car is an entirely different beast now -- mashing the gas at 2000 RPMs produces an instant wall of torque, and different rates of tip-in actually produce linearly proportional amounts of power.  Using the flex fuel setup, I can now put anything from full pump gas to full E85 or any mix in between, and the car fires right up and makes a safe amount of power for whatever ethanol mix I'm running.  The car now makes 359rwhp/279rwtq (Dynojet corrected) on E85 and 316/256 on 92.  The old power numbers were falsely inflated by the previous dyno I was using -- he applied a fixed 20% correction factor, so this is actually quite a huge bump in power over the old setup.

Suspension-wise, I knew I needed to change things.  In 2012 I was running 700/450 springs, and based on the math said we should be running 800s in the front, so for the 2013 season I tried just bumping to 800s.  Of course, the car went from fairly well balanced to ultra-pushy.  I spent the limited season trying to dial it out with shock adjustments, but for 2014 I went up to 550 rear springs and added a small rear swaybar.  Two events into the season, the balance of the car is proving to be a huge improvement over last year, which makes it a lot more fun to drive.

The brakes had always been a big complaint of mine with the car -- they were a little unpredictable and, especially on concrete, you were essentially unable to bring the car into ABS.  For 2013, I upgraded the fronts to Flyin Miata's little big brake kit with the 11" rotors, and went to a 1" 929 master cylinder with a dual diaphragm brake booster, which were a huge improvement.  I'll likely upgrade the rears to their rear little big brake kit this season as well, but I haven't pulled the trigger on that yet.  I've also changed out the rear wheel bearings and front hubs for this season, as they were starting to get some play in them noticeable as pad-knockback after hard cornering.

I'm planning on attempting to get the car out to several big events this year -- Crow's Landing, Packwood, Spring Nationals (Lincoln), and Nationals.  There's still a bunch to get working on the car for real -- Eric Anderson's car proves how effective the RaceLogic can be if I can work the demons out of my system, dialing in the new suspension setup, and learning how to drive a car that can break the tires loose at will in a straight line in 2nd gear on Asphalt.

8Jul/120

Packwood ProSolo

Posted by David de Regt

The ProSolo was an interesting weekend.  The course was the most annoyingly not-fun collection of ridiculously tight offsets and stupidly slow corners that I've ever driven.  I feel like, in 8 years of autocrossing, I have to have driven a less fun course than that, but none come to mind.  On the upsides, I managed to squeak out a win (barely -- 0.009 seconds) and it highlighted a lot of things to focus on for the next phase of development for the car.  Also, we had zero reliability issues with the car again, so I'm just waiting for something expensive to explode at this point...

One of the big obvious weaknesses this weekend was power application.  I think that part of the problem with my current build is the forward weight bias (we're at about 56% front right now), which isn't leaving any weight on the rears to put power down.  I've been working on a design to try out that will let me drop down to a 1.6 motor, non-intercooled, which would drop a ton of weight off the front end, and potentially give me room to ballast up the rear.  In the interim, I need to explore a fuel cell so we can run less fuel without starving, allowing me to ballast up the rear a bit.  I'm starting to wonder in general about just running more fuel after seeing Sam Strano, a man who is already heavily ballasted, running a full tank of fuel in his SS C6 Corvette, because the extra power application he gets out of the weight over the rear is worth the weight gain from the gas.

An unrelated huge problem this weekend was that my front splitter wasn't playing nice with the Pro light sensors. I could get the car to "stage" in a couple different places, and it was very hard to tell which one of the two places I was staged in.  As a result, when runs really mattered, I had to run extremely conservative lights (7xx-8xx) just to make sure I wouldn't redlight.  If I'd hit the deep stage hot spot, then I'd still redlight even waiting that long (see my first run of the challenge, for example -- 498 redlight, and I didn't even start to release the clutch until right before the GREEN light came on, so I must have been deep staged millimeters from the next sensor...)  I've emailed Howard Duncan to ask how high the lights are, so I can make some "endplates" for my splitter to make sure that I hit the sensors in a reliable place at the next event.  This problem made every run an unpredictable nightmare.

I also need to just learn how to launch.  I cut everything from a 1.935 to mid 2.2s 60-ft times all weekend, mostly doing around 2.10ish, which is pretty terrible.  On low grip asphalt, even after the course rubbers in, the car has enough torque to break the rears loose at basically any point in first gear, so it's extremely hard to find the balance between a little bit of wheelspin and way too much wheelspin.  I'll hopefully learn eventually, but until then I'll be running almost STS 60 foots while I burn up expensive rubber.

2Jul/120

Packwood National Tour

Posted by David de Regt

The whole month of June has been filled with rainy weekends that we didn't even bother bringing the car out to autocross in.  One of the downsides to running a high power car is that, in the wet, you just get hosed on PAX.  So, as little fun as I have autocrossing in the rain in general, it's also uncompetitive, so the idea of towing for 3 hours each way to get creamed in the rain is unappealing.  Seattle might not have been the greatest place to build this car...

Fortunately, it gave me some time to mess around a bit with the car.  I bought a 12+1 tooth crank trigger wheel from Trackspeed Engineering, and after ~20 hours of trying to make it run reliably, I gave up and switched back to the stock wheel.  I don't know if there's noise on the crank hall sensor from something or if the AEM just doesn't like getting that many tooth updates, but it's off the car until I have time to spend more time debugging.  I replaced both the cam and crank angle sensors with brand new ones from Mazda (they had absolutely no effect on the 12+1 wheel, incidentally,) and replaced the main relay (the relay that broke in Lincoln) while I was in there, which, interestingly, had been superceded by a much different part.  We also learned that we're not allowed to run shoulder harnesses in a non-rollbar-convertible, so we pulled the 4/5 point belts and put in simple lap belts, which removed a bunch of weight from the car and made it a lot easier to get in and out as well.

After having no problem with power application in Lincoln (and even limited problems with power in Packwood,) I upsized the crank pulley (that the SC feeds off) from 120mm to 125mm, and spent a bunch more time on the dyno.  We're up to 330rwhp and 258rwtq and spent a lot more time on the low RPM portion of the map.  The car is now faster and has much smoother power ramp-up, so it's more drivable too, and is better at putting its power down than it used to be.  I think this will be the end of my attempted power mods before 2012 Solo Nationals.  It's time to work on the suspension setup and on my driving.

The 2012 Packwood National Tour arrived in a real hurry.  I spent 2 weeks messing with the trigger wheel to no avail, and had to rapidly switch back to the stock wheel on the Monday night before the NT.  Wednesday the car went in to get a new muffler, since we were consistently blowing 102dB at Lincoln (over the limit.)  Thursday I spent 3 hours on the Dyno at Pina Motorsports, and then 6am Friday morning we pulled out from Redmond to head down to Packwood.

The practice course was the dirtiest (gravel/etc.) course I've ever driven on, but we did some rough shock tuning and the car was working really well.  Ironically, both of our paddock-mates' cars broke, so we spent some time helping fix them.  The rest of the weekend, the car performed perfectly (despite us driving it inadequately,) and I managed to put a 1.776 second victory on the class, running fairly consistent times.  At the end of the weekend, we rolled the car onto the trailer and drove home, ready for the Pro in a few short days.

We learned that the car's fuel starving at around a third of a tank, which will require some investigation after the ProSolo.  I am also sure at this point that the car will be faster with more confidence-inspiring brakes, so that will be another focus for after the ProSolo.  The traction control also needs some further tweaking, and the wing really needs some string+video attention and a Gurney Flap.  Time, time time...

Here's some videos of my fastest runs for the weekend (expand for 720p.)  As you can tell, I'm still driving the car incredibly abruptly.  It's still more than a bit of a shock, coming out of stock class.  At this point, I have at least as much speed to be gained in learning to drive than will be found with further car setup...

29May/120

Second Packwood weekend and Spring Nationals Leadups

Posted by David de Regt

It’s been an incredibly hectic two weeks.  I spent a couple days tearing into more of the electrics and found that the fuel pump wire down the driver’s side of the rear harness had rubbed through to the chassis, which is probably what started everything.  I ran a new thicker gauge wire and re-loomed/moved some of the harness, so it should hopefully never rub through again, and the thicker wire will be less prone to heating.  I also ordered a Deutchwerks 300LPH pump due to my decreased confidence in the Walbro, intending to use it as a spare if the Walbro died.  Finally, I moved the fusebox out from under the side of the engine bay into a spot near the intake with some airflow, in case it just needed some ventilation as well.

The subframe that we were told existed wasn’t actually the right subframe, which we didn’t discover until after close of business Thursday night.  We got to wake up early Friday morning and call every parts shop in the area to find a new subframe.  The only one we could get before Monday wouldn’t be pulled until 5pm that night, 2 hours from home.  So, we picked it up, got back home around 7:30pm, and spent the entire night swapping the subframe and front control arms.  Shaikh finished up the shocks Thursday night and overnighted them so those arrived Friday morning, and when we finished with the subframe, we got to also assemble/swap the new shocks (mounting canisters, etc.)  Around 11:30pm we finally finished with all the work, then spent until almost 2am doing a ride height set and string alignment.  We realized the alignment wasn’t going terribly well, and we couldn’t get enough camber out of the front anyway, so we called it good enough for a shakedown and went to bed.

We decided that waking up in 3 hours and driving to Packwood for the morning session was out of the question, so we went down for the afternoon, which was a nice dry day, and had no major issues at all with the car.  The car was pretty loose, but still drivable, and we settled into liking a RaceLogic setting of 20% slip and 5.0mph wheel differential.  Sunday morning we woke up to a wet ground and ongoing rain, and we ran a very wet morning session, where we got to throw on the Hoosier Wets (225/50/15 H2Os on 15x9 6ULs) and play with the RaceLogic to figure out a wet setup, which worked out quite well.  We tried 5%, which seemed pretty good, but 10% was really the sweet spot, if you had quick enough hands.  The car is remarkably drivable in the wet, which is good to know.  On a very representative course, I was only 0.4 behind the Hyman GT-R car in full wet trim, which is closer than I would have hoped to get.  We had no issues with the car in the wet either, called it a day, and drove home.

On the tow home, the alternator on the truck let go, though, adding some more insult to the existing injury.  Nothing is ever easy...

17Jan/120

Delays and the Front Harness

Posted by David de Regt

The story of this build has mostly been things taking longer than I thought they would.  I was expecting problems like things being more expensive than I thought they'd be, but I wasn't actually expecting so many things to just take exponentially more time than I was hoping.  Almost every part of the project is a month behind at this point.  Some of these things aren't a big deal, but others are.

  • The Quaife gearset I ordered still hasn't left the UK yet, and won't until at least friday of this week, despite that it should have been in the builder's hands almost 2 weeks ago according to their initial promises
  • Scope creep attacked the uprights project like a banshee.  We kept wanting to measure more things, causing me to order new subframes, uprights, control arms, etc.  We finally got everything bought and in my garage, and at this point Brian has most of the parts measured, but we were hoping to be engineering the manufacturing process by this point.  This has also delayed figuring out the shock/spring/bar package that we're going to use, but we have plans to powwow Wednesday night to determine the CG of a stock Miata as well as start doing the math on the suspension package.  This may be further delayed by the impending Snowpocalypse, though.
  • Corky Bell is an awesome old guy who knows way too much about superchargers and Miatas in general.  Unfortunately, he's also busy and very very slow.  So, I'm still in final stages of discussing/negotiating what he's going to fabricate for me.  I think we're nearly done, and he can start making stuff soon, but I was hoping to have an intercooler/radiator/intake manifold setup by now, instead of them all still being discussed.  It means that I'm going to be getting the car running again on the 99 motor on the stock intake manifold before I even start messing with the supercharger aspect. (not the worst thing in the world)
  • Getting all the right parts together for the motor took a while.  The bearings didn't come with thrust washers, so we had to source those; then we had a debate over what to do for valve locks; the windage tray didn't quite fit right and needed to be modified; etc.  Mid-build, I decided to switch to doing the coolant re-route, so I had to order a new head gasket.  Fortunately, the builder says he'll finish the motor sometime today, so I can hopefully pick it up and finish the assembly today or tomorrow.

The biggest fairly unexpected delay so far is how long it took to do the front harness correctly.  The front harness in the Miata is what manages the vast majority of the electronics in the car.  It connects the ECU to the fusebox, the dash, the engine sensors, etc.  The pictures to the right shows its extent.  It's huge, heavy, and invades every portion of the front half of the car.  As it came out of the car, it weighed 17.76 lbs.  And, as it turns out, I was going to change a huge portion of it.

There were several main goals with the harness modification:

  1. Remove everything allowed to under the rule set to save weight/complexity
  2. Convert from the 90-00 to the 01-05 ABS system
  3. Convert a bunch of sensors from the 1.6 motor to the 99 stuff (cam/crank angle sensor breakout, knock sensor) and other applicable sensors too (adding wideband oxygen sensor)
  4. COP (Coil-On-Plug) ignition conversion
  5. Convert to speed density air metering (MAP/IAT instead of AFM or MAF)

I figured, "how hard can this be?  It's just a bunch of soldering."  Well, kinda.  Mazda kinda makes their harnesses like tanks.  Most wiring in the harness is covered with a thick layer of electrical tape, most of which is then covered in a plastic stretchy loom protector, which is then coated in a final layer of electrical tape to hold the slide-on loom together (and presumably to further protect it).  Good for the car, bad for making it easy to hack apart.  It took a couple hours alone to safely remove some of the coating for the first parts I wanted to attack.  From there, the scope of the problem laid before me became more clear.  Once the basic stuff was stripped, I laid the harness out in the general pattern that it sits in the car and got to work ID'ing wires and sensors.  The picture at right is the middle of that process.

After spending at least 20 hours on this damn thing, my conclusion is that either the harness was designed in 4-5 successive stages by a bunch of people as they added more things to the car and moved them around, or the one person who designed it all at once is an idiot.  There are multiple large gauge wires that split off from a source and then run the entire length of the harness, only to split somewhere else again and go to 2 different sensors, there are wires that go long distances only to split and then come back another long distance.  So, just when you think it's safe to snip a wire out, because you're near the end of a harness strand, and you don't need that sensor anymore, make sure to unravel the coating all the way up the harness to the end.  A distressing amount of the time, the harness will split further up, and a smaller gauge wire will share the signal and travel back out the harness to another place, and you would have hosed that other poor sensor.  After realizing all of this, things went much slower.  On the bright side, it meant that anything open under the rules (anything powertrain-related, and some other stuff) I could tear apart and redo properly.  I have a crate half full of just wires I tore out of the original harness that I don't need anymore.

It pretty much just took forever.  There were hours of researching, poring over and comparing parts diagrams for a 91, 93, 99, and 05.  I had to make some executive decisions that may come back to bite me.  For example, the 01+ ABS unit just has some power leads that don't exist on the old car, and aren't really documented where they come from on the wiring diagrams I was able to find, so I guessed.  It also has a bunch of large gauge ground wires that go into the main harness on the 05 that have no counterpart at all on the 1.6, so I decided to do what the 1.6 ABS unit has and made a giant ring terminal out of all the grounds, which will bolt to the chassis, and I'll just go ahead and pray.  I also realized, the day I needed some sensor connectors, that the 05 front harness didn't include most of the engine sensors, so I needed to buy new connectors in a hurry.  Flyin Miata then sent me a wrong connector which took a little while to sort out.  Lots of little delays like that happened.  In the end, I learned a lot, and I tried to be really careful with all of my connections, since, once this is all in the car, debugging anything that I may have screwed up is almost impossible without tearing the entire front half of the car apart again and starting over.

The harness is almost done at this point, as the picture to the right shows.  The loop of whiteish wires at the center are the cam/crank sensor wires, and the coiled red/black wire are the IAT sensor, since I have no idea where I'm going to put it yet (we haven't finalized an intake manifold setup.)  It's 12.2 lbs right now, which is probably the lowest it will be.  I'm waiting to get the motor back from the builder so that I see exactly where the cam and crank angle sensors are, so that I can measure how long to make the wires to get to them, but the wires are run to approximately where they need to go, and are several feet longer than they could possibly need to be, so it's only a few minutes more soldering from here.  The Race Logic traction control box just showed up today as well, so that needs to be spliced into the harness, but that's pretty straight forward as well.  Once that's all done, I'll re-tape and loom everything again.  There's lots of ways to interpret the rules, but I'm trying to take the conservative route and will probably end up massively over-looming it.

It feels really good to finally have the hard parts done, and I actually have a decent amount of confidence that I did it correctly.  I'll still be white-knuckling it when I go to start the car, as I rewired almost every single sensor that the engine uses to run, and the fueling and ignition systems will be completely different than stock as well, all running on an AEM EMS that I've never touched before.  I'm sure that will be a delightful evening of swearing.

Updates should be coming more fast and furious-style now.  Things are finally falling into place in a more rapid fashion.  I get the motor today or tomorrow morning, tomorrow we measure the CG of the stock car for some suspension calculations, and once we do that I can tear the new car apart.  I got the car to pass emissions, which was a lot of fun since the car came with an exploded catalytic converter and a dead O2 sensor.  Finally, I got word from the Oregon DMV that they've "processed" a title and it's in the send queue.  Between those two things, I'm good to dig in.