Kevin’s 91 AWD BG Tracer build

Last Updated on July 23, 2024 by Mutiara

After spending three years on the Miata and taking third last year, it’s time to head in a different direction. Something new and quite different must be built – something other than a car with a turbo slapped on and a paint job.

Enter my ’91 Mercury Tracer LTS. That’s right, a four door sedan. Last May (2007), I came across a post wherein the poster stated his co-worker was getting rid of a Tracer, trim level unknown. It was stated that the head gasket was blown. The cost? “Just make it disappear and it was yours”. My first car was a ’96 Ford Escort sedan, so I have a bit of a soft spot for these cars and I couldn’t beat free. I was in and committed to pick up the car. I was fully expecting it to be equipped with the lethargic Ford 1.9L SEFI and automatic transmission found in most of the specimens.

To my surprise when I arrived, I first noticed the identifying lip wing of the LTS, then looked inside and noticed it had a manual transmission and power-nothing. Super-score #1. Ultimately, the car is equipped with rear disk brakes, Mazda 1.8L BP, etc. Basically, it’s a first gen Protege with all the Mazda gear in a slightly different package.

Moving on, the engine indeed had a blown head gasket and severely seized water pump. I picked up a replacement engine from a friend for the price of $50. A weekend later and a bit of wrenching and she ran perfectly. The car sat until November or so while I tried to decide what I was going to do with it. The GRM Challenge was always an option and after weighing all of my choices, I chose to go to with a second challenge build .

In order of the plan, sure there will be suspension work, a nifty paint job, and obviously a turbo. But that just isn’t enough. I thought I might undertake the task of making all four wheels turn under power. That’s right, AWD. There are several Mazda AWD options out there, foremost, the 323 GTX.

I was fortunate enough to score a rather rough ’88 GTX, all gear included with exception of the turbo, manifolds, and drive shaft. Price: $200. Super-score #2. As a side note, the seller originally offered it to me for $250, but upon arrival, he knocked off $50 AND gave me a beer. He was extremely excited to see it go (that’s a story in itself).

The build started several months ago , actually, but I thought I’d make a bit of significant headway before I threw up a build thread only to later discover that things just wouldn’t work out. That’s not to say that they still won’t in the end. There’s a TON of work left to do.

For those not familiar with the challenge, information on the $2008 Grassroots Motorsports Challenge is grassrootsmotorsports.com/events/2008-challenge

Below is the donor vehicle.

Below is a comparison of the two transmissions and what I have to work with. They’re quite a bit different and really aren’t meant to be interchangeable. Fortunately, the GTX and Tracer flywheels, clutch disk, and pressure plates are identical.

It’s pretty nasty as it is covered in oil residue and general funk. I’ll clean it, as well as everything else, later as I plan to first completely assemble everything, then later tear down of the car to clean and paint.

The best place to begin when wanting to drive the rear wheels is obviously moving the power from the engine to the back. Fitting the AWD transmission is the first step. I knew this wouldn’t be easy and that has proved true. The only thing that lined up properly between the BG chassis and the BF transmission were the block to transmission bell housing bolt points. Even the BP oil pan bolt holes don’t align. Once I actually got the transmission bolted to the engine, the headaches began.

The front subframe was the biggest hurdle. Lots of cutting. I have spent innumerable hours trying to get it to clear…..measure, drop the engine, drop the subframe, cut the subframe, install the subframe, raise the engine, measure, drop the subframe….rinse/wash/repeat. The subframe has been off of the car no less than 15 times. As it turns out I removed way too much metal and had to rebuild a bit of the center of the subframe. Using scrap 10 gauge steel plate I recovered from the work dumpster, I cut and welded plates to cover the transmission clearances in the subframe.

Rebuilding the subframe allowing transmission clearances.

Did I mention that nothing fit? This includes the angle of the engine. In the BG chassis, the engine is tilted forward about 8-10 degrees. The BF chassis has no tilt in the engine. With the BF AWD transmission bolted to the BP using the stock engine angle, the rear output shaft pointed upward toward the floor pan.

At this point, one might ask, why not just use a J-spec (JDM, y0) BG transmission. The primary reason is the cost. I’d rather offset cost by labor, since this is a hobby and all. Furthermore, that’s the goal of the challenge: minimal cost.

Anyway, the next best place to give focus to would be the passenger’s side engine mount, considering the remaining three would have to be custom made. I have tried various mounts from various cars throughout the junkyard, but alas, nothing seemed to provide the proper arrangements, though several were very close. The current arrangement includes 3 washers under the aft bolt (there are two). This will be corrected and finalized later . However, the engine sat as I wanted it to.

The first mount to be manufactured was the driver’s side transmission mount. Using most of the BG chassis mount as it sat, I made an adapter plate to bolt to the tranny. The spacers were made using a hole saw cut through a 1” steel block.

The front transmission mount was frighteningly easy. I perused the junkyard for a mount and/or bracket that would be suitable. I came across an Escort LX-E (the same thing as the Tracer LTS – Mazda bits and all) with an automatic transmission. I noticed the front mount bracket shared a similar bolt pattern to that of the GTX’s. In short, I brought it home, bolted it to the transmission using 2 of the 3 bolts and began fitting the actual mount. Fortunately, with the mount bolted the the bracket, 3/8” steel plate fit the gap between the mount and the cross member perfectly while minding the desired vertical engine geometry. A bit of cutting and drilling and the plate was fitted. To complete the attachment of the bracket to the transmission, the third bolt hole had to be drilled at a lower position.

2 out of 3 mounts were set, and the engine finally rested without the assistance of a lift. Yay! The first two images compare the differences in the mounts (From left to right: GTX, BG-Auto Transmission, BG-Manual Transmission).

Of course, no one said this would be easy. During the fitting of the main subframe, the cross member was modified to fit the front transmission mount and clear the transmission rear output housing. A bit of cutting, a piece of rigid gas pipe, steel plate, and welding and it’s all together. I need to make one more notch on the driver’s side of the piece to clear a bolt in the tranny, but I’ll also save that for later.

After reassembling everything and hoping for the best, the transmission made contact with the steering rack. I had been expecting this, but was hoping to avoid it. Nonetheless, it was a situation that required correction. Eventually, the steering rack was raised 3/4” to provide ample clearance of the transmission. This obviously creates another inherent issue of the actual steering column becoming too long. Fortunately, a roughly 3” “extension” of sorts was build into the column. I was able to cut the necessary length from this and reconnect the column to the rack. The angle of the various universal joints in the column is a bit more “acute”, for lack of a better word, but I don’t think it should present a significant problem.

Currently, I’m lacking a mounting point on the driver’s side of the steering column as it had to be cut off while clearancing the subframe. I will later add this bolting point during the teardown/rebuild of the car.

Ahh…the all important aspect of getting power to the wheels. Better yet, how about doing so with parts from no less than 5 different car models (Tracer, 626, GTX, Ford Aspire, and Mercury Capri)?

The alignment of the entire forward drive line has been a challenge to say the least. The GTX transmission moved the axles forward about 2.5” from where they were in the tracer. That leads to some funky axle geometry and reduced axle life. The goal was to find a way to move the wheel center’s forward and therefore increase the car’s wheelbase.

Enter the 3rd gen Protege. I updated the front control arms on a friend’s car about a year ago and held on to the old ones for the reason of “just in case” (I never throw anything away). Sure enough, the length and width of the CA was exactly what I needed. The first drawback was that the rear mounting point was a rubber bushing as opposed to bracketed rod found on the Tracer. The second was that the ball joint pin was slightly larger not allowing it to slide into the bottom of the steering knuckle. During a junkyard search for some parts, I noticed that the control arms found on the 2nd gen MX6/626 were nearly identical to those found on the later Protege, but the rear mounting point was a bracketed rod, nearly identical to the Tracer’s. That eliminated a significant amount of customization for said mounting point. I simply manufactured a spacer for the rod, trimmed the inboard pinch weld, and trimmed the front bushing tube and it fit perfectly and placed the knuckle nearly where I wanted it to be.

The first image compares several control arms that were likely candidates. From left to right: Tracer (BG), GTX (BF), ’96 626, ’99 Protege. The second image depicts the comparison between the 626’s and 3rd gen Protege’s control arms. The third shows the geometry differences between the Tracer’s and 626’s control arms.

Considering that this is a budget build, custom axles aren’t in the budget, at least right now. Unfortunately, those from the Tracer don’t fit the transmission. Therefore, I was forced to use the GTX’s axles. Furthermore, this requires the use of the GTX knuckles and spindles. However, using the GTX knuckle/spindle assembly required the use of 4×114.3 bolt pattern wheels as opposed to the 4×100 bolt pattern wheels that are more commonly found………….what a mess.

Bolting up the GTX’s knuckles would have worked just fine. However, I noticed that the the steering arm between the GTX’s knuckle and the arm on a knuckle I had lying around from a Capri were different….the Capri’s arm was about 2 inches lower. If I were to use the Capri’s knuckle, bump steer would be improved once the car (Tracer) was lowered. Additionally, the spindle between the GTX and the Capri are identical.

Being the stubborn person that I am, I’d rather not buy any additional wheels as I have several sets employing the 4×100 bolt pattern already laying around. My goal was to find a way to use those wheels with the Capri knuckles without the use of any adaptors or the like. My answer was found in the Ford Aspire.

The Aspire’s spindles share the same axle diameter and spline count as the GTX/Capri. It uses the same bearing part numbers as well. I picked up a pair of knuckles from an Aspire at the junker and compared what I had. The knuckle castings were nearly the same as the Capri’s with the exception that the lower control arm ball joint attachment was closer to the hub’s center. This could pose a problem with the suspension geometry once the car (Tracer) is lowered. So the decision was made to use the Capri knuckles with Aspire spindles. With new bearings installed on the driver’s side assembly, all was ready for a test fit between the control arm, knuckle, axle, brake caliper, etc.

Since the 323GTX employs the B6T, the engine block length is significantly shorter than that of the BP that is used in the tracer. This posed a problem when installing the intermediate axle shaft; none of the mounting holes aligned on the longer engine.

I discovered (the hard way) that the bearing support for the intermediate shaft was cast iron and not very weldable. I therefore fabricated a bracket of sorts to “pinch” the bearing support against the block and securing it with the mounting points found on the BP. The bearing support is secured by 3 bolts and this bracket accounts for two of them. I simply drilled a hole in the bearing support to allow access to the third bolt hole.

With the complete reworking of the front drive train, another problem was presented, of course. Since the lower mounting point of knuckle was moved forward nearly two inches, ultimately increasing the wheelbase of the car, the caster angle was severely changed. Furthermore, the OE springs rub the unibody. This will be corrected later when time permits to run through everything with a fine tooth comb.

As anticipated (read: hoped), the Tracer’s brake caliper affixed to the knuckle perfectly and the rotor was centered in the caliper as it should be. Whew…that was easy.

Now that the front drive train is nearly complete, the time arrived to sort out the gear shifter and linkage. The BG uses a rod actuated shifting mechanism whereas the BF uses cables. The BG’s shift linkage resides beneath the car in the exhaust tunnel while the BF’s cables reside on the interior passing through the firewall.

The first step was to fit the GTX’s shifter assembly. Both the Tracer’s and GTX’s assembly is secured by four bolts. The GTX assembly’s two forward most bolts matched perfectly to the Tracer’s mounting points only requiring two holes to be drilled in the unibody for the rear mount points.

The shift cables were installed just as easily. To my fortune, a hole exists in the firewall of the BG that perfectly matches that of the BF. It was covered only by a simple plate and even has the two mounting studs to secure the cables in the exact manner as the BF.

I simply removed the blocking plate, passed the cables through, and connected them to the shifter assembly and transmission. Gotta love it.

At this point I’m into this build with less than $500 spent, believe it or not! With the front of the car at a point where I’m sure everything will work as I would like it to, it’s time to see about making the rear wheels turn under power.

I started with removing the rear subframe from the GTX. Fortunately, nearly everything came off relatively easily as rust was minimal. The trailing arms are still stuck to the hubs and will need to be removed, but I’ll get around to that in the next few days.

It’s interesting to note that the suspension geometry and parts between the two cars are nearly identical. The trailing arms and lateral links share exactly the same measurements, however the Tracer’s trailing arms are a touch beefier and the GTX’s rear most lateral links are both adjustable where as only one of the Tracer’s is. At this point, it looks as though the GTX’s rear wheel spacing is about a 1/4” shorter than that of the BG’s.

Of course, the entire rear end of the Tracer had to be disassembled along with it went the fuel tank. I’ll have to use the GTX’s fuel tank since it allows clearance for the drive shaft whereas the Tracer’s does not.

Both the Tracer and GTX’s rear subframes mount to the chassis at 4 locations, 2 per side. However, the locations are quite different between the two cars. The GTX has 4 studs protruding from the chassis and the subframe is secured by nuts. The Tracer has 4 threaded bosses and two alignment dowels and the subframe is secured by bolts.

To properly align the GTX subframe, I flipped the Tracer’s subframe over and secured it to the GTX’s using common mounting points between the two, namely the inboard lateral link attachment points. This ensured proper alignment so that I could mark and drill the holes for the Tracer’s alignment dowels and bolts. After measuring several times, I was convinced that everything was square and aligned properly and thusly marked and drilled the holes in the AWD subframe. I’m only able to use one of the original bolts per side to temporarily secure the subframe to the Tracer as there isn’t enough clearance to allow use of the other two bolts.

Currently, the subframe is held in place only by those two bolts. I will mark each of the original mounting holes of the GTX subframe onto the Tracer and drill them through the unibody “frame rails”, running bolts into the trunk floor. I will also need to fabricate a few spacers to allow proper seating onto the unibody rails as it is loosely sitting in place.

A quick measurement indicated that everything will line up perfectly from the front to the back of the car. Furthermore, it appears that all of the rear suspension will also align properly. This was a little too easy, I wonder what hang ups lie in wait.

I will continue working on the rear throughout this week and hopefully have it completed in the near future. Currently, the biggest hangup I foresee is the driveshaft, foremost, locating one. Secondly, I’ll need to have it sized to fit the car.

Looking back, I neglected to mention the steering rack. Since I neither care for power assisted steering nor the added weight to the car (at least when it comes to competition), I decided to de-power the power steering rack. Three options are available: by-pass the PS pump and leave the rack untouched, source and install a manual rack from an Escort LX Pony, or remove the inner workings of the rack that work with power assist.

It’s difficult to find a Pony manual rack and I already had a perfectly good power steering rack [that came with the Tracer]. As always, I chose the more difficult route. In the past, I’d de-powered a couple of Miata PS racks and thought this one would be no different. It’s a typical rack and pinion so all that was necessary was to disassemble the rack and remove a metallic bushing (I have no idea of its proper name) and o-ring. It’s held in place by a pair of recessed circlips and is much easier to cut off. After making a pair of cuts on both sides and a quick swack of a hammer, it was time to reassemble. Furthermore, there are no costs associated other than a handful of grease.

I put in a touch more work to finalize the rear subframe attachment this evening. Since the subframe was already aligned squarely with the original alignment dowels, I proceeded to drill through the unibody rails into the trunk floor using the original GTX subframe mounting points as a template. Then using threaded rod cut to length, secured it to the car. To ensure a square fit and overall proper alignment, I added make shift spacers above the original Tracer mounting bolts and aft-most GTX mounting point. The front most mounting point already sits flush with the unibody. I’m not happy with the rear spacers, but I’ll leave them in place for the time being and will tidy it all up a bit upon final reassembly. I may box-in/reinforce the whole area with 12 gauge plate. Furthermore, I added a square of 1/4” plate above each bolt that protrudes through the trunk floor to add a bit more rigidity to the unibody sheet metal. This too may be changed later. I’ll have to see how the subframe reacts whenever I finally get the thing on the ground .

As the car sit now, everything is [loosely] assembled in the rear. Everything fit as it should. Furthermore, the input of the rear differential aligns with the output of the transmission. I’m using the Tracer’s trailing arms as they share exactly the same dimensions of those from the GTX but beefier. I’m also using the Tracer’s struts as the GTX’s top mounts don’t fit the tracer’s strut towers. I had to drill a couple holes in the lower mounts to fit the GTX’s knuckles. However, that’ll all change when I get around to finalizing the suspension – that’s all temporary for the moment.

The GTX’s spindles are still installed but will be swapped out for the Aspire’s front 4×100 spindles as soon as I source a pair. I’ve already confirmed that the spline diameter and count matches the front and that the bearings are the same size.

All that currently remains to do in the back of the car, aside from that mentioned, is to sort out the hard brake lines, mount, the rear brake calipers, mount the GTX’s fuel tank, reinforce the subframe mounting, and tighten everything up. I’m really surprised, and nervous, as to how easily everything went together.

It’s been a while since I last posted, but work and life in general have kept me aloof. In the time I’ve been able to wrench, several significant snags have crept upon the build.

As the car stands, three corners are bolted together and pretty much ready to go. I finally came to a point which allowed me to tackle the front right corner. The strut has been modified to fit the Capri knuckle and everything was ready to bolt together. In doing so, the axle seamed to want to bottom out and bind. After an hour long staring session, a touch of measuring, and about three beers later, it was clear that the front passenger side axle was too long…by about 3/4”. Whoops. Everything fits perfectly on the driver’s side including the amount of free play in the axle. The opposite is the case on the passenger side. I’m currently researching my options, but it’s seeming like I’ll have to fork over the cash to have the axle shortened. That’ll be a good ding to the budget for such a small change.

Moving to the back of the car, I’ve also been busy with the fuel tank. I have removed the fuel tank from the GTX and cleaned it. Aside from draining about 9 gallons of fuel, I also removed an measurable amount of sand and various other types of debris. I haven’t a clue how sand found it’s way into the tank. What I do know is that the car spent most of its life in Savannah, GA. I have found sand in just about every nook and cranny that I dig into.

After trying a number of ways to fit the GTX’s tank underneath the Tracer, I have come to the conclusion that it just isn’t possible without 1) modifying the tank or 2) cutting and rebuilding the rear seat pan. In the Tracer’s current state, the GTX’s tank sits unacceptably low and wouldn’t allow the drive shaft to clear. The first object that the car travels over will likely damage the tank.

The first option is out of the question and the second is somewhat feasible. To satisfy the second option, I would either have to cut out the Tracer’s seat sub-pan and rebuild it with various pieces of new metal or cut and remove the seat pan from the GTX and graft it into the Tracer. Yikes! Neither task sounds enjoyable.

I stumbled across a third option, however. Six years ago, upon disassembly of a wrecked IMSA racer, I removed an 8-10 gallon fuel cell and threw it into the boss’s pile of junk. It has been sitting there ever since since and had little likelihood of ever being used. After making an offer of $50, it was mine.

The fuel cell will me mounted in the spare tire well of the trunk as that is the only place it will logistically fit. I’m not yet sure if it has an internal fuel pump, so that’ll have to be determined and sorted. Furthermore, the mounting bracket(s) and fuel lines will also have to be manufactured.

On a positive note of the build happenings, I have acquired a worthy suspension setup for the car. With many thanks to Luke (Rogue), I purchased his old Protege5 Tein suspension for a fair price. Like all other B-series chassis’, the Protege5 has all 4 corners suspended with McPherson struts.

To fit the P5’s struts to the Tracer chassis and the aforementioned Capri front knuckles and GTX rear knuckles, the lower mounting points need to be modified. The upper mounting hole of the both the front and rear struts was lowered and moved inward to induce a fair amount of negative camber. At static ride height, found by placing a jack under the lower control arm, the front camber angle is at about -2.5°….a good starting point for a front Mac-strut. I still have to complete the assembly of the rear struts, but it’s looking like rear camber should be a about -1.75°.

In addition to modifying the mounting holes, I had to incorporate spacers between the knuckle and receiver on the the strut. A 3/8” plug worked perfectly for the front. As seen in the below images, I pushed the front strut toward the rear when installing the spacers to reduce the already severe caster angle. The rears knuckle, when finished, will be centered in the strut receiver.

Finally, I’m employing the P5’s top mounts on the front and the Tracer’s top mounts in the rear. The Protege5’s top mounts are significantly improved over those of the (also worn) Tracers and bolt right in. However, I’m using the Tracer’s rear mounts as they mount via 2 bolts whereas the P5’s mounts via 3.

I’ve been busting my hump and really haven’t seemed to accomplish much. I still haven’t solved the FR axle issue, I still don’t have a drive shaft, and now it turns out that the trailing arm bolt receiver on the rear left knuckle is severely stripped. Apparently, some schmoe at some time decided to drive the the bolt in with an impact wrench. I attempted to re-tap it tonight only to actually strip the tap. How does that happen? Thankfully it didn’t break. Regardless, I’ll try again with a new tap. The previous owner of the GTX swore that the car was cursed. With the parts and problems I’m moving over, I’m starting to believe him.

Aside from the frustration, some progress has been made. I’ve completed the rack/frame for the fuel cell. It turns out that the cell came with an internal fuel pump of unknown flow rate. Upon disassembly, I found that it was seized. A couple of taps on the floor and the pump spun again. I’ll recondition it with a bit of injector cleaner and hope for the best….it’ll be one less thing I have to add to the budget.

In other news, I have fitted an Aspire front spindle into the GTX rear left hub to allow for the 4×100 wheel bolt pattern. As I’d hoped, the spindles, bearings, and axles mated perfectly. However, the Tracer’s rotors don’t sit as far back as the GTX’s. I attempted to fit the Tracer caliper and bracket to the GTX knuckle but the mounting spacing doesn’t properly align. Overall, the Tracer’s caliper just wouldn’t work out. Of course, the GTX caliper and bracket fit, but it didn’t clear the Tracer’s rotor. To remedy the situation, I’m currently using a 1.6L Miata rotor and had to fabricate a spacer to move the caliper outward about 3/8”. Sadly, the Miata rotor is about 1/4” too small in diameter to use the entire brake pad surface. I’ll have to investigate junkyard rotors and hope I find something that works out.

Sadly, I won’t have much time in the upcoming weeks to progress as I have a business trip, vacation, and a “24 Hours of Lemons” race scheduled. I’m booked solid to say the least. I hope August yields good results.

Progress is slow. Between work, travel, and just generally very tedious progress, not much seams to have been accomplished. I’ve completed the rear right corner for all intents and purposes, I just need to clean up the brakes. I tossed on some E30 “bottle caps” to see how I like ’em and I think they’re gonna be a keeper. There’s something about the design and the relation to the old Oz (?) wheels used on rally cars.

I’m currently working on a solution for the front right corner wherein the axle is too long. I hope the solution lies in extending the control arm. I’m about half way though it’s completion, but have no pics yet.

After fooling around for so long in an attempt to figure out what to do with the front right suspension and axle issue, I finally came to a solution. The answer was found in simply extending the length of the control arm. I’d been hesitant to try this method as I was worried that I wouldn’t be able to maintain the trueness of the arm with the tools I have on hand. In the end, I cut the control arm about 2” behind the ball joint and inserted about 1” of material. When fitting the modified arm, it fit perfectly and allowed the proper amount of play in the CV. Whew.

In previous attempts, I tried to dismantle the axle so that I could have a new center shaft fabricated, but I wasn’t able to separate it completely. Furthermore, using a spare control arm, I cut out the ball joint with a hole saw and moved it further out via a piece of 1/4” steel plate. However, during welding, I cooked the nylon liner that supports the actual ball. Doh. I wasn’t aware that a liner existed. You learn something every day.

While assembling the rear struts to the knuckles, a significant amount of force was required to push them inward and allow them to mate with the knuckle. This concerned me as it places undue lateral stress on the strut rods and will induce accelerated wear.

Upon closer inspection, it turns out the angle (I don’t know the correct term unfortunately) at which the Protege5 top hats direct the struts is much more pronounced than that of the Tracer’s. To correct the issue and orient the strut angle to more closely match that of the Tracer’s, I cut about two thirds of the way around the perimeter of the raised area on the P5’s top hats. I then applied force to adjust the angle and re-welded the cut. I suck at descriptions tonight so just look at the pictures.

This weekend I’ve moved onto the interior of the car. This included removing anything from within that was bolted down, including the dashboard. The carpet was disgusting and I’m not at all unhappy to see that go. The dash will go back in, but it came out to remove sound deadening and prep for paint. I’ll likely replace the dashboard as this one is quite cooked from years of sun beating down on it. It was already cracked and just the lightest pressure fractures the surface. Doh.

I removed all unneeded wiring from the cabin harness pertaining to the interior lighting, sound system, and automatic seat belts. This left a nifty pile of wire behind. I also got a jump on removing the tar-based sound deadening from the floors. Fortunately, it’s coming up fairly easy, but still takes more time than I’m willing to devote to it. Lastly, the fuel pump within the fuel cell is now wired into a pluggable harness.

I have a 12x12x12 inch box that I have been throwing small items such as screws, brackets, clips, seat belts, trim, etc into. It’s amazing how the small stuff adds up as the box now weighs in excess of 20 pounds.

In the middle of the tear down, I threw a battery in the car to make sure all of the aft lighting still worked after tearing into the cabin harness. For fun, I cranked the car (without fuel and ECU). It was rather nice hearing a bit of life come to the car after sitting for so long. The front wheels did turn as I had the tranny in gear. Cool.

The second image shows how bad the hail damage is on this car. Who knows when it happened, but it had to be a pretty good storm. Furthermore, when I removed the head liner, I found cardboard shoved in between the roof and center support. It’s pretty funny the things one finds when dismantling a car (I.E. Empty condom wrappers in the GTX).

What a PITA. I hate bodywork. Today I moved on to making the car work for the concourse segment of the challenge. This meant wrangling a few of my beer drinking Hong Norr buddies and tackling the prep work for paint. Since the car will have no interior, this means painting just about every inch of metal exposed….we finished prep on about half of the exterior today, if that. Furthermore, it took all of about 8 hours to get there. I’m not liking where this is going, but I hope the end result will yield the fruit of the labor. I’ve said it during all three build segments of the Miata, and I’ll say it here: I’d rather make a car fast than pretty (and I’m more skilled at the fast part than the pretty part). Enough bitching.

For the first time since I started this build, I sat the car on all 4 wheels. This pretty much gunned me up. Rolling it out into the driveway to minimize sanding dust in the garage, we sanded most of the exterior. The car is riddled with dents from various sources including hail, door dings, and general carelessness from the PO. This required body filling, in which none of us are skilled in that art. Fortunately, though, most of the larger holes were filled and sanded well. Of the 4 panels that were filled and sanded completely, a filling primer was applied. It’s sorta cool to see parts of the car in another color (white). In the end, the car will be primarily metalic silver with blue accents.

Maybe I’m being ambitious, but I hope to have removable body panels in final paint next weekend, and the rest of the car painted the weekend after. If that pans out, that’ll be a HUGE hurdle behind me.

To properly change the color of a car, everything must be removed/detached from it. Everything. Doing so greatly helps in hiding the majority of the half-assidness that typically goes into a color-change respray . Taking an untested car completely apart that hasn’t run in months down to nothing with only six weeks remaining sounds like a grand idea, doesn’t it?

As the car sits now, the only removable parts that remain are the front and rear glass, rear subframe and suspension, trunk lid, and doors (gutted). Everything else is out. With the rear wheels sitting on the ground, I can lift the front of the car by hand with little effort. The complexity of a car really becomes apparent when it is disassembled. I think I can get everything back together . Most of the parts only fit one way and when it isn’t bolted in by the standard metric hardware, it just clips in. Furthermore, the parts usually only fit one way, right? We’ll see.

Painting sucks. The only thing worse is the prep work leading up to it. In all, we spent a total of 17 hours alone prepping the hail damaged and previously spray painted roof.

Even with a full breathing apparatus, the fumes are obnoxious; this paint is nasty. Without a mask, the “booth” would have been uninhabitable. Furthermore, painting the inside and outside of a car is a lot tougher than one might expect. It certainly wore me out. Holding a loaded spray gun at various angles and extensions really gave my shoulders a workout. Whodathunkit?

After 4 fumigated rounds, the shell is painted. I initially laid a bonding/sealer primer to bond with the various surfaces that existed on the car (I.E. 3 types of body filler, existing paint, existing clear coat, and bare metal). All my bases were covered. After the primer cured, the metallic silver was sprayed. The base coat laid quite nicely but once dry, it felt very rough. It turns out that was normal due to the metal flake in which the final clear coat would cure. I was worried that dust had settled into the paint. Indeed, the roughness was corrected upon the application of two rounds of clear coat. The paint looks surprisingly good in person with only a slight orange peal which was to be expected. The spray looks many times better than the paint applied to the Miata (though, I have yet to see the Tracer in direct sun light). Said orange peal will be corrected with a good buffing and compound.

The most difficult portion of the spraying is complete; that being the chassis. I still have the doors, trunk lid, hood, fenders, and side skirts to paint. I’ll complete that within the next few days as I’ll be able to clear the “booth” of the chassis. I can’t wait to start reassembling the bits and pieces. It feels as though a huge page has turned in the build of the car.

Laying the primer.

The following pictures are prior to the clear coat.

The fun begins….final assembly. Everything but the hood, trunk lid, grill, and trunk spoiler are painted. Due to time constraints, I will be painting those items after everything else is done. A fully painted car isn’t as useful or fun as one that actually makes noise and moves itself. I have 3 full weeks left to complete this thing as I will be away for the better part of a week for the SoloII Nationals.

After painting the bulk of the car, assembly has begun. The main cabin harness, blower assy, fenders, doors, door handles, side skirts and a few smaller items are assembled. It’s starting to look like something again . The paint looks pretty good, though it needs a good wet sanding and buffing to really bring it to life.

Assembly will progress throughout next week. Labor Day will yield significant cleaning of the engine and transmission. The turbo manifold has yet to be constructed, but will also be completed next week. Furthermore, the intercooler needs to be mounted and a drive shaft has arrived (YAY!).

Work is progressing and today I was able to accomplish a number of tasks. I completed the rear transmission mount, finalized the cross member, added the 4th steering rack mount to the subframe, cleaned the transmission, and trimmed the doors.

As stated in earlier posts, I would be finalizing the fourth transmission mount and fourth steering rack mount. I finally got around to completing them and am glad to be done with it. The subframe is now finalized as is the cross member.

Due to the space constraints between the transmission and firewall, neither the GTX nor Tracer rear mount would fit. After significantly trimming the rear bracket, I rigged in an Energy Suspension sway bar bushing and employed a 1/2” piece of rigid gas pipe. The combo seems to line up properly, but that will ultimately be seen when everything is assembled.

The transmission is in a lot better state, at least cosmetically than it was previously. After an hour of scrubbing, power washing, scrubbing, and wire wheeling, it’s as good as it’s going to get. I’ll say that it definitely looks a lot better. At least it has some shine .

While I was working on the above items, a friend worked on applying vinyl to the window surrounds and lower window trim pieces. In all, it came out great and added a touch of newness to the car. Later, I assembled the loose trim and mirrors…it’s definitely coming together.

I’m still making headway and really haven’t hit any snags. Everything is going back together. Most importantly, the engine and transmission as well as the axles are installed. The suspension is finalized, short of an alignment, the engine has been cleaned (for the most part), a pimpy eBay special 6-puck clutch is installed, the brakes and clutch hydraulics are filled are bled, the shifter housing is painted and installed, the shifter cabling is finished, and the hand brake is installed and working. The car currently sits on 4 wheel and is looking pretty good, I might say. Once the engine was in and the suspension buttoned up, a critical area of focus was the tie rod ends. With using the Capri knuckles, the BG tie rod ball joint assemblies happened to be too long resulting in about 2” of toe-in with the tie rod ends adjusted all the way in (wow, long sentence). After much searching, I finally found a spherical rod end with the proper thread diameter and pitch (it seems female M12x1.25 ends are hard to come by). The rod ends are about an inch and a half shorter overall than the stockers allowing the full range of adjustment. I fabricated spacers for each side of the ball to prevent binding. The end result worked perfectly.

Moving onward to new material, I’ve finally started constructing the turbo manifold. I originally wanted to build a manifold with equal length runners, but with 3 weeks remaining, that just isn’t going to happen. I fabricated the head-side flange back in January before the project even really started. I’m glad to have that part out of the way considering the amount of time that took alone. Using weldable gas fittings, I constructed a log type manifold on to it. I used a wrecked head I had lying around to weld the assembly. I still have yet to place the flange…that’ll happen tomorrow. Fitting a T3-series turbo between the engine and the radiator on the BG is more difficult than I thought. It’s pretty tight in there!

Beginning of the mo-powa! part.

Tie rods, etc….

I completed the turbo manifold today. My back is killing me . The turbo doesn’t weigh much, but when holding and positioning it to get just the right angle while bending over will where you out. After much tweeking and swapping the water pump inlet pipe with one from the GTX engine, I got everything positioned just as I wanted it. I had to use the GTX’s inlet as it was quite a bit shorter and allows just enough clearance for the turbo inlet. This Thunderbird T3 seems to have become much larger when fitting it into this car. It sits about an inch from the engine block and about a half inch from the radiator. Whew! I used a mock turbo with the mock head to do most of the welding. Expendable parts are always cool to have.

After the manifold was completed, I began construction of the down pipe. This too went fairly smoothly and the only hangup was fabricating a flange to mate with the waste gate housing. Ideally, I’d have used an external waste gate with proper piping, but again, the lack of time prevented that from happening. The outlet of the DP is just at the back side of the engine and I’ll likely throw on a resonator and turn-down pipe. While I was at it, I welded in a bung for an O2 sensor and called it a night.

Tomorrow will consist of bolting on the real turbo, installation of the intercooler, and all of the piping. Following that, the next day, I hope to finalize the fuel cell and maybe even crank the thing. Unfortunately, I don’t have the dashboard in yet, and therefore the steering wheel, so driving it will have to wait a little while.

I made a bit of progress with the intercooler and piping this evening. As usual, things don’t happen as quickly as I’d like. Nonetheless, progress is progress.

I began by fabricating a bracket to mount the IC to the lower rad support mount column (or whatever it’s called). There were two tapped bosses already there to mount a bracket to without having to drill anything. With the bumper in place, I set the angle. I was forced to angle the IC for two reasons. A) to allow the in/out piping to clear unibody members and B) to maximize the lower bumper openings. I think the arrangement will work well. I secured the IC with a pair of aluminum brackets at the top yielding surprising rigidity.

After the IC was set, I began the piping. I cut a hole in the sheet metal just above the driver’s side port on the intercooler and used a 90* silicone coupler to point the piping into the bay. All of the piping is aluminum with the exception of the bend going to the throttle body. Since I don’t have the capability (yet) to weld aluminum, the steel bend at the throttle body allows me to weld a port for the IAC feed line and BOV. As a side note, I hate the fact that all of the couplings are red. It was the cheapest thing available on eBay and really makes for an eyesore. I doubt there’s much I can do about it at this stage.

It’s crunch time. I need to get this done. That said, I’m having to short cut a few things. So far, it only involves a touch of IC piping and the post-down pipe exhaust.

The intercooler piping between the turbo and the intercooler will be finished with a piece of about 12” long 1.75” ID hose from the GTX’s intake piping. However, this isn’t a terrible thing as it allows for engine movement without binding on the hard pipes. In a programmer’s world, this might be considered and “undocumented feature” . I don’t have pics yet of this as paint is drying on that portion of the piping.

Secondly, the biggest rigging thus far is the exhaust. I needed to move the exhaust outlet back as opposed to dumping it from just under the engine. For budget purposes, I didn’t want to spend the money on a length of 2.5” piping. Fortunately enough, I procured a rather nice amount of 3” piping from a trash can at last year’s Challenge (including a Magnaflow muffler with dual outlets). I think it was from a Neon. Regardless, I used it and performed quite the hack job of fitting it to the 2.5” down pipe. This will be corrected after the Challenge. Nonetheless, it serves the purpose.

All in all, I’m about 95% finished under the hood. All of the IC piping is constructed, cut, and fitted. I’m just waiting for paint to cure on a few pieces. I finalized the elbow leading into the throttle body with fittings to allow for the IAC inlet, IAT sensor, and BOV. Furthermore, I have plumbed in all but 2 holes on the turbo: the compressor inlet and water return. The oil supply was made as a hard line from old brake lines . The water and oil supply was sourced at the OE locations on the block that are, by default, plugged from the factory in U.S. applications….reliving the BG 1.8L GTX/GTR days.

Below are a few images I neglected to post regarding the ability to run the car.

Once the intercooler was mounted and the majority of the piping fitted, I had to build a fitting at the throttle body to allow fitment of the IAT sensor, BOV, and IAC valve supply. A bit of test fitting, drilling, cutting, and welding yielded a rather complex contraption.

Once the fuel cell was mounted and, for the majority, plumbed, I had to tap into the Ford fuel lines. Mazda and Ford did things a bit differently between their respective versions of the BG chassis. Ford used a sort of quick connect fittings to connect the lines to the fuel pump whereas Mazda uses simple compression. To avoid cutting into the stock semi-hard lines and fighting with fittings, I cut the original quick connect receivers from the OE fuel pump assembly. I then flared the ends to mate with fittings so that I could easily extend the fuel lines to reach the fuel cell. No leaks and all is well.

Now that the SoloII Nationals are over, I’m seeing the light at the end of the tunnel on this build. Sunday, I finally took the car for a brief drive, minus the drive shaft. I locked her into 4WD to allow the front wheels to be tied into the center diff. The first run yielded stumbling above 2000 RPM due to lack of a tune. I fooled around with the VE bins for a few minutes and got the car to drive reasonably well. The drive was only a quick burst, enough to build a touch of boost….she’s definitely faster than the last time I drove her .

Since the last update, I have installed the dashboard, steering wheel, 4WD lock controller and wiring, temporarily mounted the driver’s seat, and mounted half of the drive shaft.

Obviously, the drive shaft would need a custom mounting arrangement. Therefore, I fabricated a triangular bracket to mount one side of the shaft support to the exhaust tunnel. The other side was secured by a welded piece of steel just beneath the driver’s side seat mount. Fortunately, the chassis steel there was thick enough to provide ample support. As expected, the drive shaft is 3.5 inches too short. I’ll have to lengthen it and hope the balance will hold long enough for the challenge.

Less than two weeks remain and the list is still long. To worsen things, I think I might be coming down with a bug. Anyway, the list includes painting the front grill, C-pillar covers, hood, trunk lid, and trunk spoiler. Outside of paint, I still need to dial-in/tune the MegaSquirt, wire in the wide band O2 (for tuning only), wire in the IAT, secure the front seat, install harness belts, install the wipers, install the rear window trim, and many other little areas that escape my mind.

Nonetheless, the car is pretty much in a position to compete, short of a tune.

The Tracer runs, drives, boosts, and turns all 4 wheels! Not too complicated, huh? Right. The car is nearly complete. Still with a significant list remaining, all of the body panels are painted and fitted, the wheels are painted, and the AWD system is complete. Monday, I will wrap up the front sway bar and rear firewall between the trunk and passenger compartment (required by NHRA regs). Tuesday, I’ll dial in the corner weights and alignment. Wednesday is a catch all day which I’m sure will be full of small odds and ends. Thursday, I head to the dyno and then off to the Challenge! 4 days left.

I had to use the passenger side seat as the driver’s had a huge hole in the bolstering. After swapping the sliders and installing it, I steam cleaned it. The third picture shows the amount of funk I pulled from it. Absolutely disgusting. I may clean it again as I was still pulling brown stuff out. I also mounted a super bright harness I had lying around.

Thanks to Todd and Shuman for swinging by and helping out. Great job on the wheels, guys!

So far, I’ve completed the auto-x and concourse segments. I don’t yet where I stand in the concourse, but the car was received very well. The best part is that the car placed 9th out of 54-ish!!!!! Top 10! I couldn’t be happier. The hard part is over and knowing that I’ll get killed at the strip, the drags will be just for run.

There happens to be another AWD convert, to my surprise, present. A Mitsubishi Mirage. The two cars were asked to pose together for a photo shoot today, so hopefully there will be some magazine exposure.

I’ve finally made it home and can say that I, and the Hong Norr team, had another fun and very successful Challenge. I’m glad it’s done, but I’m already looking forward to next year.

The drags didn’t transpire well, unfortunately. On my next to last run, I realized I was only boosting to 8 PSI. I can go to 15 safely. Woops. I have no idea why the concourse score was so poor (17th?), but I did receive an honorable mention for the best engineered at the banquet. All in all, as always the event was an absolute blast both on and off the track premesis. The car showed a lot of promise, especially in the auto-x. I think I’ll be back next year

The trip was uneventful until I was about 30 miles from Atlanta when one of the trailer tires shredded. It was impressive, really. It destroyed the left side light, knocked the license plate off somewhere in the middle lane of I75 and banged up the fender. It looks like I have a project for Monday. At least I have the time now . We didn’t have a spare tire, so we jammed a peice of wood above the axle that still had a good tire to maintain proper balance and made it the rest of the way home.

Below are a few pictures of the Hong Norr teammates’ racers. The yellow CRX is a wicked screamer in the auto-x and a low 12 second car on the drag. It took the overall win. They blue Civic, aka the s00p3rturd, took 2nd in the auto-x again this year. This was the car featured on the cover of last year’s GRM issue.

The $2009 Challenge is only 4 days away and for once, I’m actually ahead of schedule before the event (knock on wood that it stays that way). The car has been wet sanded, corner-weighted, and in general, combed over. I’ve also cleaned up the valve cover and intercooler piping, rerouted some wiring under the hood, and moved the battery to the back. Boost control is now enabled, so I should see 15-16 PSI on the drags .

I have cross weights dialed in to 99.7%, front to back weights at ~60/40%, and left to right to ~52/48%….not terrible. I still need to finish the alignment and am a little worried about the bump steer that I’m seeing just moving the car up and down, but we’ll see what happens with that. All in all, things look great.

I’ve added some vinyl to the side’s of the car as a throw-back to the old GTX. The cuts were directly modeled after the actual “4wd…” and “DOHC…” found on the GTX.

It’s sort of hard to believe that the Challenge has already arrived. It’s been a quick summer.

Another Challenge is in the books and I think I did well, all things considered. The car was slow as hell through the auto-x and most of the drag. Power delivery issues plagued the weekend, but fortunately, no issues from the drive line which responded well.

Poking around through the day of the drags, I found a few things out of kilter starting with a vacuum leak at the ECU. That fixed, things didn’t improve. I tightened up the waste gate actuator as boost was building very slowly. Things still didn’t improve. Finally, I put a timing light on the engine and found that I’d lost 10* of timing. I measured 8* of base advance when the MS thought it was at 18. A quick adjustment of the dizzy and boom, 2.5 seconds off of the quarter mile (which wasn’t impressive anyway).

In all, I finished 16/53. I’d have been deep into the top 10, maybe even the top 5 in the auto-x had I discovered the timing issue before hand, but I digress.

The event was an absoloute blast as always. I have one more left in the Tracer before I retire it from the Challenge, yet, I have a number of ideas that may bring it closer to the cover. Lets hope that it all comes together .

Below are a few images during the auto-x. The car is at least well composed.

This build brought to you courtesy of Kevin, aka kcbhiw.