March 19, 2024, 02:57:28 AM

Author Topic: Matt's LS-1/T-56 FD Build  (Read 41852 times)

Offline mdickw

Matt's LS-1/T-56 FD Build
« on: September 10, 2010, 02:44:15 PM »
I spent a lot of time thoroughly documenting my build on the.. other forum, but now that's all gone so I guess I have to start over. I probably won't put as much detail, but I'm sure I'll think of new things that I didn't mention before.

I had very little experience working on cars when I started this two years ago so I had to learn a lot as I went along. For every hour I worked on the car, I probably spent 10 times that reading and searching for information, mostly on v8rx7forum.

The car is a 93 that I got at about 70k miles with 5k on a dealer-installed, brand-new rotary. Here's me beginning to pull out the rotary.



With all the crap out from in front of the engine, this is the point when the doubters start to believe there might actually be enough room in there for that "big" Chevy V8.

Say goodbye to the wankel (got $2500 for it on the club).



Of course, before I even did any of that, this thing arrived:



It (LS-1, T-56, PCM, wiring harness--even the exhaust) was out of an 2001 Camaro with 90k miles. I was a bit worried about the high miles (particularly with the transmission), but it has ended up working mostly okay.

Obligatory shot of the empty engine bay:



Then without the front bumper:



And removing interior parts:



A word to the wise: remove the parking brake before you take the dash out or put it in. Takes two minutes and saves a lot of trouble.

I had an NC (06 Miata) and I wanted a steering wheel with more buttons for the FD, so I got an NC wheel on ebay for $45 ($15 + $30 shipping). Comparison:



They're the same diameter, so don't go this route if you're looking for a smaller wheel. It would probably save more than a few grams, though. The splines between the two wheels are identical, but the alignment holes for the clockspring don't match (FD has them top and bottom and NC has them on the sides). I drilled some new holes in the NC wheel and then it bolted on nicely. I got an airbag for it for $100 on ebay that I probably won't hook up (needed it for the cover and horn button, at least). I'll talk about wiring it up later.

Yuck:



Previous owner was a smoker, so the car smelled pretty bad. Worse than that, every interior panel was coated with soot (particularly on the backs of each part). Aside from the smell,the torn-up driver's seat, and some scratches on the center console, the interior was in great shape.

A few months later, after shampooing the carpet:



I went with Grant's mounting kit since, at the time, it was between him and Hinson (and that whole dynotested debacle). I don't really remember why I went with his over Hinson's. Here's a pic of the the new and old subframes:



I cleaned up the OEM subframe and sold it on ebay for like, $30 or something (if that). I also sold the other stuff like the PPF, driveshaft, clutch master and got about $10 for each part.

Then the subframe installed:



And then I see that the intermediate shaft (steering) is not going to make it:



There is quite a lot of wiggle-room in the system and Grant's advice was to loosen everything up, then connect the shafts and tighten everything down. I was able to make that work, but I was not comfortable with it. I eventually got my shaft lengthened:



Machinist friend of a friend cut it in half, then welded a section in the middle and machined it down to the original diameter. It's about 3/4" longer than it was originally. In the end, I still need to raise the rack back up, though (more on this later), so it probably wasn't even necessary.

One of the fun mysteries I had to figure out was how to plumb the clutch. It took countless hours of googling before I finally had a setup worked out. I ended up with this:



I got the wilwood master from hinson with the correct threads welded on. The clutch line goes from the master to a male 3 AN adapter (came with it). On the slave, I used a nifty adapter that Russel makes. You knock a pin out on the slave that holds the quick-disconnect fitting in and then you attach this thing instead (so, no more quick-disconnect fitting). The adapter comes out to male 3 AN so I used a 36" 3 AN (F-F) brake line to connect them. Then there's the bleeder. You'd have to be batshit insane to try to bleed it with the stock bleeder, so a remote bleeder is a good option and very inexpensive if you order all the parts yourself. You need an adapter from 3 AN to 10mmx1.5 (male). You'll need to use a 10mm crush washer and copious amounts of thread sealer to mount the adpater to the slave. I have read some discussion about how this will not form a reliable seal since the surface has not been machined-flat. So, you do this at your own risk, I suppose, but lots of people have done this successfully. On the opposite end, you're going to use a brake speed bleeder and you'll have to find a corresponding 3 AN adapter (then another 3 AN brake line). I used a Russel 3/8" -24 bleeder. For the AN adapter, you'll need something with a bubble flare (this is important) and threads that match the bleeder. I used this one.

I did not mount a second reservoir, but instead hooked it up to the stock brake/clutch reservoir where the previous clutch master was connected. Some people don't like this, saying if the clutch line fails, you'll lose your brake fluid, but if you look in the reservoir, there's a baffle that keeps the fluid separate if it gets below a certain level.

At some point, I unbolted the T-56 from the LS-1 and decided to change the clutch. I went with a SPEC stage 2 (I think) and the Fidanza aluminum flywheel. Here's a shot of the pilot bearing I took out using a puller from autozone:



I did not replace the oil seal when I did this, though now I'm kind of wishing I had. I also didn't replace the flywheel bolts. There was some discussion about this on the old forum with people saying the bolts are supposed to be replaced. None of the instructions I read on doing the job said to do it, so I never had thought about it. I did call up the local GM dealer and asked them what the service manual said (after all was said and done) and they said that it doesn't say to replace--so hopefully it will be okay.

I took the intake manifold off before I tried installing the engine. It was pretty gross under there:



You need to take the manifold off in order to clear the lip on the firewall. Then when you install the intake, you may need to bump the lip in to get it to fit. Mine required very little modification.



It took about 5 minutes to get the thing in there all by myself. Use a load-leveler. I assumed that prior to shipping, they would have drained all of the fluids. It was a nice surprise when I started tilting the tranny back and fluid poured out all over the place. Oh well.

Once everything was generally where it needed to be, I used several jacks to line things up. Had one jack underneath the oil pan, another in front, and another on the side (while the tranny rested on 2x4s). I had things where they needed to be in only a few minutes and got one of the bolts in, but then the second motor mount was very stubborn:



The two brackets from the mount were a bit narrower than the corresponding point on the subframe. I stuck a pipe in there against the mounting tabs and banged with a 4lb hammer a few times and then it all fit.







It's at this point that you call the neighbors over and say, "see, it does fit."

Now, the Steeda handle I got mounted a bit oddly--at quite an extreme angle.



The stock shifter has a bit of a bend (I think I read somewhere that the T-56 sits at an angle in the Camaros/Firebirds or something) and then the Steeda handle bent in the same direction. That would make it very unwieldy. I took the shifter apart and reversed the little nub thing and then the angles canceled each other out quite nicely.



The Steeda handle (got it from Hinson) used the Mustang thread size for the knob (I forget the specs). I got a Mustang Cobra knob on ebay for cheap and it screwed on nicely:



Figuring out how to do the fuel setup took way more time than the clutch plumbing and then I still really didn't have a handle on it (though in the end it's quite simple). It doesn't help that there are millions of ways of doing it and everyone does it differently. Basically the gist of it is that you've got 5/16" lines coming out of and going into the tank. You've got some rubber lines and some steel lines. Lots of people reuse some or most of that and lots of people replace it all. Most people replace the pump but some people have used the stock unit. I didn't realize that you could use the stock pump (or I would have), so I got the Walbro 255. The Walbro had a 5/16" hose barb on it. I sawed off a bit of the tube in the tank and attached a compression fitting screwed to a hose barb. In retrospect, I could have probably just run the hose straight onto the 5/16" tube and clamped it.



Interestingly, a while after I did this, I pulled it out and found that the rubber had not responded well to the gasoline (it was sort of melting). I replaced it with a short length of braided line (but otherwise clamped on just the same).

Then I took out the stock fuel filter and used one of the rubber lines to bypass it. I did have to slightly bend the tubes so that the rubber hose wasn't kinked all crazy-style. From there you've got three lines: one out of the tank, one return, and one to the charcoal cannister.

I used the Corvette FPR/filter like almost everyone else. This thing uses fancy quick-disconnect fittings and you can get adapters (to AN fittings) if you want to be really cool. You can also user rubber hose and slide it onto the inlet and return if you're not trying to be cool. Then there's another quick-disconnect fitting on the fuel rail. To make things very simple, I connected the FPR/filter right up to the rail and bolted it to the frame that holds the igniters.



Not the greatest, perhaps, but elegant in its simplicity. I'll redo this at some point, though, probably with fancy-style braided lines like most people here seem to like.

Now, for cooling, I went with the JTR kit



This makes use of a C3 radiator and comes with everything you need except for the hoses (uses standard f-body hoses with slight modifications). The least attractive aspect of this option is that the radiator is slightly too wide and you have to bash the frame rails in slightly in order to make it fit (something like 1/8? on either side).





You also need to take out the coolant loop and the mounting tabs for the original radiator. The brackets that hold everything on will mount between the frame and the tow hooks.



I cut off the lower mounts off of the sway bar mount things and repainted





I noticed that the sway bar bushing was squeaking pretty badly (always heard it but didn't know what it was) so I threw some brake grease on it and now it's quiet.

Here's one of the brackets. I ground off some of it so that it will fit into the big aluminum piece that the radiator sits in



Here's the lower piece installed (this is before I modified the brackets, so the big aluminum piece is sitting on top of the brackets here).



Everything mounted:





They don't give you anything to mount the top end of the radiator, so for now it's only being held in by the mounts at the bottom. It stays in place just fine, but I'll have to make something to prevent the lower brackets from being subject to too much fatigue.

To modify the upper hose, simply cut out a section in the middle and add the coupler that is tapped for a bleeder valve and a hose barb. The barb goes to the surge tank.



The lower hose kinks pretty badly



so they give you a coupler and you're supposed to cut it in the middle, twist everything so that it's all at the right angle, and then clamp it together. I don't have a picture of that, but it ends up working nicely, if not beautifully.

The kit also comes with this restrictor thing that you place mount in the hose going to the heater core to reduce flow.



I ended up using the OEM fans. They're not a great fit, but I guess we'll see. They seem to be working great so far, but I don't have the AC going yet so it's hard to say.

I hooked the brake master up to vacuum using part of the f-body hose and part of the FD hose coupled together with an adapter I got at the hardware store. I forget the hose sizes?like 3/8? to 1/2? or something.



I should have checked out the brakes as soon as I got the car. They were in bad shape



Replaced them with rotors and pads from rotorpros. Probably shouldn't have used their pads, but I was feeling cheap. I wanted some cheap blank rotors (like the kind you could get at NAPA for $20), but I couldn't find any anywhere. Rotorpros doesn't do blanks, unfortunately.



I had a speed-bleeder left over from the remote clutch bleeder, so I hooked up some hose and just plopped it onto the regular bleeders and bumped until no air came out. Worked pretty well.

For the power steering, I went the DIY route with parts that were posted in a thread (I think by eat-pez). This is convenient since you can order everything and put it together yourself. It makes use of this adapter that goes from AN to the GM fitting on the pump:



So you make a hose with AN (6 I think it was) on one side and a banjo fitting on the other. I don't remember the thread size for the banjo fitting or the part numbers for any of this stuff (had it in the old thread :( ).





Now, the problem with this setup is that the hose is only rated for 1000psi. The GM pump normally puts out 1500. The mazda pump only did around 900 (these are just numbers I'm repeating from having read on the old forum?I haven't verified them or anything). So, you use the spring modification to bring down the pressure but a) you don't know what it is exactly and b) even if it's 900, that's not much of a margin of error. I seem to recall reading of these failing on people, so tread lightly. If you have a competent hose shop nearby, just take them the mazda and GM hoses and have them mate them together.

I'm sure the spring modification has been documented here by now, but it's pretty simple. Just undo the bolt where the outlet fitting goes and use a magnet to pull the spring and regulator thing out



Cut a few threads off (3) and put it back together.

When I was messing around with the engine before mounting it (I don't remember what I was doing), I broke the barb on the PS reservoir. Rather than paying out the ass for a new one, I did a sort of ghetto repair. I melted a piece of plastic over where the barb broke off and put a new barb at the top of the reservoir. Since the fluid would return above the level of fluid in the reservoir, it got aerated and made a ton of noise. I put another barb inside and ran a 2? rubber hose down into the reservoir and now it works fine.

BTW, to get the reservoir off, you have to take the pulley off. To get the pulley off, get the pulley removal kit from Harbor Freight. It's pretty cheap and there's no way you'll get it off without it (you can get it back on with a hammer, but that's not recommended).



For the intake, I used the JTR kit. This comes with one 4? (I think) to 3.5? 90 degree tube that slides onto the throttle body. Then you slide the MAF onto that and slide a 90 degree elbow (3.5 from here on out) onto that, pointing down. A largish aluminum tube goes onto that (like a foot long or so). You get two more 90 degree elbows and a two short aluminum sections to finish it off. You have to buy the filter separately (I don't remember the size I used).

The tube that goes down ended up a bit too close to the PS pulley



So I cut a bit of the fan's frame away



I ended up cutting one of the 90 degree elbows in half (so, 45 degrees) and that got the filter in a pretty good spot, right at the (former) oil cooler opening.



I took a piece of aluminum strip, wrapped it around that aluminum section of the intake and that bolts to the radiator bracket, holding the intake in place.

I decided to do the exhaust myself, which was an interesting undertaking. I had never welded anything before. I have an old-school stick-welder that I inherited from my grandfather which I used to weld everything together. The welds look like garbage with slag and splatter everywhere (well, I ground most of that off, but it still looks like shit). Oh well, it was cheap and works great. Sounds even better.

I ordered all of this stuff from Columbia River (except for those cats which I got on ebay for like, $30 each, and the v-band clamps which came with the headers)



I have JTR headers and the goal was to go 3? off the headers, through the tranny brace, back to the two cats, then into a y-pipe and out to the original cat-back. I'm not sure what cat-back I have, but I seem to recall the previous owner saying it's a Borla. It's a single muffler with two outlets and 2 3/4 tubing going in.

Here's the route



Getting the tubing to bend up and over the tranny brace was fun. I cut out small sections of the 180 bend and taped them together with electrical tape as I went along.

Here's the stuff mostly mocked up:



Then welded together (don't laugh!)





Okay, fine, you can laugh. Also, the headers are mild steel (along with everything else). I very much regret not getting them ceramic coated. I ended up painting them, but they've gotten pretty rusty on the bottom. The rest of the exhaust is now coated with a nice layer of rust, too, which I expected. This just has to last long enough for me to finish school, get a real job, and, I don't know, go turbo?

I went with Hinson's lower tie-rod-ends. These aren't put together the greatest, but they're supposed to wear out in 20k miles so I'll wait till then to get new ones (I'm not going to throw away $150). Interestingly, the hole for the clip on top was way above the castle nut



So I cut the top of it off and drilled a new hole



I had a bit of a surprise when I mounted my newfangled 17? wheels:



The tie-rod-ends do not clear the inside of the wheel. Now I know why these suckers weigh 25lbs :( Okay, so I cheaped out on the wheels (they're voxx). At least they look pretty damn good. They're 17x9 front and back (45mm offset, which is pretty damn good, actually). I've got 255s on the front and 275s in the back. You can tell the wheels in back are a bit on the narrow side for the tires if you look closely, but it's not very noticeable.

After getting an alignment and driving it around, I finally fully understood this damn bump-steer thing everyone's always talking about. The car felt like a death-trap, darting this way and that at will. Apparently, Grant's subframe lowers the rack much lower than Hinson's. I had become aware of this long before driving it, but I was somehow hoping that the bump-steer thing would only be something more skilled drivers would notice. Hah.

So I did some experiments. I Bolted the car down to the lift very very tightly (cargo straps, then tensioned the shit out of them by raising the car up a bit more using jacks). Then I used a jack to lift the suspension up. You're supposed to do stuff like this without the shocks in place, but I felt like cutting corners so I just strapped the car down tight enough to ensure it would *not* move. I made a little contraption that shot a laser down and then I marked where it hit at various suspension positions.



I did all the calculations and figured out how bad it was, but I don't remember what they were. It was close to 1 degree toe-in per inch of suspension travel, which is obviously very bad. I was able to verify that the rack did indeed need to move up (not much room for the tie-rod-ends to go down any further). Meanwhile, Mattster, who has a similar setup, got a bump-steer gauge and got some more solid data. At this point, I need to get some new mounts to raise the engine up a bit and then machine some spacers to raise the rack up.

Right now I've got it raised up about 3/8? on washers and aluminum strips and it is much improved, but still a big hindrance.



Now I have significantly more to add about electrical/electronics stuff and the interior. Really, those are the more interesting topics, but I'll get to them soon when I have some time.

Offline digitalsolo

Re: Matt's LS-1/T-56 FD Build
« Reply #1 on: September 10, 2010, 03:24:36 PM »
Thanks for reposting the build thread, I know it's a PITA.    I'm curious about the Miata wheel wiring.   :)
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.

Offline mdickw

Re: Matt's LS-1/T-56 FD Build
« Reply #2 on: September 10, 2010, 08:21:17 PM »
Before I forget, when I started up the car, it squeaked a lot. At first I thought the tensioner was worn out, but then I found it was the idler pulley (went away briefly when I sprayed it with WD-40). After tolerating it for a while, I read somewhere online that you can pull the bearings apart, clean them out, and regrease them



So I did that (using wheel bearing grease). And it still squeaked. So I bought a new pulley. And it still squeaked. So I bought a new tensioner. And it still squeaked. So then I bought a new belt?and it turned out to be the exact same belt (NAPA's more expensive option) that was already on it?but now it doesn't squeak anymore. I took the tensioner back and my regreased pulley probably would have worked, but I left the new one on. The internet consensus seems to be that you should use Goodyear Gatorback belts on these things, so keep that in mind.

Onto the electrical stuff. I'm a senior electrical engineering and computer science (double) major, and I had done a handful of alarms, remote start, etc-type stuff before I started this, so the electrical part is really my bread and butter.

We're extremely fortunate to have access to the FSM and the full factory wiring diagrams. I have found Grant's instructions to be very useful and I'm not really sure why people talk so much trash about them around here?though I haven't really followed them very explicitly. They're a great way to start and provide a lot of good ideas. I haven't made a single connection, though, without checking the FSM and (when applicable) GM PCM pinouts and Camaro wiring diagrams. With all of these resources readily available, you'd be stupid not to.

First thing I did was relocate the battery. I used some big fat welding wire (either 2 or 4 AWG, I don't recall) to run from the passenger-side bin area out to the starter. This ensures the shortest possible path from the battery to the starter, where you'll have the most current. Keep in mind that the electrical resistance of wiring is directly related to the length of the run, so if you keep the runs down, you don't necessarily need gigantic cables (good luck finding anything as big as 4 AWG in a car from the factory). I used a nice big lug which I soldered generously, then covered in dielectric grease and taped up pretty well.



From there I believe I used the wire that went to the old starter and used that to get power from the junction at the starter up to the main fuse box. This works great, but the line from the battery to the starter is currently not fused. I need to throw a nice big fuse on that thing in case the starter goes nuts or the wire's insulation gets compromised.



I used one of those fancy, expensive Braille racing batteries for a bit



It didn't have any trouble starting the car, but I destroyed it by (many times) letting it drain to empty for months on end and  forgetting to take it off the charger. Then I just used the original FD battery for a while (despite it sitting without use for a year and a half). Now I have a Miata battery from autozone in there. Smallish with a fantastic warranty. Miata batteries aren't totally sealed, but they vent to little hoses that you can easily route outside the car. For now, I just make sure to keep the sunroof open all the time.

I don't have the main fuse box secured except for a zip tie that's holding the big fat wires going into it in place. I should come up with something better, but it's fine.



I took the two grounding cables coming off the engine and bolted them into an unused hole on the frame rail.



Note: make sure you have your engine grounded in at least two points, because if the single ground point fails, it will ground through your PCM and miscellaneous other things that won't be able to handle the current. I added these ground lines (green) later for this purpose.



In order to power  up the PCM and starter, I tapped into the wires behind the dash, immediately after the ignition switch (constant, ignition-on, and starter). This isn't the greatest, though, because these wires are only fused with the 40A bus fuses, so if there's a problem with the PCM or ignition or something, it's going to have a draw more power than the designers probably intended in order to blow the fuse. In retrospect, I should have just tapped into the wires in the passenger foot area per Grant's instructions.



Be careful when wiring the starter, since you can tap into it in many places: after the switch, after the clutch interlock switch, and after the starter relay. If you want to use the interlock switch and/or starter relay, make sure you tap into the wire afterward.

Since I brought up the interlock switch, I'll go ahead and remind everyone that the FD has two ?clutch switches? and they seem to be a constant source of confusion. One is the interlock switch, which is a high-current switch that is closed when the pedal is switched all the way down. The other is a low-current neutral switch which closes when the clutch is pressed even slightly. The latter has nothing to do with starting, but is used for cruise control and to alter the tune of the engine when in neutral. The f-bodys used a similar switch and the PCM expects to see it, but the switch works backwards from the FD switch (e.g. one of them closes when the clutch is pressed and the other one opens). If you want to connect the FD clutch switch to the PCM, you'll have to invert the signal somehow (a relay would be the simplest way).

My clutch switch was actually broken, but I found it was very easy to take apart and I replaced the original spring (broken into three pieces) with one out of a pen. Works like a charm (note the plunger is actually backwards in this pic).



In order to wire up the fuel pump, I took the white/red wire that normally goes to it and cut it in the driver's foot area, then I pulled it back and there was plenty of slack to get over to the passenger-side bin area where the battery, a small fuse box, and some relays are mounted. Here's the wire coming out of the area behind the driver's seat (white/red).



I used a trick which I had seen in another build thread to get the wiring through the firewall. There's a toilet gasket that fits reasonably well as a grommet.



I used some black silicon to help ensure it stays in place (it doesn't look this brown in person)



Most of the rest of the miscellaneous stuff was done pretty much the standard ways, similarly to Grant's instructions. All connections are soldered, of course.

Since the FD only comes with one power door lock actuator, I had to install another one. The crazies out there will get an actuator out of a RHD FD since installing a traditional actuator in the FD may seem difficult on first inspection, but I saw it done on the Club so I got a cheapo actuator (these things are like, $10 online) and made it work. I mounted it at an angle, behind the lock lever. One side is screwed into a rivet-nut. The other side would have been, but it turns out the regulator motor is behind where I wanted to put it (and drilled a hole). I had to cut down the coupler piece that attaches the actuator rod to the cable.



For wiring, I used the wires that had been used to power the Bose amp. I then spliced them into the wires coming out of the door lock controller that go to the passenger side. This was very nice, since you'd have to move the lock only slightly and it would flip the rest of the way on its own, but for the life of me I could not figure out a way to get control of the door lock controller (to override the position). The wiring diagram makes it look pretty trivial (ground one wire to lock, ground the other to unlock) but that won't do it. If you scope it out, you'll see sine waves on some of the wires and some pulses on others. This is because it uses capacitors to control the timing so that it can flip from one state to the other.

Anyway, after I spent hours trying to make it work the way I want, I just took it out and used the relays built into the alarm I got. Not what I wanted, but I'll deal with it later.

For the alarm, I went with a K9 Sombra. This thing is very cheap ($30 new on ebay) and does everything I want. It uses an RFID-based remote that has one button. You can either use the single button on the remote to arm/disarm, or you can enable ?RFID mode? where it will disarm based on proximity. This actually works out pretty well. I don't much care for the beepy-honky part of alarms, so I didn't bother to hook up the siren, horn, or the shock sensor. Pretty much just the locks and the starter kill. I was skeptical due to the price, but I couldn't be happier with it (though I wish the remotes didn't say ?K-9? in giant letters, but I may just paint over that).

My NC used an RFID keyless system that I liked very much, so I'm trying to go in that direction. One of the features was a button under the trunk lid that you could press if you've got the key thing close enough. I decided to implement something similar on the FD. There are trunk release solenoids you can get, but I decided I try to see if I could do it with a door lock actuator ($10!).

I took the lock mechanism out and drilled a hole in the little arm thing



Next I bent some pieces of the bracket stuff it comes with:



And finally:



This has no problem popping open the trunk. Now I just need to wire it up. The alarm has a built in starter kill relay that I'm not using, so I'm going to use that to provide power to a button only when the alarm is disarmed. I need to order a waterproof button. There aren't any really great locations to mount it, though. Either I'll go the easy route and mount it down by the license plate, or I'm thinking about mounting one in the center of the Mazda logo. That would be pimp, but I'd have to take the whole damn bumper off.

Pretty much all of my Bose stuff had had it. The subwoofer thing just made whistle noises and whined, while the driver's speaker was completely blown. I decided to stick new speakers in the doors and a sub in the back.

For the doors, I got some 5 1/4? component speakers. I wanted to mount the tweeters up higher, but I couldn't really find a good spot and I wasn't sure I could do it in a way that would look good. I used the stock speaker boxes, then I cut away some of the plastic to make a bigger 'opening,' then I surrounded it all with spray-foam.



Trim it up..



And you've got yourself an enclosure. I mounted the tweeters nearby in the box. You can see what the driver's side one looks like in one of the pics above.

For the sub, I went with the ?DamonB? design. Some guy on the club (DamonB) came up with a design that mounts between the shock towers. He posted some crude blueprints which I used as a starting point, along with my own measurements. Here's what I came up with (pretty much the same thing)



He said to use a 10? subwoofer, which I did (a 10? Jensen unit I got for $20 on partsexpress). I calculated it out and don't remember the exact number, but this box has an internal volume around .7 cubic feet. This is way, way smaller than just about any 10? sub calls for. This is much better-suited to an 8? sub. Unsurprisingly, when I hooked it all up in the garage, the sub sounded like a sub with a box that's too small?it sounded like shit. BUT, when I got this thing in the car and shut the hatch, it ended up sounding very, very good (well, or at least, much, much better).



Whoops, that pic is out of focus. Also, that fancy carpet? About half of a rug I got at Walmart for $20, trimmed to fit (that gap at the back is because that panel is out until I install the hatch button).

I got a 4 channel amp from walmart (online only) during a christmas special.



I'm quite happy with how I wired it up. I cut the connector off the Bose subwoofer thing and then I pulled the wires back, crimped and soldered new spades on, and put the connector back together and used the other side to wire up the amp. This is especially nice because the factory speaker wiring is shielded (you don't see that every day). I then spliced the lines that used to go to the Bose subwoofer thing into the lines that go to the speakers and voila.

I don't have a head-unit, instead I'm doing the whole carputer thing. That space to the right of the amp is for the carputer. I haven't installed it since I haven't finished the LCD faceplate thing (what a pain in the ass that has been).

I was interested in using the Bose center speaker if it wasn't blown (it didn't seem to be), but I figured it would probably not be worth the effort to mess with. Turns out, the Bose amps use high-level (so, speaker-level) inputs and as soon as I wired up the door speaker lines, the center channel amp got a signal and started driving the center speaker. But, it was way, way louder than the door speakers and that's no good. I threw a stereo 10k audio-tapered pot in before the center channel amp and was able to reduce the volume to a suitable level. This pot isn't ideal for this application (only a tiny portion of it is in the right range, so adjustments are tricky), but it works.



I couldn't be happier with how the sound system turned out, actually.

And since I took a pic of it, here's what I did with the shifter hole:



It's a miata (NA/NB) shifter boot bolted to a piece of sheet steel with is screwed/bolted to the tunnel with some thick gasket material in between. I had the shifter boot laying around and I didn't think I'd need it any time soon (though the shifter boot in my NA has since torn apart and it would have been nice to have had a replacement on hand).

Now for the interesting part. I'm using a microcontroller to deal with some of the interesting electronics issues. It outputs the VATS pulse when the alarm is disabled. People online will say that it should either be 30hz or 50hz, and they'll say either ground pulses or 5v pulses. I've found that only 50hz works (30hz, I believe, is for Corvettes). I'm using 5v pulse with 50% duty cycle and it works fine. Crappy thing is, though, that the car will often start up without it. This is pretty damn annoying, as it prevents me from using that as an extra security measure (reliably, at least).

I am also reading the VSS and tach frequencies to get speed and RPM. The plan is to generate the signal to run the speedo, but I haven't gotten to that yet. For now, at least I have the speed and can do whatever I want with it (currently I have it set up so if I hit a button on the wheel, my laptop will read me the speed in computer-voice over the speakers). The GM VSS does indeed output at 8kppm. If you want to get the actual speed, you have to multiply it by the ratio of the stock f-body tire diameter to whatever you're running and by the ratio of the f-body diff to the FD diff. I did the math and the speed came out to match GPS exactly, which was quite fun.

I am also reading the water temp and oil pressure (using the GM sensors) and driving the Mazda gauges with PWM. To reiterate, yes, I'm using the GM sensors to drive the Mazda gauges.

Here's the microcontroller board, mounted securely in the cabin:



So, here is how the gauges work. The PCM gets a voltage signal from the temp sensor (look on the pinout, you'll see the line). I monitored the voltage while looking at the temp over OBD-II and derived the following equation:

f(v) = -54.3v + 288 where v is the voltage

That correlates 99.6% with the forty or so data points I noted from the experiment, so this will pretty accurately provide at least the same reading as the PCM.

Oil pressure wasn't as easy, since the PCM doesn't read it. The LS-1 oil pressure sensor is a simple resistance-to-ground sensor, so it only has one wire that comes out and the sensor grounds it out with a certain amount of resistance (this is also how the FD sensors worked, too, but with different curves, I'm sure). GM calls this thing an ?oil pressure switch,? leading many to believe that it's on-off or something, but no, it provides discrete levels of resistance for different pressures.

I tested it out by attaching it to a big PVC tube, mounting a pressure gauge on the tube, and then pressuring the tube with my air compressor and noting the readings. I did not get stable numbers when I used the multimeter to measure the resistance, but when I hooked the sensor up using a voltage divider (5v with 100ohm resistor), I got stable numbers. So, in a 100ohm, 5v voltage divider, you can use this equation:

f(v) = 1.91 ? 4.47^v

It's exponential, so the readings become less accurate as pressure rises (my data correlated 99% through 80psi, btw). The resistance, in case you're interested, calculates out to about 6ohms at 0psi, 60 at 30psi, 85 at 60psi, and so on.

Many people have said that the Mazda gauges are garbage, aren't accurate, aren't linear, etc. I found them to be 100% linear. As in, no shit, I wrote down 10 positions, did linear regression and the equation I got had 100% correlation. The gauges work, of course, by driving them to ground through some resistance. The water pressure gauge works over a much larger resistance range. Fascinatingly, the gauge seems to have been designed to work with a sensor that runs at 1ohm per degree fahrenheit. The oil pressure gauge uses much lower resistance values, possibly intended on 1ohm per PSI. I didn't get too finicky about figuring the resistance out since I'm using PWM to drive them. Basically, the microcontroller calculates where the gauge needs to be an generates pulses that drive the gauges to ground some fraction of a period of time. I'm driving them at 100hz, which is pretty low for PWM, but it works fine. I set the water gauge up so the low point is 150 degrees and the white line is about 230 or so (when the fans come on) and the red line is 250. It's weird driving a car where the water temp gauge isn't buffered and actually move around?and it moves around a LOT.

Someone wanted to know about the steering wheel? The buttons on the steering wheel work like pretty much all steering wheel buttons these days (the same, even, as the stock FD buttons). Pressing a button drives a resistance to ground. So, you provide the wheel with some voltage and when you press a button, the voltage will go down to a specific level depending on which button was pressed. The NC wheel uses separate lines for the left-side buttons (stereo) and the right-side buttons (cruise), but I wired them up with a resistor on one of the lines to make sure all of the buttons have discrete values. The NC wheel is also illuminated. The FD had two wires going to the cruise control buttons, so I used one for illumination and the other for the buttons. The buttons are then wired up in a voltage divider to the microcontroller, so it constantly sends the voltage level over serial and I can do what I want with it on the carputer (or my laptop at the moment). If you want to wire one of these up yourself, you can get modules online that will do all of this for you.

By the way, here's a comparison of the FD airbag to the NC airbag



The exterior is in pretty good shape, though I'd like a new front bumper. A paint job definitely won't happen for a few more years. It's not lowered or anything and that's not even in the cards, especially with the subframe as low as it is.

The interior isn't finished yet. I reupholstered the seats with leather seat covers straight from China. Took me about 8 hours to do the driver's side and a bit less for the passenger side. The leather is quite a bit lighter than the original stuff. I used VHT vinyl 'dye' to fancy up the tan plastic parts and it comes out very close to the shade of the seats. I've been refinishing the black plastic stuff with 'gunmetal' paint and that seems to be working out pretty well. Here's a teaser shot:



Now we're mostly up to date. I'll add more as I think of it and as progress continues.

Offline lackskill

Re: Matt's LS-1/T-56 FD Build
« Reply #3 on: September 11, 2010, 03:48:37 AM »
There are a few shallow mount 10" subwoofers that require less than .7cuft such as this Pioneer:  http://www.sonicelectronix.com/item_18211_Pioneer+TS-SW2501S4.html (.35 min, .7 max)

I have 2 of those in my truck, and while they're not overpowering by any means, I had serious space restrictions.  My enclosure is actually too small for 2 of those, but it only sounds like it when pushed really hard.  I achieved the more balanced sound I was looking for anyway.  There were some other shallow mount 10" speakers that would probably be a better fit for you like this fosgate: http://www.sonicelectronix.com/item_9155_Rockford+Fosgate+Punch+P3SD410+Shallow-mount.html or this Kicker: http://www.sonicelectronix.com/item_21923_Kicker+CVT102+-10CVT102-.html

As I understand it, the airspace requirement is a function of the excursion of the driver.  So the more airspace a driver requires the more excursion, and therefore more energy, the speaker can deliver.  Or at least in theory...  The size and shape of the box also plays directly into frequency response and harmonics.  Some speakers will perform better than others in a space deprived box, but wouldn't be the right person to tell you which speakers do or don't.

Sorry if I'm not telling you anything you don't already know.  I have a tendency to ramble when I'm tired.

Offline ScottMan

Re: Matt's LS-1/T-56 FD Build
« Reply #4 on: October 12, 2010, 05:04:48 PM »
This is the first time I've seen this build and I'm blown away by some of the creativity and innovation!  Awesome job!

Quick question about the microcontroller from a complete electronic n00b.  Do you need to run the computer in the car constantly for the microcontroller to function or do you simply calibrate or configure the controller then you can remove the computer and retain the functions?  I ask because I would love to have calibrated, quality stock gauges like you have, but I don't want to carry a laptop or carputer all the time.
1993 RX7 - GTO LS1

Offline BrownBoy

Re: Matt's LS-1/T-56 FD Build
« Reply #5 on: October 12, 2010, 06:31:49 PM »
Great detail in your posts, keep up the good work! :)

GregD

  • Guest
Re: Matt's LS-1/T-56 FD Build
« Reply #6 on: October 13, 2010, 11:09:14 AM »
This is the first time I've seen this build and I'm blown away by some of the creativity and innovation!  Awesome job!

Same here, thanks for re-posting this.   :cheers:

Offline Shavel

Re: Matt's LS-1/T-56 FD Build
« Reply #7 on: October 13, 2010, 08:27:30 PM »
This is the first time I've seen this build and I'm blown away by some of the creativity and innovation!  Awesome job!

Same here, thanks for re-posting this.   :cheers:

Again, first time I've read this.  The electrical part I'll have to read again (a few times).  Thanks!
Ant

Offline mdickw

Re: Matt's LS-1/T-56 FD Build
« Reply #8 on: October 21, 2010, 11:46:14 AM »
Do you need to run the computer in the car constantly for the microcontroller to function or do you simply calibrate or configure the controller then you can remove the computer and retain the functions?

No, it's a completely self-contained system. You need a computer to calibrate it (and program it in the first place), but otherwise, it just runs 24x7 like any of the other systems in the car. You can also use a computer to log the data, which is fairly useful.

As I have more time, I plan on making the system more robust and easy(ish) to replicate so that others can implement it. Hopefully during the holiday break between semesters I'll have enough time to finish it up and publish all the details (code, schematics, etc) and even put together kits and assembled packages. You can pretty much do anything you can think of with a system like this--it's mostly just a matter of programming and sometimes additional hardware.