GSpeed's 2015 BRZ- SCCA Super Touring Build
 

We bought this car from Hiley Subaru in Ft. Worth, TX with the express purpose of learning as much about the car as we can, in order to make parts to help people get (and stay) on track.

This little guy showed up on the 4th of June with all of 52 miles on the odometer.

http://gspeed.com/wp-content/gallery...zbuild0001.jpg

http://gspeed.com/wp-content/gallery...zbuild0002.jpg

The very first thing we did was get it on the scales. We've got to know what we're working with here. The following numbers were taken with 7/8ths of a tank of gas, and 175lbs of lead shot in the driver's seat.

Total Weight- 2917 lbs
Front Weight- 54.5%
LF/RR Cross- 50.1%

LF: 820 RF: 769
LR: 688 LR: 640

http://gspeed.com/wp-content/gallery...zbuild0003.jpg

http://gspeed.com/wp-content/gallery...zbuild0004.jpg

Clearly a very well balanced car. It'd be nice to get a little more weight over the rear, but looking at that crossweight, I can see why they put the battery where they did.

The second thing we did was swap out all the fluids with Motul stuff. We have no intentions of being nice to this car, so we want to make sure the fluids in the car are up to task. Why Motul? We believe in it so we sell it, not much more to it than that. It's not cheap, but compared to both the problems failing fluids can cause and the overall costs of racing, it's a drop in the bucket. Cheap insurance, if you will.

http://gspeed.com/wp-content/gallery...zbuild0005.jpg

http://gspeed.com/wp-content/gallery...zbuild0006.jpg

The next thing on the list was a Cool Shirt tank in the trunk. Conveniently, it's within 2 pounds of the spare tire we removed, so the balance of the car won't be adversely affected. Texas track days can get really hot in the summer, and we don't want to be running the AC.

http://gspeed.com/wp-content/gallery...zbuild0007.jpg

http://gspeed.com/wp-content/gallery...zbuild0008.jpg

http://gspeed.com/wp-content/gallery...zbuild0009.jpg

http://gspeed.com/wp-content/gallery...zbuild0010.jpg

Our goal for this project is to be as scientific as our budget allows, so the first significant modification to the car was an Aim MXL2 dash and datalogger. I've got a detailed thread here, but here's a picture of the finished product:

http://gspeed.com/wp-content/gallery...zbuild0011.jpg

Now we're moving on to the engine sensors we want on the car at the moment. We bought a TRD sandwich adapter from FT86 Speed Factory (Thanks guys!), and installed an AEM temperature sensor in one of the ports. It's even got a DTM connector built in. Isn't that cool? All NPT connections were sealed with Loctite 567.

http://gspeed.com/wp-content/gallery...zbuild0012.jpg

http://gspeed.com/wp-content/gallery...zbuild0013.jpg

http://gspeed.com/wp-content/gallery...zbuild0014.jpg

Since one of the problems we hope to address is fuel starvation, we need a good way to quantify the problem. We designed a little sensor block to hold a pressure transducer and mount to the shock tower brace. This pre-production one isn't anodized, but future ones will be hard anodized black.

http://gspeed.com/wp-content/gallery...zbuild0015.jpg

http://gspeed.com/wp-content/gallery...zbuild0016.jpg

http://gspeed.com/wp-content/gallery...zbuild0017.jpg

Next on the list are shock pots, tire temp sensors, and lap times! Hopefully next weekend we'll be able to get it on track and start getting some baseline lap times to use for comparison as we add modifications.

Jake

Subbed

Looks super cool, interested to see where this goes.

Thanks guys.

I decided to start off the morning by installing our little oil pressure sensor. We bought an adapter from FastWRX.com to install in the little galley plug on top of the block.

http://gspeed.com/wp-content/gallery...zbuild0018.jpg

http://gspeed.com/wp-content/gallery...zbuild0019.jpg

It's a little close to the throttle body connector, but not so close that the wires are binding. I'd prefer a little more clearance, but it'll do. If you go this route, be sure and clock the tab on the connector in a direction you can reach it, it'll make detaching the harness MUCH easier.

We need to capture video of all our shenanigans (tests), so we bought and installed a SmartyCam HD. It'll talk to the dash, so it'll start and stop recording automatically, and it's got a nice data overlay.

I didn't have a mount to install it, but I had some bar stock and a Bridgeport, and that's basically the next best thing.

http://gspeed.com/wp-content/gallery...zbuild0020.jpg

Looking forward to seeing the development on this car.

In for results. Loving all of the AIM products.

First track weekend!

We put the car on track for the first time last weekend. Our goal for this weekend was to get ideal cold tire pressures, and establish some baseline lap times for a bone stock car.

The car is completely unmodified, save for the data system, camera, and cool shirt. As of the first session, the car had 62 miles on it. All laps were done on the 1.7 CCW.

Thursday's first session was with VSC and traction control off with the buttons on the console (no pedal dance). We ran 13 laps, and the best lap time for this session was 1:34.9.

Thursday's second session was an hour later. Tire pressures were adjusted slightly, and we did the pedal dance to completely remove the electronic nannies. Immediately, the car was "WAY MORE FUN!!! :)" based on driver notes, and the best lap time from that session was 1:32.9. So we've confirmed what we already knew, the pedal dance is well worth the hassle until a more permanent solution is worked out. It got us two entire seconds.

Thursday's third session had a best lap of 1:32.7. We adjusted the tire pressures a little bit more, and settled on hot pressures of 34 psi front, 33 psi rear. Driver notes indicate the car could get into the 1:31 range with a little more seat time.

http://gspeed.com/wp-content/gallery...zbuild0021.jpg

(Please excuse the cell phone pics, we'll get some better ones next test day.)

On Friday, we got some more time on track. By 9:30A, the track temp was already 97F. At the end of the 5th session for the weekend, the OEM pads were done. They hit metal on metal, so we called it a day. Best lap time up to that point was a 1:33.3. Damage to the rotors was superficial, so we turned off 0.012" off of each rotor and got some NAPA "Pro-Former" replacement pads from across the street.

Saturday morning was a little bit cooler, with track temps around 84F. It was fairly humid, and the track felt noticeably slicker than days prior. The best lap from the day was a 1:33.0, though. We learned exactly how worthless the NAPA pads are from a track standpoint. Within about 5 laps, they had faded horribly. The diff was smelling a little toasty, too, so we called it a day after 12 laps and changed the fluid just to be safe. It didn't seem to hurt anything, but the fluid did smell pretty cooked.

I decided to see if I could recognize the brake fade with the data system.

Here's a GPS trace showing longitudinal acceleration. Blue is negative, so the blue spots on the track are the braking zones.

http://gspeed.com/wp-content/gallery...zbuild0022.jpg

I took the data from the Aim system and exported it to Excel. I wrote a formula to sequentially number each braking event, and then average the pedal pressure and deceleration during each event. Messed with the units and signs a little bit, then I was able to get a brake gain, in units of Gs per ksi of line pressure.

Here's a graph of the brake gain per braking event. Each lap is a different color. You can see after the 5th lap, the braking force generated by the pads is falling off rapidly. It's always nice to see data matching driver notes.

http://gspeed.com/wp-content/gallery...zbuild0023.jpg

It would be really nice to be able to factor in pad temperature as well. I'll have to add some brake temp sensors down the road when we get heavier into brake development. The pads were probably falling off even earlier, but without temperature data, it's hard to say that for sure.

I also need to brush up on my Python, too, since processing the data in Excel took waaaaaay too long to use as a regular analysis tool.

Here's a quick shot of the owner next to the car:

http://gspeed.com/wp-content/gallery...zbuild0024.jpg

So, lessons learned:

-Stock car can run 1:32 flat.
-OEM brakes are good, but last about 5 sessions (2 hours).
-NAPA brakes are no good.
-Fastest cold tire pressures are 29 psi front, 31 psi rear for these tires on this track in this weather.

Agenda for next test:

-Wider, stickier tires
-Tire temperature monitoring
-Fuel pressure monitoring
-Oil temp & pressure monitoring
-Better pads

We've got some good video from the Smartycam, but I'll have to go over that later.

Jake

subbed...

I can't wait to read your tire temperature monitoring ? do you plan on installing some IR sensor ;) ?

Yep! We've got some little guys going on this week.

OMG I am so looking forward to this .

this is one of the install that would push me to get an AIM MXG...if I ever see someone documenting it.

Great seeing the car on Friday and all the work you guys are doing with it. Looking forward to watching this progress and hopefully getting out on the track with Dave sometime to play.

Nice build!

I think you guys would become quite popular if you could sort out a solid solution for E85 fuel starve.

I like the oil pressure sensor, definitely going to look into that.

Keep up the good work.

Thanks for the kind words. Fuel starvation is definitely on our radar. I don't want to start posting about it until I have something worth talking about, though.

http://gspeed.com/wp-content/gallery...zbuild0025.jpg

We had another test day today. Since we cooked the NAPA pads last time, we installed some PFC Z-rated pads on the front. We were hoping to run some full race compounds, but they (PFC) are shut down doing annual inventory right now, so we couldn't get any in time. We had these on the shelves, though, so it was a great opportunity to test them out.

We got on track about 9:30 while it was still cool out, and snapped off a 1:32.4, our new fastest lap for the stock configuration. Predictive was showing a high 31 was possible, but we just couldn't get a clean lap in (lots of traffic). We ran 11 laps for the first session, then after looking at the tire temps, decided to add some camber bolts. We added about 0.8 degrees of negative camber, which really changed the balance of the car. It went from pushing through slow corners to rotating with a little throttle. It clearly could use some more camber in the rear to match what we've added to the front. We actually slowed down a bit, but that's likely due to the increased temperatures.

This was our first run with the engine sensors, wheel pots, and tire temps (rear only right now). It'll be a lot of fun to analyze all the data we collected this morning. I'm looking forward to getting some actual pitch and roll information before we start messing with springs and sways. I'll put up some stuff in the Aim system's thread about the installation.

We've logged 167 miles of track time in the past two weeks. Tomorrow morning, we're going to run a session or two while it's nice and cool on the RE71Rs and see how much time that gets us.

Summary
Configuration: Stock suspension and powertrain, PFC Z-rated front pads, OEM rear pads
Best Lap: 1:32.4

can't wait to see the temperature monitoring setup !

Subbed. Awesome!

Since you have re71rs on stock wheels please run them back to back with 245s on 9" wheels (making the assumption that you're heading in that direction.) Except maybe on the longest tracks I'm convinced that 245s are net faster.

While that would be an interesting test, we don't have stock size RE71Rs. I'm sorry if the picture was misleading. The picture is of the stock tires, and the RE71Rs were installed (on RPF1s) after that picture was taken in preparation for tomorrow's test.

I would enjoy getting that information, but I don't think we could justify buying a set of tires for one or two tests. We bought 245s since we're planning on running them for a while as we test other components. Maybe once we're through the RE71Rs we'll test some other competitive tires like Rivals, but at that point, honestly, it'd probably be time for slicks.

Fair enough. Carry on! :burnrubber:

http://gspeed.com/wp-content/gallery...zbuild0026.jpg

Our test day today went really well. We bolted up a set of 245/40/17 RE71Rs on some 17x9+45 RPF1s, and immediately dropped 4.8 seconds! Quite a substantial improvement, which really shows how important tires are to a car.

http://gspeed.com/wp-content/gallery...zbuild0027.jpg

Another member was out with his modified BRZ, so we had a great opportunity
to directly compare to another car.

http://gspeed.com/wp-content/gallery...zbuild0028.jpg

With his massive Stoptech brakes, he was able to outbrake us pretty handily and passed us in short order.

http://gspeed.com/wp-content/gallery...zbuild0029.jpg

We were clearly losing time to him in the twisty section as well. On the stock, soft suspension weight transfer just takes so long, so we've got some coilovers on order to get the car lower and speed things up a bit.

We ended the second session about 11:00. Track surface temps were nearing 100F, and we were roasting the RE71Rs. The car was being pushed hard, and the tire temps showed it. Taking measurements in the pits, we saw the shoulder of the right front was over 200F. The right front tire is always abused on counter clockwise tracks like this.

For our first session, we intentionally started the session with a slightly low tank of fuel (5/8ths full). We wanted to provoke some starvation and see how it looked on our fuel pressure sensor. The results were pretty interesting.

Here's a plot of about 15 minutes of track time. You can see pressure is sitting pretty right around 58psi or so, but when it starves, it dramatically plummets. There are six major events there, each one lasting for anywhere between 5 and 15 seconds. Every one of them occurred after an extended corner.

http://gspeed.com/wp-content/gallery...zbuild0030.jpg

What's interesting is that the driver did not always notice, and the car certainly wasn't stumbling for the entire time pressure dropped. So this tells us the problem is much more prevalent and serious than it would initially appear. I would love to see similar plots from other users with modified cars. We're not really supposed to sell parts in these build threads, so I'll make a separate thread, but we just got our first batch of fuel pressure sensor adapters in. The more data we can get the better we'll be able to identify a solution.

Oil temperatures were also pretty high for the ambient temperature, so we'll start working on an oil cooler setup. Anyone have a wishlist, or anything specific they want in a cooler setup? Speak up now!

Summary
Configuration: Stock suspension, camber bolts, 245/40/17 RE71Rs
Best Lap: 1:27.6
Notes: Fuel pressure is ugly, oil temps are high. Stickier, wider rubber is good.

PM me for the fuel pressure sensor ;) I like what you are doing :)

Awesome stuff. Much respect. Keep it up!

Great to meet you today. Stop by tomorrow to check out the JRSC

As of today, I've finally got all the sensors working on the BRZ. The analog inputs on the Aim dash used Binder 719 series connectors, which are meant for 24ga wire. I didn't realize that at first, and ordered a spool of 20ga 3-conductor cable out of habit. I wish I'd paid closer attention, because that made soldering much more difficult than it had to be:

http://gspeed.com/wp-content/gallery...zbuild0031.jpg

Little tiny solder cups, over and over and over. My eyes hurt at the end of each day. I did not want to cut the Aim harness, though, just to better future-proof the setup, so it was worth the hassle. The other end of the cables have 3-pin DTM connectors, which are fairly common. I didn't get any pictures of those.

http://gspeed.com/wp-content/gallery...zbuild0032.jpg

The wheel pots were a challenge to figure out. Here's what I finally settled on for the fronts. I'm not a HUGE fan of the resolution on the front, but it should be sufficient. The MXL2's ADCs (analog to digital converter) are 12-bit, which means over the full range of the potentiometer, we've got 4096 (2^12) discrete values. For a 4" pot, that's near-as-makes-no-difference 0.001" resolution. I measured my install ratio as 3.822:1, so the front pot will be able to pick up a wheel displacement of 0.004". Good enough for a sports car.

http://gspeed.com/wp-content/gallery...zbuild0033.jpg

Here's a close up of the lower mounts I machined for the fronts. I didn't want to drill into a control arm, so this guy just clamps on to the little hole in the control arm with a big turned "washer" on the bottom.

The AEM pressure transducers had crappy connections supplied. If you're going to use those, I'd highly recommend replacing the connectors with Delphi PN 13602480 (Mouser link here.) Much better quality, and it matches the connectors under the hood a bit better, too.

http://gspeed.com/wp-content/gallery...zbuild0034.jpg

The rear shock pots were somewhat easier to install (since the wheels hopefully keep pointing the same direction the whole time). The right rear was mounted to the filler neck bracket that had a hole in the exact right location, with a little aluminum spacer to keep it at a good angle.

http://gspeed.com/wp-content/gallery...zbuild0035.jpg

The left rear needed a little aluminum bracket added out of some scrap sheet and rivnuts, but it wasn't too much trouble either.

http://gspeed.com/wp-content/gallery...zbuild0036.jpg

The lower ends of the shock pots were mounted to the lower shock bolt with an adapter turned out of some 3/4" hex stock. I like this method since it prevents anyone from accidentally using the pot as a droop limiter. :D

Front and center in that picture above are the rear tire temperature sensors. They're little 1/4" x 1" IR units that work with any thermocouple reader. Really neat little sensors. The lower half of that aluminum arm is held on with nylon bolts, so in the event of an impact, it'll tear away cleanly rather than get all caught up in the tire and cause problems. The wires lead up to some conventional thermocouple connectors that would be able to pull free as well.

http://gspeed.com/wp-content/gallery...zbuild0037.jpg

http://gspeed.com/wp-content/gallery...zbuild0038.jpg

The TC hub for the rear was mounted in the spare tire well with rivnuts, and some TC wire was used along with Binder 712 connectors to make some extensions to outside the trunk.

http://gspeed.com/wp-content/gallery...zbuild0039.jpg

Since the TC hub was so far from the dash and CAN expander, I had to make an extension. Once again, 20ga was a mistake. 24ga would have been much easier to work with for this application.

http://gspeed.com/wp-content/gallery...zbuild0040.jpg

Just like with the wheel pots, the front tire temp sensors were no fun. I finally settled on mounting them to the chassis rather than the upright. The data logs show no more than 90° of steering input at any point on a clean lap, so the sensors should have a good view of the tire at all times.

http://gspeed.com/wp-content/gallery...zbuild0041.jpg

Here's a view from the top to give you a better feel for where they are. The fender liner was trimmed and replaced, but you can still see through to the tire from the engine bay. Once we're done testing tire temps, we'll probably replace the fender liner with an intact part.

http://gspeed.com/wp-content/gallery...zbuild0042.jpg

The Aim thermocouple hub was mounted to the right side of the engine bay, so we didn't need to make an extension like in the rear. The fourth wire there is a reference IR sensor mounted in the belly pan by the right wheel to take track temperature measurements real time, so as to make it easier to compare tests from different days.


http://gspeed.com/wp-content/gallery...zbuild0043.jpg

Thermocouple wire was run across the bay and along the strut brace. You can see the track reference sensor splitting off from the three tire temp sensors.

http://gspeed.com/wp-content/gallery...zbuild0044.jpg

Here's how the engine bay looks currently. Thermocouple wire and instrument cable everywhere! We wouldn't have it any other way. It's hard to make informed decisions about performance modifications without the right data. :D

Once everything was hooked up and double checked, it was time to calibrate the wheel position sensors. Here's how we did it.

Tools required:

-Bore Gauges
-Calipers
-Floor Jack
-2/4/6 precision gauge block

http://gspeed.com/wp-content/gallery...zbuild0045.jpg

A particular position was marked on the rocker as near to the wheel well as we could get. The distance from that point to the floor was measured and recorded, as was the shock pot position. The floor jack was used to very slightly raise the car, and a new set of values was recorded. This was repeated 20 times or so for the front and back of the car.

http://gspeed.com/wp-content/gallery...zbuild0046.jpg

The values were put into Excel, where a near perfect linear fit was found (R² values of 0.9998 and 0.9997). For the front, the wheel moves 3.82" for every inch of shock pot movement, and the rear wheel moves 1.27" for every inch of shock pot movement. Given the rear installation location, this is probably very, very close to the rear shock motion ratio.

http://gspeed.com/wp-content/gallery...zbuild0047.jpg

Now that we've got wheel pots on the car, we can do really fun stuff like calculate chassis roll and pitch. Here's a plot showing lateral acceleration (purple) vs. chassis roll (green). We've got a strong inverse relationship like you'd expect, and you can actually see the roll damping working. (Green downward slopes are smoother than purple upward slopes).

This information was also put into excel where we saw an average roll stiffness of 2.5°/G. I say "average" because the progressive spring rates are apparent in the data. Low lateral loads give a higher amount of roll per G. Once we get the new coilovers on the car, I'll put the springs on the spring tester and plot out the actual spring rate curves.

I hope everyone is enjoying this post as much as I'm enjoying doing the work! :D

I enjoyed it. The pics could be a little bit bigger though as they don't quite fill up two 24" monitors.

21:9 monitors are out. Fill one of those!

So skimming over the hopefully obvious sarcasm... Are the pictures resizing properly? I had hoped if I left them big, people could blow them up if they wanted detail, but if it's making it harder to read, I can resize them.

Photos are perfect on my end. Great info!

Pictures are good.
I personally think it would be fun to come watch one of these sessions.

That'd be great! We usually test on Thursday and Friday mornings, though, rarely if ever on weekends. If you want to swing by, just let me know.

There are a lot of them, so sometimes when I load this page not all of them resize. 3110x2073 is a damn large photo though. :lol:

I like the "after driving" notes, great thread!

[ame="https://www.youtube.com/watch?v=3pxHWPTUJnM"]PST BRZ- July 10th Fastest Lap - YouTube[/ame]

Finally got a video trimmed and uploaded! Here's our current fastest lap.

Very nice lap.


Our lines are very similar except for a couple of places. I take the same approach to rattlesnake but come back a little more center track to turn into the right slightly sooner to allow me to be farther right on the track when I turn into the left hander and can then apex it a little later and carry more speed down into wagon wheel. After Bryan's insistence, I stopped tracking all the way out on Buzzard's neck to be in a better position to enter Horseshoe from mid track, which allows me to carry a couple miles an hour more through there. I also short shift from 3rd to 4th between boothill and tombstone so I am not fighting to get the shift in on track out (same thing on Ricochet, short shift right before turn in). I shift in the same spots on the rattlesnake/wagonwheel section.

I'm really enjoying this thread. :burnrubber:

Between the video and your notes. I have my homework :) Great tips. 1:36 is my best time. I'll take it for my first time ever on the track.

I would be cautious taking my notes over Dave's (the driver in the video) as he is one hell of a driver and has a lot more seat time on that track than I do (although I think a lot of that is in a very sweet viper). I am interested to see if the slight differences are a personal preference, if there is a good reason for doing them (which I could benefit from) or if he is still figuring everything out as the car is new to him.

thank you again for sharing all this information with us.
It is very valuable to see how and where the IR temp and shock system can be mounted.
Looking forward to see what these sensor are saying on the track !

Jonesy/Bramick, here's Dave's response:

"The comments Jonsey made are spot on. I have tried both of his suggested changes previously as they are the conventional wisdom of some. I found no improvement on the predictive lap time readout when doing so. Still, I am not convinced there is not something there. His comment on track positioning entering Rattlesnake is correct and very important.

On Rattlesnake: The fast guy with the most laps around MSR enters Rattlesnake very tight. Most instructors advocate a wide entry. Either way, Jonsey is correct, the place to focus on is being track-right before the left-hand entry to Rattlesnake. FWIW, my money is on the tight line - even though I try the wide entry occasionally.

On Buzzard's Neck/Wagon Wheel: I am convinced I am giving time up here but have not made the wider entry position work for me. I will keep at it though because I think this is my weakest point on the track. I will get Bryan [PST's Instructor] to ride with me and advise me. It is easy to get into a habit that is not the best thing to do (for example, tracking all the way out on Buzzard's Neck or driving cars around a race track for all, or part, of your adult life)."