MASTER THREAD: Comprehensive Road-Course Modification Guide — Optimizing the 3 for the track

[Mash]

Would a vented hood let more air in and the front tub / front under panel removed (I am already assuming you are doing this)
View attachment 608296

Also when are you planning on going back out?

Due to the debacle happened after news media claimed that my Tesla is 2 seconds from Taycan, which it's not, I have to take low profile here and wait for the spring. But I will be keep improving it on Autobahn, superchargers.

 

[Mash]

Without ducting the air that passes through the box directly out of the hood I don't think it would make any difference at all. That would be impossible using the stock radiator/condenser box as the air is currently being directed out of the bottom of the car (so opposite direction).

Letting air easier going up to the windscreen should be helping. I don't have data proof of that though. But I believe that removing frunk tray helps.

 

[MasterC17]

Letting air easier going up to eimdscre should be helping. I don't have data proof of that though. But I believe that removing frunk tray helps.

I agree removing the frunk helps because the air now has a much larger volume of space to evacuate from (including out of the space around the hood by the windshield). I just don't think that adding a vented hood would make a measurable difference unless you ducted the air out of the hood. I could be wrong!

 

[Mash]

I agree removing the frunk helps because the air now has a much larger volume of space to evacuate from (including out of the space around the hood by the windshield). I just don't think that adding a vented hood would make a measurable difference unless you ducted the air out of the hood. I could be wrong!

I wouldn't mess with car aero without any need. Its easy to remove tray for the track and I doubt that any hood holes add anything useful. Btw, it's possible that tray removal decreases airflow around front drive unit...

 

[Davidss2]

I was able to fit 18x11s on my car. This is with mpp coilovers, borg motorsports front camber bushings, diy slotted rear upper control arms, and a very light fender roll.

The wheels are konig hypergrams 18x11 +40 with a 21mm spacer front and anywhere between a 10-20mm spacer rear. I am running 275/18 federal tires.

 

[MasterC17]

I was able to fit 18x11s on my car. This is with mpp coilovers, borg motorsports front camber bushings, diy slotted rear upper control arms, and a very light fender roll.

The wheels are konig hypergrams 18x11 +40 with a 21mm spacer front and anywhere between a 10-20mm spacer rear. I am running 275/18 federal tires.

View attachment 608768

Looks good, and good to know - though I am curious why such wide wheels on a comparably narrow tire? I'm looking to move to a 10.5" wheel with 295's.

 

[Mash]

Looks good, and good to know - though I am curious why such wide wheels on a comparably narrow tire? I'm looking to move to a 10.5" wheel with 295's.

I also don't get it. 275 fits very nice with ET30 wheel on 9.5 wide wheels.

295 on 10.5 with et25 requires 10mm spacer (and therefore something to make wheel hub centric) in front and more front camber.

11 for 275 sounds like too much trouble for what?

 

[MasterC17]

I also don't get it. 275 fits very nice with ET30 wheel on 9.5 wide wheels.

295 on 10.5 with et25 requires 10mm spacer (and therefore something to make wheel hub centric) in front and more front camber.

11 for 275 sounds like too much trouble for what?

I am going to see if APEX will make 18x10.5 ET25 wheels. How many degrees of camber are you running up front?

 

[Mash]

I am going to see if APEX will make 18x10.5 ET25 wheels. How many degrees of camber are you running up front?

I need to double-check that mine are et25. I was running 3.5, but I think it was too much - inner edge was much hotter after run on both sides. I think 2.5 should also fit.

 

[Lunares]

How much camber are people getting with MPP front arms? With a 1" drop from stock P3D my guys were only able to get -2.7 degrees

I am running 18x9.5 ET34, currently a 265 width. Will want to try to move a 275 but alignment shop wasnt sure it would fit. Maybe eventually a new wheel to go to 10.5 and wider if possible

 

[Mash]

How much camber are people getting with MPP front arms? With a 1" drop from stock P3D my guys were only able to get -2.7 degrees

I am running 18x9.5 ET34, currently a 265 width. Will want to try to move a 275 but alignment shop wasnt sure it would fit. Maybe eventually a new wheel to go to 10.5 and wider if possible

275 fits with normal street 1.5 camber for me on et30. Unless your tire is in reality much wider than 275 from Michelin

 

[Dangerous Fish]

How much camber are people getting with MPP front arms? With a 1" drop from stock P3D my guys were only able to get -2.7 degrees

I am running 18x9.5 ET34, currently a 265 width. Will want to try to move a 275 but alignment shop wasnt sure it would fit. Maybe eventually a new wheel to go to 10.5 and wider if possible

You'll struggle to get more camber than that with MPP arms. I'm switching to Redwood arms which can go to over 4 deg., not that you'd need it. Personally, I'm aiming for -3 on the front.

 

[TwoK4drSi]

How much camber are people getting with MPP front arms? With a 1" drop from stock P3D my guys were only able to get -2.7 degrees

I am running 18x9.5 ET34, currently a 265 width. Will want to try to move a 275 but alignment shop wasnt sure it would fit. Maybe eventually a new wheel to go to 10.5 and wider if possible

You can get up to -5.8 on UPs FUCA. Their system can “flip” back and forth between street and track camber. Looks like they are able to achieve this by adjusting the FUCA length itself to make it physically shorter. Can also adjust caster by doing this as well.

 

[subieworx]

The reason we've elected to go with an oil cooler rather than larger heat exchangers is due to the reason that unlike an internal combustion engine, there are two systems on the Model 3 with slightly different operating temperatures. The battery really needs inlet coolant temperature in the 25-30C range, and the Drive Unit can accept inlet temps in the 35-50C range.

Tesla already has a method of keeping the two systems separate - when the AC Compressor is running and the system is in Parallel mode, the AC Compressor can focus on removing heat out of the battery.

The oil is the hottest fluid, so using an oil cooler with fans allows a direct interface with the hottest fluid to the air. It is also the simplest to install. Any heat that is removed from the oil cooler is heat that is not going into the heat exchanger - so that means cooler loop temperatures and reduced temperature into the battery when the system goes into series mode.

The fan can also be independently controlled, so to Mash's point if you are concerned with heating the battery quickly for supercharging, not running the fan will result in simply a little bit more thermal mass in the oil system, but not a significant delay in heating the system (this is something I know for sure as we tested supercharging and there was no issue for the battery loop to get up to temperature for supercharging).

So all in all, pulling out a ton of heat with oil coolers seems like the logical way to go.

Beyond that certainly, a more hardcore approach would be a larger AC condenser, chiller, and radiator.

However since the system is so complex, and Tesla has multiple different ways of doing the control systems, it's harder to make a perfect back to back comparison. What I can say is that when we disconnected the AC Compressor from our RWD car, without the oil cooler the drive unit overheated in 2 laps. With the oil cooler it didn't overheat for over 15 minutes.

You can put your hand behind the fan and feel the amount of heat coming out of that thing when the Oil Temperature is 70 degrees C!

Here's an overlay from some road testing we did on our LR RWD. The testing consisted of full-throttle power pulses from 60-110km/h. The stator temperature is down 8degC, the oil temp is down 2degC - but the big difference is that the powertrain and battery loops are down 2.4 and 1.3degC respectively - because the oil cooler is working in parallel with the existing heat exchanger. You can also see the significantly different slope of the rear oil temperature.

The colored traces are without the oil cooler, and the white traces are with the oil cooler.

View attachment 608304

Was this done on the dyno?

 

[subieworx]

You can get up to -5.8 on UPs FUCA. Their system can “flip” back and forth between street and track camber. Looks like they are able to achieve this by adjusting the FUCA length itself to make it physically shorter. Can also adjust caster by doing this as well. View attachment 608931

Is this actually possible though? Most of the time changing camber heavily changes toe. I haven't modeled the suspension on the 3 yet, but given my experience aligning street/race cars over the last 20 year I can't imagine it would work differently.

 

[Dangerous Fish]

Is this actually possible though? Most of the time changing camber heavily changes toe. I haven't modeled the suspension on the 3 yet, but given my experience aligning street/race cars over the last 20 year I can't imagine it would work differently.

It will change the toe, yes. Some will count the turns and mark their tie rods so they can re-adjust this without a full alignment, but small errors multiply and eventually a re-alignment will be needed. I don't see a great need to re-adjust every time I go on track, so these quick-change arms are a bit pointless for me. The Redwood arms are similar to the the UP ones in that you can adjust castor too, so I'll dial in 3 degrees on them and leave it.

 

[subieworx]

You could adjust caster with the UP arms as well.

Yeah, I never recommend user adjustable alignments as they always turn out bad.

 

[Adrian Cockcroft]

I've been working some more on my open source telemetry analyzer, made some good progress on another update adrianco/rs-tesla-telemetry
For now, it needs to be run locally after downloading the free RStudio app and connecting that to GitHub, and loads a local file before it displays in the browser.

Multiple plots are working, with time as the x-axis. I also clean up the lap list and sort it so that the fastest complete lap is at the top and is pre-selected.

Stuff still to do that I have ideas for: pick and zoom on plot, distance rather than time on the x-axis, metadata about the event, turn by turn analysis, saving best laps for a specific circuit.

Uploading files to a central service could be done, but the telemetry files are fairly big, security and making it work for multiple users is an issue. I'm thinking that saving the best reference laps into a curated leader-board for each circuit might be a better approach, then have the cloud based version be a read-only browser for people to try out the tool and look at a few interesting laps for different car setups and drivers.

 

[fastp3d]

You can get up to -5.8 on UPs FUCA. Their system can “flip” back and forth between street and track camber. Looks like they are able to achieve this by adjusting the FUCA length itself to make it physically shorter. Can also adjust caster by doing this as well. View attachment 608931

+1 on the UP arms! I run them and have great results with -3.5. Also running UP 18x10.5 on 315's

 

[Dangerous Fish]

I've been working some more on my open source telemetry analyzer, made some good progress on another update adrianco/rs-tesla-telemetry
For now, it needs to be run locally after downloading the free RStudio app and connecting that to GitHub, and loads a local file before it displays in the browser.

Multiple plots are working, with time as the x-axis. I also clean up the lap list and sort it so that the fastest complete lap is at the top and is pre-selected.

Stuff still to do that I have ideas for: pick and zoom on plot, distance rather than time on the x-axis, metadata about the event, turn by turn analysis, saving best laps for a specific circuit.

Uploading files to a central service could be done, but the telemetry files are fairly big, security and making it work for multiple users is an issue. I'm thinking that saving the best reference laps into a curated leader-board for each circuit might be a better approach, then have the cloud based version be a read-only browser for people to try out the tool and look at a few interesting laps for different car setups and drivers.
View attachment 609017View attachment 609019

What Adrian's done here is really useful for anyone serious about analysing their track driving and who doesn't run a dedicated data logger.
Take a look if you haven't already and see what you think.