flatlander937
flatlander937 HalfDork
11/20/22 12:31 a.m.

It's been about a year with this car so it's time I document it here for GRM. This will be a bit more of an abbreviated "build" thread... most likely there will be more info in the golfmk7.com forum in the members build section (just search for derhase HPDE shenanigans and you'll find it).

Anywho... My last car was a 2011 Mazda2 named "Tot". It met a tire wall at Mid Ohio August of 2021. Build thread on GRM for that HERE. I mulled over what to replace it with for 3 months... partly because I wasn't sure if I was even going to be able to drive on track again. I had a pretty bad concussion and was out of work for a week, and my balance and focus was off for a solid two months. Right side of my head felt heavy for a long time. I could tell I wasn't as "sharp" as I was before just driving in normal daily traffic. When that started to improve we decided to start looking for something. I have some VW friends who've been trying to get me into a MK7 for a long time... and I drove one at some point and it was just really damn good. Then the exact car I wanted popped up for sale locally... a 2019 Rabbit Edition in Pure White. I was kind of indifferent to DSG vs 6MT since IMO the factory shifter and clutch feeling is pretty terrible overall. This one happened to be DSG. The wife liked it too.

Note: for the MK7.5 facelift cars, the Rabbit Edition is the only way to get the LP (lighting package LED adaptive headlights), PP (performance pack which has bigger brakes and a electro-hydrualic LSD, referred to as the VAQ differential), AND no sun roof. Avoid the sunroof at all costs on these cars. They are notorious for leaks. For pre-facelift cars the only way to get that combo is a 2017 Sport model FYI. In 2018 there was no PP available on the S trim. All years SE and Autobahn come with sunroof standard. IIRC the Sport trim in 17 was a one year only thing. 

Also the Rabbit Edition was limited to only 3000 cars. 500 white, 500 black, 1000 urano grey, and 1000 cornflower blue. All were supposedly 50/50 as far as DSG vs 6MT goes. So I'm basically on my way to ruining a relatively uncommon car. cheeky

Der Hase is German for "the rabbit" or "the hare" FYI.

 

The thought behind buying this car was that even IF I got on track and found out I wasn't up for it, it was still just a friggin nice car to drive around and fun to drive in anger. Not to mention silly power can be made if you want to for relatively little effort.

 

Spoiler alert: I've been to the track 7 days so far this year, with another 2 coming up in December for the TSCC December to VIRmember HPDE on December 10-11.

Anyway before it's first event I replaced EVERY fluid in the car except for the coolant. I also got a VCDS and did a few of the common track focused adaptations as well (disabled hydraulic brake assistance, changed XDS from normal to weak, increased lockup of the VAQ differential, and lowered the brake boost slightly for better pedal feel).

I did 2 days at VIR in February with the car bone stock aside from Ferodo DS2500 pads, some new Zimmerman blank rotors, and Motul 600 brake fluid. Still had some all season Michelins on it. The car did great, and I managed to run a 2:26 at VIR in 50F weather and being very rusty myself. About 10+ sec faster than the Mazda2 was the year before.

The event went great. I wasn't super happy with the pads. They felt inconsistent from one corner to the next. Temps were in the 50s so I figured maybe they were cooling off a bit too much over the long straights? Pedal would be soft on some applications. I hate not being able to trust the brakes every time. The factory all seasons suck but still had fun.

Avatar photo VIR head on.JPG

....

flatlander937
flatlander937 HalfDork
11/20/22 12:49 a.m.

...

Had a blast and went back to the track early in May with PCA FSR. I got some 245/40R17 RT660s and 17x9 Apex SM10s first...

 

 

 

This is why I love hatchbacks for track duty:

 

The only other thing I had added prior was the Audi RS3 brake ducts... which were not enough to help out the DS2500 pads (despite all of the internet saying they are great for the track). They're not BAD by any means... their falloff is slow and gradual and would be perfectly adequate for a novice student... but I like to be able to stand on them repeatedly.

 

On the plus side there were plenty of MK7s to play with there! (the one has a Golf R front bumper and is STUPID fast - 2:10s at VIR and actually this past weekend just did a 2:09).

22-04-30 VIR 3 white GTIs spiderman meme.jpg

 

So the problem has been established it needs better brakes and it needs some camber up front. I did get down to a 2:22.97 though, so still a new PB.

...

flatlander937
flatlander937 HalfDork
11/20/22 1:09 a.m.

...

So on went some 034 Motorsport camber mounts. Nice pieces. Honestly the stock shocks/springs feel pretty decent overall. I don't want to drop the car at all and frankly based off of some stuff I've seen I honestly thing most aftermarket coilovers for this platform are complete garbage. Off the shelf rates are completely wrong for basically every single one of them. Lowering springs as well.

 

I also put some Ferodo DS3.12 front pads on (and kept DS2500s for the rear). Note for the MK7.5 GTI *rear* they do not sell anything more than the DS3000 rear compound. I didn't think there'd be enough difference to warranty changing them too. These pads are $$$.

 

I also wanted to start logging IATs (and also oil and coolant temps). I know they're not great but I wanted to know HOW not-great they were.

 

So I squeezed in one more VIR trip not even a month after the last one...

 

 

 

 

So one of the things I was trying to accomplish (in the background and not having to think about it) was data. Temperatures for IAT vs AAT, coolant temp, and oil temps.

What a lot of people seem to say is that coolant temp is not able to be kept under control, and fixing that sometimes fixes the oil temps... on a STOCK tune my data at least says otherwise. Note that this was an incredibly slow session because of a ton of traffic. First session of Saturday... so it is mostly only useful for IAT monitoring.

Refresh rates are intentionally slow for the temperature channels because really they don't need to be super fast... the engine RPM was more important to log as a higher refresh rate. Speed and G Forces come from my AIM data.

 

 

So the front pads were amazing. Rotors up front are starting to show minor heat checking so I'm guessing 2-4 more days left on them (already have a pair new in box waiting). The only issue is they grip harder and harder with heat as the session goes on... and the rear starts to drop off and it becomes VERY easy to get into just the front ABS. It beats losing the pedal entirely... but could be better with more aggressive rear pads. The DS3.12 pads never felt like a "race" pad daily driving them a bit beforehand... I can't speak for COLD temperatures just yet (been 60F+ mostly), but on track the first couple applications you could tell they didn't bite as hard as you'd expect (but still better than the DS2500s when kind of warm). They get better after a lap or so and seemed to stay pretty consistent. I think my problem is rears getting hot and front tires greasing up.
 

Oil temps at 270F are not hard to hit on a hot day at VIR. Definitely needs an oil cooler at some point, especially if I tune it. Temps are the main reason I'm not in a HUGE rush to do so. The IATs were kinda silly though as you can see in the videos above.

 

I also got down to a new PB again at 2:20.93 (long video, it is lap 4). I also managed to take the upper esses a lot faster than I was probably ready for... nearly pooped a bit hitting at 114mph (tapering to 109mph through T9) before hitting more traffic. It was REALLY bad that day.

 

flatlander937
flatlander937 HalfDork
11/20/22 1:26 a.m.

Then the car got ceramic coated by Gilmore's Detailing in Newport News (highly recommend!)

 

Ordered H&R 26mm sway bars front and rear after running a bunch of numbers through some calculators. I haven't 100% modeled the suspension just yet, but here are the bar rates I've measured so far if anyone is interested.

Note H&R normally sells a 28mm front paired with 26mm rear, or a 26mm front paired with 24mm rear.

Tirerack had some on sale so I got a 26mm front and 26mm rear. I suspect for actual performance use that would be my best bet to experiment with. 

I also bought a DO88 intercooler. It's one of the largest stock location ones available. I did not want a FMIC to leave room for a (future) oil cooler.

 

 

 

Better lighting in the form of (24) 4ft long LED lights + quickjacks have been fantastic in the new house as well. Makes working on stuff so much less of a pain.

 

 

You can see the 26mm rear sway bar installed. Also did an Apex wheel stud kit. Wasn't impressed with their packaging and had to clean up threads on basically every single wheel stud but they do the job.

 

 

And also got a group buy with a couple other guys together to get a more aggressive DS1.11 rear compount made for the GTI rear calipers (with mechanical e-brake... vs the Golf R that has an elecronic one with different pad shape). It was a $450 set of pads but well worth it. Got to test them out (and the new intercooler) at Fastivus at Summit Point on the Jefferson Circuit. LOTS of fun... and in the intermediate group actually never had to give a point by all day. 

 

 

 

 

Braking was a LOT more confident and consistent. Also was able to string a bunch of quick laps one after the other as opposed to stuff varying so much lap to lap due to IATs/heat soak/etc.

flatlander937
flatlander937 HalfDork
11/20/22 1:35 a.m.

Fastivus was a LOT of fun... finally got my dad out to the track and he rode along with me in my second and third (final) sessions of the day. He had a blast and wants to come see what VIR is all about now too (probably next year as I'm planning to be there another 7 days in 2023).

 

 

 

 

 

 

 

 

And we're basically caught up to today.  The only other thing that'll be new on the car before the December 10-11 event is the APR DSG tune (already done after Fastivus). I don't think it's necessarily going to do a lot for laptimes, but should keep clutch slippage under control when things get hot since clamping force is increased.

 

I'm still on the stock ECU tune (which is speed limited to 124mph). Won't be messing with it before VIR, but I am working on an open source tune file for it. I plan on making a very conservative tune targeting maybe 275ish whp. Not looking to give myself a ton of overheating problems on track.

docwyte
docwyte PowerDork
11/20/22 10:22 a.m.

I was super happy with the Bilstein PSS16 DCC coilovers on my Golf R, paired with the big rear Neuspeed swaybar.  As you discovered, DS2500's aren't track pads.  I did the IE stock placement intercooler, seemed fine but I never tracked my car

flatlander937
flatlander937 HalfDork
12/19/22 8:22 p.m.

Had a great weekend with the Tidewater Sports Car Club at the December to VIRmember HPDE!

Was great to see a bunch of other MK7 buddies, some new ones from Facebookland and some old friends.

The intercooler: awesome. I never got logging from a full session, but I got a few laps in. I think I have a problem with my dongle or need to update firmware to work with the latest version of Simostools. Not a big deal but in a nutshell IATs were never more than 12-14F over ambient when on throttle. They'd creep up a bit when off throttle of course but go down and stay down to +12-14F. Oil temps never exceeded 245F. Coolant stayed fine at 190-200F.

The sway bar feels great out back. I do think I will go forward with installing the front bar this winter since I already have it. For science. If I hate it in February then the stock front will go back on before a May event.

The Garmin Catalyst: friggin awesome. VERY useful tool for comparing lines and different braking techniques. I still have a lot of work to do myself as a driver as far as getting back to consistently doing the same thing every time. Keeping myself in place in the seat seems to be the biggest issue. Arms and shoulders are so sore  I was able to work on individual sections of the track throughout the weekend and it pieces together an "optimal lap" video where you can figure out which ways were fastest.

The main thing is it doesn't tell you what the CAR is capable of... only what YOU have already done. As long as that is understood, and you can feel comfortable pushing yourself and know WHERE you can do it safely, you can figure out what works and what doesn't pretty quickly. I think the value in this thing will be huge in speeding up learning a new track. Try a few different lines and it'll slowly adjust the recommendations to piece together what is a decent goal.

On the APR DSG tune: prior to the event I tested out the manual mode and found out it did NOT auto-upshift as I instructed to... unless you lift off the throttle. Even then it didn't work every single time. I took the car back and had it flashed to run manual with no auto-upshift as I'd rather it be consistent. MAN was it hard to break the habit of letting it upshift itself on the stock tune... At some point I'll have to make a montage of me yelling inside the car "DAMNIT SHIFT" etc as it hits the rev limiter, I remember to hit the paddle, and then it takes FOREVER to complete the shift. Finally on my 4th session on Saturday I started to get a rhythm down... Basically only used 3rd-4th. Stock ECU tune was speed limited to about 121-122mph GPS speed (due to the shorter 245/40R17s).

 

New personal best is down to a 2:17.31:

 

We had a bunch of MK7 owners there including two wagons, an R, and the rest GTIs:

 

 

In other news it took a LOT of getting used to manually upshifting due to the way I was driving the stock DSG tune up until now (letting it auto upshift, only worrying about downshifting as I entered every turn)... It was REALLY hard to unlearn. I spend the first 2.5 sessions figuring out just WTF to do since the car is so darn quiet. This video is from session #2 Saturday only.

This is NSFW - Don't watch with kids around... it's nothing but cursing every time I hit the limiter followed by a delayed shift.

https://www.youtube.com/watch?v=ecEVFWoWWSk

 

And then some photos from the weekend:

 

 

 

 

 

flatlander937
flatlander937 HalfDork
1/14/23 9:52 a.m.

Been busy with the holidays and stuff but here are some updates:

 

Temp data from VIR:

https://datazap.me/u/derhase/stock-tune-temp-data-track-vir?log=0&data=0-2-4-5-7-8-9-13-15

https://datazap.me/u/derhase/stock-tune-temp-data-track-vir?log=1&data=0-2-4-5-7-8-9-13-15

 

Pertinent info I got from it:

-DO88 IC heat soaks to about 12F above ambient on the stock tune in a 20min session. Perfectly acceptable. Honestly due to the IAT location, it wouldn't surprise me if it's actually lower than reported because of the sensor/intake getting heat soaked on track. What will be important is checking the delta again once tuned.

-DSG temp is fine and increases slowly throughout the whole session. I'd probably feel comfortable running it to 220F or so.

-Coolant temps not a problem.

-Oil temps don't seem to be an issue. We'll see what happens once tuned. Depending on how quickly it approaches 270F once tuned... I MAY see if it's possible to get away with just doing some massive hood vents. I 100% believe that this chassis' primary cooling problem is airflow. More through the radiator and everything, the better it'll be.

-I was running the heat on speed 2 or 3. Nothing crazy but there was a little extra help there.

-On one of my first days at VIR on stock IC, I was able to get to 270F a couple times even in similar 50ish F temps. I wasn't able to do that at all this time. Perhaps the lower IATs help the engine more efficiently get rid of heat while also making more power? Or perhaps the stock IC was heat soaking, so any airflow through it was getting pre-heated before the radiator (moreso than typical driving conditions). I don't think this is right because a more efficient (for IATs) IC is removing more heat from the charge air... meaning that heat has to go somewhere. Not sure what to think of this. But it feels good and temps are not an issue yet.

 

 

The car is now tuned:

-I finally took the plunge with the OS Tuning using Simos Tools. I purchased a tune from 8 Bit Performance (who is on the OS Discord and helps a lot of others out frequently). Currently on V5 (V1 was flashed exactly one week ago today), with V6 waiting to be flashed and logged. Nearing the end just refining timing mostly. He is tuning me VERY conservatively as IMO this car doesn't need a ton of power to be fun... just needs to not feel like a tractor with zero top-end like it does stock. We will see what the difference is come February. I have a stock tune (just unlocked to enable 3E HSL) to flash back to maybe do some back to back testing (though I may wait until May to do this... I'll have 8hrs of open track time at a private event which will be better for testing different things. At some point I plan to take it to a dyno for some actual verifiable numbers.

-I did actually start to play with a tune myself (not flashed to the car though) before realizing I just don't have enough time to dedicate to learning to do it all myself at this time. May fiddle with some in the future. For science.

-My FAVORITE part is the 3E HSL (high speed logger). It is a patch applied to the tune which allows full access to powertrain bus data messages. So now I can log:

  - Individual wheel speeds

  - Brake pressure

  - SAS

  - Plus a ton of other stuff engine and trans related.

 

Here's a couple quick videos going over them real quick. Ignore my stupid voice.

 

 

 

 

 

 

 

I still haven't dropped the front sub frame. Procrastination may happen until after February. HOWEVER I have a decent reason: I ordered some Superpro poly LCA forward bushings, plus some Audi RS3 rearward front LCA bushings. I really only want to drop the subframe once, so depending on how long it takes for everything to show up, I may delay putting the FSB on until after February.

 

My schedule this year for track stuff:

Feb 11-12: VIR Full HPDE with Tidewater Sports Car Club (Oak Tree Bowl III) - Registration here

Mar 4: Champcar VIR South 12hr race with Team PistnBroke in an SVT Focus - should be fun. Just signed up today.

May VIR Full: Private track day Friday + 2 day event, hasn't been announced yet.

July-Oct: Not sure yet since a lot of schedules aren't out... but wanting to hit Road Atlanta, Summit Point Main + Shenandoah again.

+Fastivus 2023 whenever that gets announced. It may take care of checking Shenandoah off the list above.

December VIR Full: Also not listed yet.

 

Basically the plan is to try and do an event every other month roughly, with other stuff going on permitting.

flatlander937
flatlander937 HalfDork
1/14/23 9:58 a.m.

Final tune revision:

https://datazap.me/u/derhase/8bp-v7-data-logs?log=0&data=6-10-15-26-33-49-54-55

 

I also recently went down the rabbit hole of trying to understand the ECU oil temperature calculations since nobody really knows exactly what is going on. With the OS 3E HSL patch applied, you can log basically anything within the ECU as long as you have the address and know the formula for it.

 

There are 5 main temperature calculations:

toil and toil_raw - "toil" is what gets logged on a Cobb I believe. It's the only thing available without the high speed logging patch on mode 22/unlocking the ECU for the 3E patch.

 

toil_srv_intl, toil_srv_intl_raw, and toil_srv_intl_can all follow each other exactly with the cluster under normal conditions. In fact if everything is working just fine, I don't recall ever seeing any of those three changing in relation to each other (aside from one blip during going from key on engine off to engine starting).

 

You can also log the oil sensor signal directly. It is a Hella sensor that sends 3 pulses every few seconds. First is for temperature, second is for oil level, third is for diagnostic purposes. All the documentation I can find shows the temp pulse should be 23-82ish ms long. But what's really weird is while monitoring the 5 temperature PIDs above, plus the sensor signal, it slowly creeps up to about 25.5ms pulse time, then drops to zero at about 130F. Then continues trending upwards. I have graphed the values vs all the oil values... it is directly linear with toil_srv_intl, toil_srv_intl_raw, and toil_srv_intl_can (the instrument cluster). Based on my estimates, the sensor might potentially "reset" to zero again at around 280F... but I hope to never have to find that out. I did go back and forth between C and F a few times. The base calculations and formulas are all in C so I found for diagnostic purposes that made it easier to recognize if I got a formula wrong and to prevent rounding errors.

 

You can see the temp values below:

 

Normal drive, warmup.

https://datazap.me/u/derhase/oil-temp-data-logs?log=0&data=2-3-4-6-7-8-9-10-11-12

 

More digging through available PIDs found a reference value available which stayed pretty consistent between 22-23ms. Also take a look at the toil vs the toil_srv_intl lines... as toil_srv_intl is trending upwards towards the toil line, as soon as the sensor pulse time drops to zero, it begins trending away. There are references in the ECU documents that indicate there are times the temperature is modeled and times that it is being reported in real time. I suspect that once the sensor pulse time changes it is being used for real time measurement without any (or as much) background estimations going on.

https://datazap.me/u/derhase/oil-temperature-data-logs-cold-start-trip?log=0&data=1-2-4-7-9-10-11-12-13-14-15-16

 

Datalog from a long drive after already warmed up:

https://datazap.me/u/derhase/oil-temperature-data-logs-warm-trip?log=0&data=1-2-4-7-9-10-11-12-13-14-15-16

 

Then I decided I'd try unplugging the sensor while already warmed up to see what happens about here.. it just continues along using the last known good value for everything (notice no changes in ANY of the values):

https://datazap.me/u/derhase/oil-temperature-data-logs-unplugging-and-plugging-sensor?log=0&data=1-2-4-7-9-10-11-12-13-14-15-16&mark=446

 

I high idled the car for a little bit (can see the tail end of that at the beginning of this log) to try and change temperatures a tiny bit. Then plugged the sensor back in and you can see the jump in all the temperature readings. There was a brief period where there are 3 different readings among all the sensors before they re-stabilize like usually with the toil/toil_raw following one trend, and the other 3 "srv_intl" PIDs following their own:

https://datazap.me/u/derhase/oil-temperature-data-logs-unplugging-and-plugging-sensor?log=1&data=1-2-4-7-9-10-11-12-13-14-15-16&mark=1991

 

I let the car cool for several hours and decided I'd try unplugging the sensor on a "cold" start (about 100F coolant temp) to see what happens...

 

https://datazap.me/u/derhase/oil-temperature-data-logs-cold-start-unplugged?log=0&data=1-2-4-7-9-10-11-12-13-14-15-16

 

toil and toil_raw were completely modeled for sure because without the oil sensor data they are now trending up as the car heats up.

toil_srv_intl and toil_srv_intl_raw remained at the last known good temperature (90C).

toil_srv_intl_can was pegged at -40C and never moved.

t_temp_sig_loil (sensor direct signal) was zero.

 

Once plugged back in, everything goes back to normal:

https://datazap.me/u/derhase/oil-temperature-data-logs-cold-start-unplugged?log=1&data=1-2-4-7-8-9-10-11-12-13-14-15-16&mark=230

 

So not a TON of answers aside from it appears that the cluster is as close to accurate at least at "normal" running temperatures. Trying to find out more about the sensor itself to find out why the signal drops to zero.

 

Signal vs temperature:

1673319139051.png

 

 

So yeah... I'm doing pretty good procrastinating dropping my subframe before the February event. I got my Superpro forward LCA bushings, but waiting on my RS3 LCA vertical bushings to arrive. Still have not shipped but have my fingers crossed.

jfryjfry
jfryjfry SuperDork
1/14/23 10:06 a.m.

What's the plan to remove/install the bushings?

flatlander937
flatlander937 HalfDork
1/14/23 10:29 a.m.

In reply to jfryjfry :

I'm dropping the subframe to install the H&R 26mm front sway bar, and because I have a DSG I'd LIKE to replace my LCA bushings while the subframe is down. (The DSG case is very close to one of the LCA bolts which prevents complete removal. I think I could probably undo the lower engine torque mount and rock everything out of the way if needed though). 

All the hardware on these is 1x use. So if I'm replacing $30-40 worth of bolts, I'd rather do as much as I can at once.

One of the guys from the OS Discord group I think is sending me his RS3 bushings to get me going before the event, and I'll just send him mine whenever Urotuning ships them out.

Our Preferred Partners
Hhh8RH4zsMNSEwzaYJJVcfkEnzda2rkDhjm5dW38t3ohBCpDQW4Om0MFdvHtqtdU