CMP Test Day

Posted on 2015/09/21 by Chad

Note: This event took place on 30 July 2015, just getting caught back up with posts.

My last test day at Carolina Motorsports Park ended prematurely when my radiator popped off its mount and the belt cut into the coolant expansion tank. With the car back home I pulled the plugs to check the compression since it saw temperature spikes over 250F. Fortunately the numbers came back good, actually matching the numbers I had taken when I first installed the engine.

I put the car back together, this time with a new CSF OEM+ radiator and a cooler 80C thermostat while ditching the stock fan, and headed back to CMP to get some testing in. I knew the car needed some chassis tuning after its first race post-crash.

It was a low turn out at the track, so we were able to double up on track time. That’s 8 full 30 minute sessions. That’s a lot of driving in a single day, so I cut a few sessions short and skipped the last one. What I did discover was that I needed a little more rear end grip. The week prior I had done some ride height adjustments and corner balanced the car at a friend’s house. The first couple sessions were a mess, I either spun or got into some serious tail out action.

I brought the car in and decided to stiffen the front bar one notch. This had an immediate positive impact, the car was much more neutral, at least as much as I could get it on CMP’s slick surface. Either way it felt much better and gave me more confidence going into my upcoming NASA race at Road Atlanta.

As for the engine temps; the car ran much cooler. I was seeing a high temp of around 202F, albeit in completely clean air. Looks like removing the stock fan and shroud allowed for better flow through the radiator. I also had done a little work with foam rubber insulation to force air through the radiator rather than around it.

Custom Oil Temp Sensor for AiM MXL

Posted on 2015/09/21 by Chad

The AiM MXL datalogger is a great piece of equipment, but it does have a few shortfalls. One of the major issues is its inability to interpret the BMW stock oil temperature sensor over the CANbus. I’m no computer scientist, so let’s just say what it does; it reads a constant -54F. Obviously this is no help to me as the driver trying to determine whether my engine is healthy or not. Since there was no solution on the horizon from AiM, I decided to just go with an aftermarket sensor.

The best place to check for engine oil temps is the pan, and Turner Motorsports makes a very nice adapter to easily allow you to add a sensor. The mounting plate actually replaces the stock oil level sensor located in the bottom of the pan. This isn’t an issue for a race car, you should be checking your oil level the old fashioned way before every outing.

The adapter flange is threaded in M12x1.50mm, allowing you to simply thread in the BMW stock sensor. Unfortunately I need an aftermarket sensor to connect with the AiM MXL. AiM only offers the sensor in 1/8″NPT/10mm/5mm, so an adapter is needed. I used an Auto Meter 2277 adapter, which is a simple brass bung adapter with a copper crush washer.

Note: in the pictures below, the sensor shown is not correct. AiM accidentally sent a brake pressure sensor. I caught the mistake before installing in the car, and the correct fluid temp sensor was sent in exchange.

With the new sensor installed and plugged into the AiM MXL, all I needed to do was hookup the laptop to setup the new sensor. That’s it, now I had a proper oil temp reading… only one problem; the AiM PT100 temp sensor has no probe. Its length terminates at the ends of the threads, so it does not protrude into the oil sump. This problem is compounded by using the Auto Meter adapter. Essentially the sensor is located down in a hole, away from the normal flow of oil. Sure it’s better than no oil reading, but there has to be a better solution.

The photo at the right shows the problem with using the AiM PT100 and adapter.

With no other sensor option from AiM, the only other possibility would be to create a custom sensor. AiM’s RaceStudio allows for the creation of custom sensor profiles, you just need the resistance readings at various temperatures. After a little searching on the internet I found a GM Delphi temperature sensor that came in M12x1.5mm thread and had a proper long probe.

See more about the sensor here at Ballenger Motorsports’ website.

Another hurdle to overcome is that the AiM analog inputs use voltage difference sensors rather than resistance based. So this GM temp sensor would require a pull up resistor soldered inline to achieve a mV reading. A stripped down AiM 719 (4 pin plastic) patch cable would provide the basis for the harness, and the GM plug on the other end was provided by the same store where the sensor was purchased.

The photo above shows how I soldered in a 1k ohm resistor, bridging from the Blue (Vref) wire to the White (Vout) wire. The Black wire is ground and the Red wire is 12V and not used.

The following photo shows everything wired up and covered in heat shrink. The GM sensor uses Weatherpack style crimped pins. I picked up a cheap set of crimpers from eBay for the job, which is very simple. You can see the difference between the GM (long probe) and AiM PT sensor in a side-by-side comparison.

Now with everything wired up, installed in the car, and plugged up, you have to build the custom sensor in RaceStudio. This requires a little math, as you will be converting the resistance based table provided with the sensor to a variable voltage table that the AiM MXL can use. To make life a little easier, I made an Excel spreadsheet, where all you have to do is plug in the temperature, corresponding resistance values, and the resistance of your pullup resistor. The reference voltage should be set to 5v for this application.

Since the GM sensor data was provided in Celsius, I also inserted a formula to convert to Fahrenheit. Essentially plug in the numbers in the green fields to get results in the white fields. From there you simply copy the Vout millivolt (mV) values and corresponding Fahrenheit temps to your custom sensor in RaceStudio.

Note: RaceStudio will only accept 20 values per sensor. Your beginning and ending values must contain your entire range of expected temperature readings. In other words there is no point in including values of -40F when your car will never see that. However if you start your values at say 77F and on a cold day your sump is 50F, you will get an error message on the MXL as it doesn’t know how to interpolate a reading off the chart. So pick the best 20 value range that covers all expected temperature readings.

Click the screenshot of the Excel spreadsheet to download a .zip file containing a copy that you can edit for your own custom temp sensor. The tables are already filled in for the GM Delphi sensor. I also included images of how to fill in the custom sensor fields, AiM 719 wiring colors, as well as wiring of a pull-up resistor.

Known Issues: Your custom sensor will display on you MXL (or any other AiM display logger) just like an AiM proprietary sensor would. I have my dash set to always display water and oil temps. Your logs will also capture this information correctly. Unfortunately the AiM SmartyCam does not handle custom sensors properly. The SmartyCam overlay will try to convert the Fahrenheit temperature into Fahrenheit, which of course leads to erroneous readings. AiM (both America and Italy) are aware of this problem, but a fix doesn’t appear to be coming very soon. The screen capture on the side shows my oil temp at 511F, which of course is incorrect. It’s a minor nuisance, as the temp is correct where it really counts (on the dash and in the logs.)

I’ve been told by AiM US that there is a new temp sensor in the works that has a proper probe. I will update if and when it becomes available.

Update 8/24/20: AiM did come out with a new liquid temperature sensor, with an elongated probe. It can be found here:
https://www.moreheadspeedworks.com/product/aim-tr-oilwater-temp-sensor/

And even a better solution for M52TU and M54 engines, is MSW’s flange kit, which is already threaded for 1/8th NPT sensors, which saves the hassle of having to use adapters:

https://www.moreheadspeedworks.com/product/msw-bmw-oil-temp-sensor-flange-m5x-s54-e36-e46-z3-z4/

Spec E46 Test Day Ends Early @ CMP

Posted on 2015/09/21 by Chad

Note: This event originally occurred on 12 July 2015, I’m a bit behind in my posts.

Following my first full race weekend, which was a mad dash to ready the car, I knew I had a few bugs to work out. Chief among those issues was a persistently warm running engine. Of course this was in the heat of the South Carolina summertime, but I wanted to get my temps a little lower.

To that end; I installed a larger CSF all aluminum radiator specifically for the E46 non-M chassis. Unfortunately the mounting points weren’t ideally designed to interface with the factory plastic mounts. I did my best to modify the mounts to work, installed it, and off I went to Carolina Motorsports Park for a day of testing.

I went out on my first lap and found that the temps were much lower. Prior to the new radiator I was averaging ~212-215F in the heat of the day. Now I was sitting at around 202F, much more reasonable. I pushed on harder for the next couple laps, besting my previous times by more that 2 seconds. Then on the third or so lap I got a warning light briefly for engine temps. It went away as I looked down at the dash to decipher what the problem was. Then a few seconds later it reappeared. By now I was paying closer attention to the numbers, and I watched as the water temp spiked to over 250F then back down.

I immediately let off the throttle, short shifted, and brought it in to the paddock as quickly as possible using as little engine as possible, coasting the last hundred feet or so with the engine off. Hopping out I could see steam rising from the engine compartment. A peek under the car showed hot water dripping to the ground. This certainly wasn’t any good. Popping the hood confirmed my concerns about the mounting of the CSF radiator. It had slipped from its mounts and shifted rearward where the expansion tank made contact with the serpentine belt. The belt made quick work of the plastic tank, sawing right through it.

My day had come to an early end with a destroyed expansion tank, a radiator that wouldn’t sit securely, and possible engine damage. Later on, after everything had cooled, I also discovered that high temps had destroyed the expansion tank cap. The lower portion that retains the sealing o-ring was completely missing.

The good news: CSF was quick to help rectify the situation. They sent me one of their OEM+ radiators, covered the cost of the expansion tank, and promised to redesign their all-aluminum radiator. Hopefully they get it corrected soon, because other than the mounting problems, it’s a solid piece.

And finally a short video of my one hot lap before the radiator mounting failed:

httpvh://www.youtube.com/watch?v=-DuvY_UuF9I

Spec E46 Race @ Carolina Motorsports Park

Posted on 2015/09/21 by Chad

Note: This Event originally took place on 27-28 June 2015, sorry for the delay in posting.

Having finished the requirements for my provisional racing license, I was looking forward to the next opportunity to race my Spec E46. The upcoming NASA-SE event at Carolina Motorsports Park (CMP) looked very promising, with a possibility of five other SE46’s in attendance. One slight hiccup; my car was still in pieces. Several delays in parts shipping meant I was way behind my anticipated competition date.

I had already committed to this race, having already cancelled the previous Road Atlanta event due to the crash, and I really didn’t want to miss another opportunity to get some much needed track time. The week leading up to the event became a mad thrash. I was working until the wee hours of the morning right up to the day before the race. The good news: CMP is only an hour away from home, so at least the few hours of sleep I did get were in my own bed.

I also needed an annual tech inspection the morning of the event, so everything had to be just right. I escaped with a few minor discrepancies which were easily corrected and was able to make the race. I had barely got the car off the alignment rack the night before, so there was zero time to shake down the car, much less tune the chassis. I’d run what I brung as the saying goes.

I missed the first practice, fiddling with a few last minute things, but was able to go out in one of the HPDE sessions to get a feel. I knew the track was slick from my previous test day, but having an untested car made it much more of a challenge. My goal was to just keep the car on track and avoid any contact, speed was way down my list of things I was worrying with.

I qualified at the back of the SE46s, as expected given the all-star veteran competition I was up against. No big deal, I was here to learn. We, as a class, decided to start at the back of the Thunder run group in our own separate start rather than gridding on pure time. The first race started off cleanly, but I was out of it both mentally and physically. The lack of sleep had caught up with me big time, and I couldn’t maintain my concentration and felt like garbage, my times were reflecting my condition.

I continued to press on, just looking to finish when a red warning light illuminated on my AiM MXL digital dash, water temps. I was at 230F and it looked as though it would keep climbing. I had pressed my luck enough already, so I pulled off that lap and putted back to the paddock area. I couldn’t find any obvious issues with the cooling system, all signs pointed to trapped air bubbles. This is the price you pay when rushing to get a car ready at the last minute.

I decided to pack up for the day, let the car sit and cool overnight and get some rest. Another SE46 driver had already had worse luck when one of his rear shock mounts failed, resulting in a destroyed MCS shock. I left early to get caught back up on sleep and hopefully have better luck on Sunday.

The next day I jacked up the front of the car and bled the cooling system thoroughly. It looked like my suspicions were correct and the engine had some significant amount of air trapped in it. Feeling recharged from a long night of sleep I headed back on track, this time in a much better frame of mind. The car still wasn’t working as well as I had hoped, oversteering terribly on exit and pushing on entry, but it would have to do.

(Note how I was still missing the trim under the headlights and decals from the new fender)

Again I started at the back of the SE46 pack, where we started as group behind the rest of the Thunder field. This time I felt a little more confident making passes and tried my best to hang on to the back of the SE46 in front of me. I certainly wasn’t pushing for overtakes, rather waiting for the safest opportunity to get by slower traffic. Towards the end of the race I could see I was closing back up to the next SE46 up the road, but every time I tried to push, the car would just lose traction on exit. I was at opposite lock over and over, watching the Mustang and other SE46 in front grow the gap.

Despite the headaches and lack of a good setup, I still had a blast driving with other SE46s and finally getting a chance to mix it up in traffic in an actual race. This was, after all, the first race where I got beyond the first lap. It was great to meet and hangout with other racers in the paddock, gleaning knowledge wherever I could.

Finally a little video from the weekend. Due to my zombie-like state on Saturday, I neglected to bring the memory card for the camera. I made sure to get it for the Sunday race. Here’s the race in its entirety:

httpvh://www.youtube.com/watch?v=GFrz5KTHV20

SpecE46 Wreck Repair

Posted on 2015/09/08 by Chad

After the Spec E46’s debut at Barber Motorsports Park, I was left with a crumpled and undriveable car. The initial visual inspection at the track showed that I’d be in need of the following:

Hood
Left Fender
Bumper
Left and Right headlights
Left control arm
left tie rod
Airbox
Left aluminum engine mount
Apex Wheel
Radiator core support

Once I had the car home, I immediately unloaded it directly from the trailer into the shop and up on jack stands. From there I started to dismantle the front end. With the hood, fender, and bumper off, I got a much better look at what lay ahead. I then found the following would need replaced:

Front left wheel housing
Front subframe
Left and right engine mounts
Steering rack
Power steering pump pulley
Radiator
Front bumper crash bar

The whole left front had been pushed to the right and rear, forcing the wheel well up into where the DME box and brake master cylinder/ABS/booster are located. Several welds had pulled apart, and I don’t think a simple frame pull would get it back to where it needed to be. So I opted to order brand new OEM sheet metal. BMW offers the entire wheel housing and frame horn as a replacement part, though it’s certainly a special order item.

The engine came out, which was in surprisingly good order, other than the torn Bimmerworld engine mounts and cracked aluminum mounts. I got lucky in that none of the bolts pulled from the block. The crash did manage to put a ding in the Y-pipe, but I decided to leave it alone.

The engine subframe was twisted at the engine mounts, so it went in the trash. I sourced one from a junk yard, as well as a used core support and crash bar. The subframe was cleaned up and a set of Turner reinforcements were welded on.

I completely cut away the original wheel well by drilling out all the factory spot welds, leaving the front left firewall completely bare. The new housing was squeezed into place, and leveled using tape measures and a plumb bob. I welded it back on using the factory spot weld locations. The interior of the transmission tunnel needed a patch plate where some original metal tore during removal.

The left control arm bushing had stripped the threads from the frame, so I ended up drilling it up a size and inserting a helicoil.

With the new metal final welded into place, I gave everything a good scuff with a 3M pad, followed by cleaning, then a coat of automotive primer. I then sealed up the gaps with some plain seam sealer before painting everything a matched 300 Alpine White and a coat of catalyzed clear.

Everything went back together as it did the first time. I did find that both frame horns slightly shifted (~1/8″) to the right when checking the frame for squareness. A little extra effort on panel gaps brought everything within factory spec and the car aligned properly as well. I’ll let the photos do the rest of the talking.

Note: While everything was apart, I took the opportunity to replace the brake master cylinder and booster. These weren’t damaged in the wreck but were original to the car (~220k miles).

[foogallery id=”7886″]

Eat, Sleep, Tinker.

No, it's not done yet…

Category Archives: BMW Spec E46

CMP Test Day

Custom Oil Temp Sensor for AiM MXL

Spec E46 Test Day Ends Early @ CMP

Spec E46 Race @ Carolina Motorsports Park

SpecE46 Wreck Repair