[Max Well]

Mid-size SUV Group for analysis established:

Here is a corrected chart of the contenders for the acceleration analysis in the X3's category. The definition for this group remains vague to me, and it has required a fair amount of time to try to understand the dynamics of what drives someone to purchase an X3 or 'similarly-sized' vehicle in this group. I have no definitive answers but am sharing my perspective.

It has been confusing over the years as I read those who wish to trade in their X3 for another Mnfr's offering when it does not match the basic physical constraints which makes the X3 so appealing for many of us. Case in point, the X3 and the Macan. The latter is no doubt an amazing work of technology, but the cargo space is too small for my needs, just as the old X1 was too small and the X5 is too large for the requirements my lifestyle sets forth.

What leads me to the 'X3-sized category' in the first place? A performance-oriented vehicle which can carry the load and cargo I need to transport in an internal environment comfortable for myself as driver and my family. The performance and handling characteristics desired whittles the field down significantly. Then one has to determine if the interior dimensions are comfortable while (whilst) still providing the cargo capacity with the rear seats upright.

This list has been narrowed to include vehicles with a rear cargo capacity of roughly 20-30 cubic feet, which meet the SUV/SAV concept, and which are considered to be performance-oriented (this latter point is subjective). Also, an attempt at forcing exterior dimensions to be within a certain range has been made, but because of the length, width and height variations this has been difficult to achieve. I will now run the acceleration profiles for these so an acceleration curve can be developed for each. Note that neither Porsche nor Mercedes vehicles have not been included in this group, as I have simply not been able to find one that fits within the cargo space requirement. This is not meant as an afront to the great vehicles they produce. It simply indicates a model with all of the aforementioned attributes has not been identified, but if an error has been made in this selection process I do apologize in advance and will certainly add it after suggestion is made. After studying this, it seems Mercedes' introduction of the GLC, which appears in all preliminary reports to fit into this group perfectly, will no doubt become a major contender in this field but data has not been found to fill in its fields to allow analysis at this time.

[Max Well]

Apologies for the delay in sharing the acceleration data as discussed above, but I ran into some difficulties with running the other Mnfr's entries in the construct. Whereas the equations seem to capture the recognized 0-60 mph, 1/4 mile, and top speed figures very well for BMW, Audi, Mercedes, and Porsche, the other Mnfrs' offerings did not match what the model indicated they should be hitting.

That can be from a number of variables this model can't pick up, so rather than trying to change multiple parameters of the equations for those Mnfrs, I have opted to simply diminish the 'rated HP' for for those examples to get them in line with their published numbers. I have tried to validate those publicized numbers by cross-referencing across at least four different sources, but as those may actually be collating data from each other, this is far from an exact science.

As this has taken a lot of time to even run the data I have not yet graphed the results, but will as time allows. And interesting that BMW, Mercedes, Audi, and Porsche are spot on with the modelling - it would seem to suggest that they are each at the top of their game with regards to maximizing 'engine efficiency'. And by that I mean, the same variable (79.2% of rated SAE HP) has been reached by each of them.

Is that the thermodynamic limit as to what can be reached? Perhaps. This is becoming more interesting...

[Max Well]

Here are Acceleration and Distance analyses for this ‘Luxury SUV’ category of 8 vehicles of fairly similar size and curb weight with cargo capacities 20.0 cu. ft. - approx. 30 cu. ft. In order of acceleration performance, the rankings results from top to bottom are: Audi SQ5 3.0TFSI; BMW X3 xDr35i; Infiniti QX70 3.7 AWD; Volvo XC60 T6 R-Design; Cadillac SRX AWD; Acura RDX-SH AWD; Land Rover Evoque Si4; and Land Rover Discovery Sport Si4.

As indicated earlier, whereas Audi, BMW, Mercedes, and Porsche vehicles fit into the model’s construct quite well with 0-60 mph, ¼ mile, and top speed results similar to publicized numbers, the other Manufacturers’ vehicles did not. The model wanted to make them faster than their published values indicate. For example, in order to have the Acura RDX model reasonably agree, the HP had to be decreased from 279 SAE rated to only 210 HP (-24.7%), for the Cadillac SRX from 308 SAE to 245 HP (-20.5%), for Infiniti QX70 from 325 SAE to 285 HP (-12.3%), for the Land Rovers from 240 SAE to 210 HP (-12.5%) and even then, the Discovery Sport modeling was still faster at 6.9 sec than its published 0-60 mph time of 7.8 sec, and for the Volvo XC60 T6 from 325 SAE to 275 HP (-15.4%).

The reason for this discrepancy is unclear, but it might suggest these vehicles may not utilize their power during acceleration as efficiently as the vehicles from Audi, BMW, Mercedes, and Porsche, at least in this dynamics model.

Note the hypothetical X3 – M and X3 – 40i (if they existed currently in the LCI frame) have been included to see where they might fit in this current competitive arena. It provides a clue as to why BMW may be bringing them into the mix, and why Mercedes is introducing the GLC and upping the ante with possible AMG variations on the horizon. It wouldn’t be surprising to see Audi following suit with a more potent engine option in future SQ5 iterations as well. To be clear, I have no affiliation with automotive manufacturers or businesses even remotely affiliated with automobiles, their parts, or sales. This is just my attempt at understanding the SUV category my family prefers.

The final Tables show a summary of the vehicles and a ranking for 8 variables with a simple average. This can be misleading as some of the differences in rank levels are negligible, but this makes it easier to compare and one can refer to the table to see if they matter. These analyses do not take into account other important variables when purchasing a vehicle (initial cost, reliability, comfort, handling, etc.), but at least a quick comparison is made possible for these objective parameters.

[FSociety]

Here are the fastest 3 time slips of 9 runs

339 is me
14' X3 xDrive28i (n20) w/ RaceChip Ultimate otherwise stock car
Full Tank of Gas, 93 octane from 76Gas + 32oz of Torco Accelerator, 245/45-19 Continental DWS

The other car is mge92
13' X3 xDrive28i (N20) w/ Dinan Stage1 & Supersprint QuadMuffler
Full Tank of Sunoco 93 + 32oz of Torco Accelerator,
245/40-20 | 275/35-20 Pirelli P-Zero RFT


Now Crashnbrn5 post your timeslip

[Max Well]

Here is a Table summarizing ROBNYC's, mge92's and Crashnbrn5's runs at Island Dragway (from the slips they provided) and the resulting graph. Because these analyses are aimed at assessing 'performance' of the X3 in its different forms and tunes, I chose to include the simulation's curves for the xDr35i and xDr28i so one can see where they stack up against the model. [Edit 9 Sep 2015 - only the straight drag slip times are now being used, as the previous methodology of subtracting the R/T times was skewing the data.]

[FSociety]

Max Well unfortunately Crashnbrn5 wont be able to make it this friday to the drag for Rematch
But I will be there along mge92 who will be ECU Flashed and it will be a Piggyback tune on a stock car vs ECU flashed otherwise stock car. None of us have intakes or downpipes to make it a fair challenge.

So those time slips I posted of F25 X3 xDrive28i w/ 245/45-19 Conti DWS (non-RFT) with RaceChip on D and E setting. This week will be RaceChip on F, 1, 3, 4 setting (depending on weather too i dont want to push it too high).
I know at 2/3/4 I was able to match the CrashnBrns5 X3 35i in the winter on a 0-60 run but his looked faster because he kept going to 70mph while I let go off accelerator.

At the highway yesterday on a rolling, RaceChip on E was able to match the X3 35i once again even with A/C on. I'd bet with the A/C off I would of been able to match the 35i even better. C setting is stock, D is above, E is 2, F, 0, 1, 2,3,4,5,6,7 settings are available too. Although in my testings I have no reached above 4 since i dont want to test the limits and blow off my chargepipe lol

[Max Well]

Thanks for the update, Rob. Even though Crash will be missed by you folks I am sure, as we already have his 'stock' 35i data in the database, unless he changed things up by unloading weight, it seems likely those previous runs are a good indication of the 35i's performance. What we don't have is 'stock 28i' data, however.

My gut feeling is that the model slightly over-estimates the efficiency of the X3's power usage, as one can see Crash is consistently below the model's prediction (albeit very slight). As the equations are identical for the 28i, and assuming BMW was able to match the efficiency of power transfer during acceleration between the two engine types, then by having mge92 even with the 28's prediction with his Dinan-tuned engine, I am guessing a true 'stock' 28i should land below the model's 28i curve similarly as Crash did.

So if you can make 2 or 3 runs with setting 'C', if that is truly considered 'stock' (or if someone else is coming who owns a stock 28i), then that will provide a good validation for the 28i's curve, which we still do not have.

[FSociety]

Originally Posted by Max Well

Thanks for the update, Rob. Even though Crash will be missed by you folks I am sure, as we already have his 'stock' 35i data in the database, unless he changed things up by unloading weight, it seems likely those previous runs are a good indication of the 35i's performance. What we don't have is 'stock 28i' data, however.

My gut feeling is that the model slightly over-estimates the efficiency of the X3's power usage, as one can see Crash is consistently below the model's prediction (albeit very slight). As the equations are identical for the 28i, and assuming BMW was able to match the efficiency of power transfer during acceleration between the two engine types, then by having mge92 even with the 28's prediction with his Dinan-tuned engine, I am guessing a true 'stock' 28i should land below the model's 28i curve similarly as Crash did.

So if you can make 2 or 3 runs with setting 'C', if that is truly considered 'stock' (or if someone else is coming who owns a stock 28i), then that will provide a good validation for the 28i's curve, which we still do not have.

I can do a 28i (2 runs "if my 1st i don't get a good launch then i'll attempt a 2nd run" on stock with the RaceChip dummy on so it will have RCU disabled), then 1-2 on C, 1-2 on F, and hopefully i get there on time to be able to test higher settings.

Although my gut feeling says RCU on C (stock setting) will be similar to the Dinan stage1 performance and I rather not waste more than 2 runs on stock 28i and RCU stock performance since I was only able to test D and E setting on RCU (which is 1 above C, and 2 above C).

We will have a F30 328i with ActiveAutowerke8 tune so it will be nice to see how the smaller sedan N20 engine with a finely tuned piggyback will do.

[Max Well]

Thanks, Rob, and I certainly agree - you are there to have fun and to run fast with your installed system. It'll be a great opportunity to try a number of its different settings, so I wouldn't worry about trying to fit in the stock runs. Maybe another time somebody with an 'off-the-shelf' 28i can join you for a night at the races. It might help to see what the 28i's baseline really is on that track so you can each compare your different tunes and their actual effects.

[FSociety]

Originally Posted by Max Well

Thanks, Rob, and I certainly agree - you are there to have fun and to run fast with your installed system. It'll be a great opportunity to try a number of its different settings, so I wouldn't worry about trying to fit in the stock runs. Maybe another time somebody with an 'off-the-shelf' 28i can join you for a night at the races. It might help to see what the 28i's baseline really is on that track so you can each compare your different tunes and their actual effects.

Indeed, I've been trying to bring out stock 28i but don't know any in tri-state and no one in the forums.
And it doesn't help if its a Non-X3 because of the weigh difference etc.

So I will be running in fully stock with RCU off for 2 runs then w/ RCU on after every run.

[Max Well]

That is commendable, Rob, for without the baseline data it is difficult to compare the effects of different tunes.

To make it easier to identify differences, I ran a focused analysis on the 1/8 and 1/4 mile times:

[Max Well]

Analysis run on the BMW models Crashnbrn5 is considering when his X3-35i lease runs out. Summary table with ratings located in his thread, 'I'm in such a bad "time zone" HELP' http://x3.xbimmers.com/forums/showthread.php?t=1163269 .

[Max Well]

Summary Chart for Island Dragway runs 21 Aug 2015 (documented slips provided to me by ROBNYC in a PM):

[Max Well]

Performance Data from Island Dragway 21 Aug 2015 (images of actual slips submitted via PM)

As hsrashid5 participated last Friday in his F10 535i xDr, an Acceleration curve was prepared for that model type and is included. As mentioned previously it is clear the simulation is over-estimating engine efficiency (using 0.792 in the construct) to a slight degree in all model types. However, as Crash had shown previously with his stock X3 - 35i runs, the track data mirrors the sim's reference curve well - just at a different set point. So once one validates their 'stock' runs against the model's curve, it should prove easier to identify how each modification affects performance on their vehicle type. If you can recall the specific settings you ran with for each of your nine runs (and include here as refererence), Rob, it might assist others who have your setup (or are considering such) as to what they might expect since you have now provided 'real-world' data. I placed all of the runs in chronologic order based on the 'Round number' as written on each ticket if that helps.

[FSociety]

Thanks Max Well we need this as a sticky!

jamoka3 mge92 Crashnbrn5 Stubok

[Max Well]

Summary Table for ROBNYC and jamoka3 at Island Dragway 21 Aug 2015:

I'm not sure if this provides clarity or induces confusion, but it seemed reasonable to compare the stock vs piggyback-tuned times of Rob's and jamoka3's fastest runs. Rob used RaceChip Ultimate and jamoka3 used Active Autowerke's ACTIVE-8 modifications.

As the Simulator provides a reference point for what the model calculates the speed should be at any 1/10th of a second out to 88 seconds for the model type with an engine efficiency of 0.792 (likely over-estimating), a relative standard (albeit coarse) can be derived. For example, if someone in Colorado wished to run their X3 - 28i at Bandimere Speedway, which is 5800' above sea level, the Simulation could be calculated for that altitude and provide a rough guide as to how one's tunes compares to those of others across the globe. The more data points which could be averaged would enhance the confidence but that isn't feasible when one has to pay for track time at certified raceways, and a host of other variables come into play (temperature, track conditions, tires, etc.), but it should provide at least a layman's perspective.

I've also included a percentage of the individual's Stock Time (in seconds)/Piggyback Time (in seconds) for the 1/8th mile and Qtr mile times which is independent of the simulator.

[Max Well]

As one more way of analyzing Rob's data from 21 Aug 2015, I took the Simulation model's X3-28i xDr construct and ran it with different permutations of HP in 10 HP increments from the stock HP +/- 50 HP. That allows a series of curves to be generated showing how HP may affect the slope of a vehicle's performance graph in this estimated simulation.

These 'reference curves' (I use that term loosely) thus allow one to make a rough extrapolation of the HP differences between runs in the two short time intervals in which the drag slip provides the distance, time, and the speed. I am attaching the global graph from 0-16 seconds, then highlight the two areas I zoomed for more detail (the 8.5-10 sec and 13.8-15.4 sec intervals). In Rob's examples in the first interval, his stock run data point is roughly -22 HP from the model's stock X3-28i curve, whereas his best run is at +10 HP in c/w the simulation's stock curve. So it would seem reasonable to conclude, in this small sample graphing one stock run and his two best tuned runs, his 'tune' allowed a 32 HP increase in the early time (and speed) interval, and about +25 HP in the second interval.

I would leave it to the statisticians to determine how many runs of stock vs tuned one would need to reach statistical significance, but that is not the point of this layman's analysis. This is simply a chance for those of us without dynos and high-end software to be able to put at least some science behind the performance of our vehicles. If folks would be interested, I can share graphs of the simulation's curves for these three timeframes which one could print and then simply draw in their own data points for personal recording.

[FSociety]

Thank you Max Well your work is priceless!

Everyone please take advantage of this time consuming work!

[Max Well]

Thank you for your kind words, Rob, pleased to assist.

To make it easier for you and your X3 28i colleagues at Island, I thought if I prepared a templated chart on which you could record your times and notes on one sheet, along with the reference curves from +50 HP to -30 HP for the 8-10 and 13-15 second intervals (which are the only two in which distance, time and speed are provided), it would be easy for you even while at the track to plot your own data while you are there. Then you would be able to have a more real-time assessment of how your differing settings are affecting the runs. If other xDr models would like their curves prepared I can. The chart will work for anyone.

[Bimar]

Thanks for the work Matt, much appreciated.

[Max Well]

Originally Posted by X4guy

Thanks for the work Matt, much appreciated.

Thank you for the kind words, X4guy - I'm happy to be able to share data on these amazing BMWs. If you think any of the X4 folks would be interested in dedicated curves for your line I'd be pleased to prepare those as well for you.

[Bimar]

Originally Posted by Max Well

Thank you for the kind words, X4guy - I'm happy to be able to share data on these amazing BMWs. If you think any of the X4 folks would be interested in dedicated curves for your line I'd be pleased to prepare those as well for you.

That would be most welcome if it's not too much trouble.