Panel Gaps, MY vs BMW X3

[sjmcman]

In preparing for our change from BMW to Tesla, I thought it would be interesting to see the measured panel gap differences. So here we go...

I made approximately 40 measurements (in millimeters) across our 2017 X3 and new MY resulting in:

BMW X3:
AVG 4.646
MODE 4.500
STDEV 0.783

Model Y:
AVG 4.329
MODE 4.500
STDEV 1.078

That the mode was the same for each vehicle was interesting, as was the MY's lower average. What also stands out is the higher standard deviation in the MY measurements; lots of variability. Here are a couple box plots, for those with interest:

 

[sjmcman]

I should add the largest gaps on the BMW were where the hood meets the front grille. MY was the bottom of the rear lift gate. Max on each was 7.5 mm.

 

[Koolio46]

Very interesting...I’m coming from a 2018 Audi SQ5 and I’ve thought: imagine if Tesla had Audi’s build quality?

In all honesty though, mine isn’t horribly worse than the Audi I sold...just not as good for the same price point.

 

[theothertom]

Thanks for the data.

 

[azentropy]

Thanks for the info. I was going to do something similar with my Y and my 2015 X3, among other comparisons. However with the whole pandemic thing and my Y getting delayed, I ended up selling the X3 before I received my Y.

 

[sjmcman]

Thanks for the info. I was going to do something similar with my Y and my 2015 X3, among other comparisons. However with the whole pandemic thing and my Y getting delayed, I ended up selling the X3 before I received my Y.

It’s a good way to put some specifics around the panel gap question. Tesla is capable of high quality, but consistency is an issue.

 

[Gizmotoy]

Do you still have the raw data? I wonder if a scatter plot might be illuminating?

The reason being that for most panels a large gap on one side is naturally going to be offset by a small gap on the other side. Those will average out to the center point of whatever the spec for being perfectly aligned may be. Example: (5mm side_1 + 5mm side_2)/2 == 5mm == (2mm side_1 + 8mm side_2)/2. One of those is clearly better than the other, but they have the same average, which is this example's perfect 5mm panel gap. The STDDEV mismatch is hinting at this, I think. The panels move in 3D so it's not exactly that easy, but it seems to me that the logic holds. Or maybe I'm losing my mind. I don't know. It's late.

 

[sjmcman]

Do you still have the raw data? I wonder if a scatter plot might be illuminating?

The reason being that for most panels a large gap on one side is naturally going to be offset by a small gap on the other side. Those will average out to the center point of whatever the spec for being perfectly aligned may be. Example: (5mm side_1 + 5mm side_2)/2 == 5mm == (2mm side_1 + 8mm side_2)/2. One of those is clearly better than the other, but they have the same average, which is this example's perfect 5mm panel gap. The STDDEV mismatch is hinting at this, I think. The panels move in 3D so it's not exactly that easy, but it seems to me that the logic holds. Or maybe I'm losing my mind. I don't know. It's late.

I do have the raw data in Numbers and here are two snapshots of how I initially logged the measurements. Your thoughts are spot on as far as standard deviation, that's exactly what it's telling us. The MY has more measurements with left-right imbalances, whereas the X3 was more consistent overall. That variability isn't that noticeable when you look over our MY, but the numbers sure bear it out.

 

[Koolio46]

It’s a good way to put some specifics around the panel gap question. Tesla is capable of high quality, but consistency is an issue.

I heard Elon say in an interview that they do have to recalibrate the robots from time to time...I wonder if consistency is due to a robot on the manufacturing line gets slightly out of spec/calibration and/or the robot that does 3D scans for alignments gets out of calibration.

 

[dandrewk]

A walk in any residential neighborhood with lots of parked cars on the street will confirm your results. Panel gaps just -are-, no matter what brand/model.

So, new and/or prospective Tesla owners - put away your damn micrometers!

 

[Jimbonic]

I’ve been in manufacturing my whole life working with tolerances of less than .0005 of an inch, so I can understand how variances can and do come up with all the pieces being assembled. I wish I would have taken some measurements from my last car, a Lexus GS 350. Great solid car but in all honesty the gaps in my MP3 are no worse than that car. Only areas the Lexus was better was the smoother ride and very little noise except when accelerating hard. The model 3 seats are just about as comfortable, but I did love the ventilated air from The Lexus seats but now I really love not having to dump $40-60 in for gas every 4-5 days!!

 

[Bad Dolphin]

I heard Elon say in an interview that they do have to recalibrate the robots from time to time...I wonder if consistency is due to a robot on the manufacturing line gets slightly out of spec/calibration and/or the robot that does 3D scans for alignments gets out of calibration.

Probably happened on a Monday and the robot partied too much all weekend...

 

[kcrane]

I can add to the findings. I had a 2019 E 220d estate and as a exercise before taking delivery of an Model 3 weren't around and checked the gaps, seals etc.

The Mercedes was no better overall than the M3, but on balance I'd say the Merc had bigger outliers. The bonnet gaps on the Merc were large and the petrol filter cover was terrible, standing 7 or 8 mm proud of the body on one side.

 

[sjmcman]

I can add to the findings. I had a 2019 E 220d estate and as a exercise before taking delivery of an Model 3 weren't around and checked the gaps, seals etc.

The Mercedes was no better overall than the M3, but on balance I'd say the Merc had bigger outliers. The bonnet gaps on the Merc were large and the petrol filter cover was terrible, standing 7 or 8 mm proud of the body on one side.

Thanks for adding in!

 

[sjmcman]

I can add to the findings. I had a 2019 E 220d estate and as a exercise before taking delivery of an Model 3 weren't around and checked the gaps, seals etc.

The Mercedes was no better overall than the M3, but on balance I'd say the Merc had bigger outliers. The bonnet gaps on the Merc were large and the petrol filter cover was terrible, standing 7 or 8 mm proud of the body on one side.

I assume you don’t see many Tesla’s in the UK yet, any observations from your driving?

 

[asus389]

We have a GTI and a Q5 (2017 and 2013). A family member just got a Y. Subjectively the GTI has the least panel gap issues. The Q5 had some noticeable gaps near the lift gate. The Y had a few, but it wasn’t much worse than the Q5. None of the cars have had body work. /shrug

 

[Technomad]

I assume you don’t see many Tesla’s in the UK yet, any observations from your driving?

UK? Can't move for the things - the M3 is consistently in the top sellers every month! In terms of panel gaps, our M3 is no worse than our previous X5s, and subjectively about the same as my 911. One thing I have noticed - in a recent ice storm (golf ball hail), the Porsche was in damaged, whereas the M3 suffered a number of dimples.

 

[WillN to Go]

On my Model Y the panel fit issues include: The driver's door right edge sticks out, above the rear door about 4 mm, the thickness of the panel. The rear left door looks like it is rotated clockwise, so tight gap at the top wide gap at the bottom. Like some huge weight yanked it down when it was open. These flaws are readily noticealbe from a distance. There are several other discrepancies. And while I'm function over form (If I could save $1000 buying a dented Model Y, I'd have done it) these are noticeable and unacceptable.
In a YouTube video there was discussion of Tesla panel fit. A friend of the video maker pushed Tesla's service about as far as it could go, and then he took his Tesla to a body shop and they tweaked it that last bit. In another video the Tesla service techs were yanking and hammer (malleting?) to align the panels. It's definitely a skill that grows with experience.
One might think it's just bad quality control, but I suspect that Tesla is trying to produce cars as fast as they can and these mostly cosmetic aspects are being tolerated so as not to slow down the line. Eventually I expect Teslas will come off the line more perfect than any other car. And you might think, well there ought to be a few teams working their way through the Tesla lot fixing the panels on the new cars - and I'll bet there are, it's just they can't keep up with production and deliveries.
I've hung a lot of house and building doors. It's a three dimensional problem. A door in your house I could probably tweak into place within minutes, on a car... you don't want to make the panel look right and end up with a gap in the seal.

 

[Sunshine State]

One thing to keep in mind that those BMW vs Tesla measurements compared were on the "older" generation BMW X3 being compared to the new Model Y. The new generation X3 is light years better in design and tolerances from the last generation X3. My daughter had a old generation 2017 X3 that we replaced with a new 2019 X3 last year and the new car is at a much higher level of build quality. It would very interesting but likely a huge embarrassment to compare the new BMW X3 to a new Model Y unfortunately. I also have a new X5 that is amazing in quality but not electric.

 

[sjmcman]

UK? Can't move for the things - the M3 is consistently in the top sellers every month! In terms of panel gaps, our M3 is no worse than our previous X5s, and subjectively about the same as my 911. One thing I have noticed - in a recent ice storm (golf ball hail), the Porsche was in damaged, whereas the M3 suffered a number of dimples.

A fine Scottish sky in your photo!