FUD - Businesses insider "The star of 'Aladdin' claims a defect in his Tesla Model 3"

[brianman]

Exhibit B is the EDR data

Some things that caught my eye:

• Odometer at Event Time Zero (km): 96.0
• Time (sec) = -1.4s | Vehicle Speed (mi/h) = 56.0 | Accelerator Pedal (%) = 0.0
• Time (sec) = -1.0s | Vehicle Speed (mi/h) = 54.0 | Accelerator Pedal (%) = 26.8
• Notes indicate wheel detaches within [-1.4s, -1.0s] yet the vehicle accelerates (-0.2g to 0.6g) immediately after? (Perhaps that's a vector thing with rotation.)

 

[AlanSubie4Life]

Time (sec) = -1.0s | Vehicle Speed (mi/h) = 54.0 | Accelerator Pedal (%) = 26.8

Yeah I was thinking a sudden jerk or bump briefly caused an accelerator "stab" or something, since typically the foot would still be on the accelerator in the Tesla with standard regen. He lifted, then this quick stab happened (presumably when the "bad thing" happened), then it quickly goes to zero % accelerator again.

To me it seems like the wheel fell off; I just don't know why! Would be nice to have some video or a witness.

 

[mongo]

tl;dr; Whoever marked up the EDR report doesn't know that they are talking about. Steering matching roll matches lateral acceleration...

All the EDR data is actually included in the lawsuit- with the lawyers claiming it shows the wheel came off ~1 second before the actual crash

Complaint Filed Copy

Exhibit B is the EDR data

A few issues I see:
The filing claims roll is the distribution between front and rear axles. That is incorrect. Roll is left/ right body tilt (not turning). Fore/ aft is pitch. Event Data Recorder (EDR) Reports: Understanding the Basics - Expert Article | Robson Forensic

I'm not sure what they meant by the note "Yaw rate measures the angle on the vertical axis", but that is wrong. It is change in angle, and is it rotation around the vertical axis (car turning). The data shows the car was rotating right then left then right (assuming they got R and L correct).

The roll rate aligns with the lateral acceleration.

The filing claims the steering wheel locked (Item 10), However, the data Steering Wheel Angle in Exhibit B shows this to not be the case.

The steering input matching the lateral acceleration and roll rate.

The EDR data also shows no ABS warning light at the time of crash, depending on the time it takes to throw a fault, this may indicate that the wheel was still attached. Especially when combined with the log showing stability control was active at 0.5 and 0.0 pre-event trigger.

The Longitudinal acceleration shows the car was speeding up until the impact, expect for one data point which could have been due to traction control.

 

[Twiglett]

I'm assuming Tesla already have the logs

 

[Daniel in SD]

It's disappointing that there is no individual wheel speed data which would pretty conclusively tell us when the wheel came off.

The Longitudinal acceleration shows the car was speeding up until the impact, expect for one data point which could have been due to traction control.

The polarity of that plot is clearly inverted. The car is decelarating after -1.6s.
This is very interesting:

The ratio of vehicle speed to rear motor speed is nominally 1:118 but at -1.2s it is 1:108. Perhaps something did happen to a wheel speed sensor. I wonder how the car calculates vehicle speed?

 

[AlanSubie4Life]

Yaw rate measures the angle on the vertical axis", but that is wrong. It is change in angle, and is it rotation around the vertical axis (car turning).

I like your description much better, but I think they are describing the same thing (other than their confusion on the rate of angle change vs the measure of the angle, which is a significant difference).

The data shows the car was rotating right then left then right (assuming they got R and L correct).

I am assuming it hit the curb and that tended to rotate the car counterclockwise. That looked like positive values for yaw rate in the plot near time zero, so I used that to determine the sign, and to assume that sign for yaw rate corresponds to a left rotation.

So it looks like left, right, left to me. I could be wrong.

The Longitudinal acceleration shows the car was speeding up until the impact, expect for one data point which could have been due to traction control.

The velocity data shows the car was slowing, lost about 12mph before event time 0.

Also the longitudinal acceleration went positive (deceleration in this case) to match it. I don’t know why the sign of the acceleration does not match the post-event acceleration data (where negative looks like deceleration).

 

[mongo]

It's disappointing that there is no individual wheel speed data which would pretty conclusively tell us when the wheel came off.

The polarity of that plot is clearly inverted. The car is decelarating after -1.6s.
This is very interesting:View attachment 416384The ratio of vehicle speed to rear motor speed is nominally 1:118 but at -1.2s it is 1:108. Perhaps something did happen to a wheel speed sensor. I wonder how the car calculates vehicle speed?

@AlanSubie4Life
Yeah, i tried to find a reference to EDR coordinate systems and failed. Deceleration makes more sense.

As to speed, it seems like they would use the wheel speed signal from an undriven wheel (if available). A far off (missing) reading should invalidate the input. The mismatch to rear motor speed could be due to wheel slip, that would explain traction control turning on.

The 26% throttle input at 1.0s seems to line up with the changing in turning.

 

[brianman]

The polarity of that plot is clearly inverted. The car is decelarating after -1.6s.

The 26% throttle input at 1.0s seems to line up with the changing in turning.

Can you elaborate on this? I'm confused by the relationship between throttle and turning you're describing. Note that "rear motor speed (RPM)" is also different at that data point relative to the decrease shown in preceding and subsequent data points.

Thanks.

 

[Glamisduner]

This model 3 clearly crashed because of a model S issue that was not properly addressed.

 

[Daniel in SD]

Can you elaborate on this? I'm confused by the relationship between throttle and turning you're describing. Note that "rear motor speed (RPM)" is also different at that data point relative to the decrease shown in preceding and subsequent data points.

Thanks.

I think the theory is that maybe the "bad thing" that happened caused the driver to inadvertently tap the throttle at -1.0s. Hitting something large or the wheel buckling under the car would certainly do that.
It makes sense that the rear motor rpm would increase with throttle input of course.

 

[mongo]

Can you elaborate on this? I'm confused by the relationship between throttle and turning you're describing. Note that "rear motor speed (RPM)" is also different at that data point relative to the decrease shown in preceding and subsequent data points.

Thanks.

Here are the steering, lateral, and roll plots. You can see the inflection point around 1 second.

Steer wheel crosses from negative to positive.

Car switches direction of rotation

Car reverses sideways acceleration direction

And rolls to the other side

 

[Cirrus MS100D]

I'm confused how anybody is able to draw conclusions based on the complaint filed. Where is the police report? Where is the accident diagram? (Unless I missed it?) Showing a picture of skid marks and a gash in the road don't tell us anything without context. He was going nearly 60 MPH so traveled at least 100' from when "the bad thing" happened, plus whatever amount of time it took to slow down. Do they not do a police report if there were no injuries or other cars involved?

I used to watch rally cars religiously and while I realize the circumstances were probably quite different in this case, you don't have to be going THAT fast to hit something hard (like a rock) and rip the wheel off. I believe these cars might be a tad bit strengthened over a TM3 too:

(And yes, I realize the complaint says, "The police said he didn't hit anything!", which I guess is in the police report too?)

 

[AlanSubie4Life]

I don't think anybody here is drawing firm conclusions, really. I guess I think the wheel came off prior to the final collision, but beyond that I don't know. And I think I might be wrong.

FUD - Businesses insider "The star of 'Aladdin' claims a defect in his Tesla Model 3"

See the above post. It sounds like you have seen it, rereading your post...and it does provide SOME context. But true, no accident diagram, etc. Would be nice. I'm sure someone here can draw it up.

you don't have to be going THAT fast to hit something hard (like a rock) and rip the wheel off.

Probably true. Could have been a (large) branch or whatever. It depends I would think on what it is and what it hits. That's why video would be nice.

The police indeed said they didn't find anything (specifically that there was no pothole and no evidence that Massoud hit another vehicle). But that doesn't mean anything beyond that, really, since they didn't see it happen. It's just a piece of information.

 

[AlanSubie4Life]

Cool. I wonder if they include individual wheel speed data. I don't see it in the document. To me it just looks like lost of control from overcorrecting a skid. I'm no expert though.
He was going 55mph in a 35mph zone in rush hour traffic (it's always rush hour there). That doesn't seem safe!
Google Maps
Makes me suspicious...

I just looked at this link. Google even has new data from April 2019 so you can see the palm tree he hit. It seems to have survived. Seems like 55 might be too fast for this road.

And you can see the gouge in the road, it really is not far from the tree (assuming I have identified the right gouge, could not find any others). Not a lot of time for skidding in this accident. EDR says no ABS, so that skid on the road...wouldn't it have to be from a tire that no longer COULD rotate? Anyway, not surprised there wasn't a lot of skidding as he seems to have hit the tree at around 43mph...lucky guy.

Here's my angle on it - tree on the left, gouge in the middle (faded).

Google Maps

 

[Cirrus MS100D]

I’m totally speculating, but given that it was a brand new car, one presumes he may not have been THAT familiar with it. (My wife’s S feels like it’s 20 feet wide compared to my FIAT 500e that I wear like a tight shirt.)

What If the guy changed lanes a bit too quickly and just smacked the curb? Kinda explains everything, doesn’t it? The curb wouldn’t necessarily show any real damage if it hit the rubber. Looking at the Google Maps photos, the curb is low enough to clear the bumper but definitely high enough to destroy the wheel and rip it off.

 

[adam0785]

I’m totally speculating, but given that it was a brand new car, one presumes he may not have been THAT familiar with it. (My wife’s S feels like it’s 20 feet wide compared to my FIAT 500e that I wear like a tight shirt.)

What If the guy changed lanes a bit too quickly and just smacked the curb? Kinda explains everything, doesn’t it? The curb wouldn’t necessarily show any real damage if it hit the rubber. Looking at the Google Maps photos, the curb is low enough to clear the bumper but definitely high enough to destroy the wheel and rip it off.

Have you or your wife been involved in an accident like this when driving each other's car?

What about everyone that uses rental cars? Are they plowing these unfamiliar cars into trees? Keep in mind you can rent pickup trucks, sports cars, exotic cars....

Obviously the Aladdin actor is a short, FUDster, big oil shill, etc etc

 

[kengchang]

I mean this happened on 9/20/2018 right? I think Tesla already told him to pound sand after their own investigation that's why he's suing.

 

[DialMforMordor]

One thing to consider is that his lawyer probably has a financial motivation to file this suit even if he knows it's BS because he's making billable hours off it. I'm thinking the kid sees the separated wheel after the crash, and maybe he legitimately thinks it came off, who knows. The lawyer then gets the data and says "yeah kid, this data totally backs that up, let's sue" when really the data shows the exact opposite.

 

[AlanSubie4Life]

I’m totally speculating, but given that it was a brand new car, one presumes he may not have been THAT familiar with it. (My wife’s S feels like it’s 20 feet wide compared to my FIAT 500e that I wear like a tight shirt.)

What If the guy changed lanes a bit too quickly and just smacked the curb? Kinda explains everything, doesn’t it? The curb wouldn’t necessarily show any real damage if it hit the rubber. Looking at the Google Maps photos, the curb is low enough to clear the bumper but definitely high enough to destroy the wheel and rip it off.

Yeah, this seems possible, perhaps. Narrow enough road, and going fast enough that I wonder if he smacked the curb just hard enough to break a control arm but not hop the curb, just as he tried to straighten at -1.6s, then wheel came off at -1.0s.

I have no idea anymore. Tesla must have more data they can dig up to use to defend themselves...

It really is interesting to me how he never used the brake (service brake) until time 0 at the earliest. He had an entire 1.6 seconds once things started going south. Really did remarkably well overall given he hit a large tree at 43mph (sounds like curb hop was still negative time and did not trigger the event). Good that he did not hit it straight on.

 

[GolanB]

One thing to consider is that his lawyer probably has a financial motivation to file this suit even if he knows it's BS because he's making billable hours off it. I'm thinking the kid sees the separated wheel after the crash, and maybe he legitimately thinks it came off, who knows. The lawyer then gets the data and says "yeah kid, this data totally backs that up, let's sue" when really the data shows the exact opposite.

This case is quickly becoming a high profile case. Plenty of free advertising for the law firm.

The data will ultimately settle the case. Right now you’ve got two parties that believes the data shows Tesla is at fault, and one party that does not (Tesla).

Cases have been settled for years with the EDR data alone. Tesla May have additional data available to it, including video and sensor data which we’ve not yet seen which may provide new details.

This will be an interesting case to see!