Possible outcomes of NHTSA investigation and Tesla response

[brianman]

Will you back that statement up by not bothering to have the retrofit applied to your vehicle?

I'm considering not installing it on my vehicle, yes. But that has nothing to do with my assertion that you quoted.

I'm not a fan of government or business agencies taking away flexibility from citizens or customers willy nilly.

 

[PeterJA]

When the battery-puncture fires occurred last year, there was lots of discussion at TMC about possible solutions. Several people suggested cowcatcher-like deflectors for road debris. But as I recall, no one suggested Tesla could build a shield that simply smashes debris to pieces before it reaches the battery. Several engineers read this forum, but none of them thought of that.

Tesla's ability to think outside the box, and invent surprising solutions that smash obstacle after obstacle, is one reason I am heavily invested in their stock.

 

[capt601]

Everyone should take some time to read the "Complaints" section on that NHTSA site for the Model S.
Apparently the "shorts" have nothing better to do than to comb threads on this website, then file reports on the NHTSA website. Not one report had a VIN number, I believe most of the reports are from non-owners, looking to damage TSLA

Edit: 2 of the 19 reports have a VIN

http://www-odi.nhtsa.dot.gov/owners/SearchResults.action?searchType=PROD&prodType=V&targetCategory=A&activeTab=1&searchCriteria.model_yr=2013&searchCriteria.make=TESLA&searchCriteria.model=MODEL%20S&searchCriteria.prod_ids=1839811

It sounds like the same person for each one. Lots of "allegedly". Really? And they keep these on and don't verify?

 

[smorgasbord]

It sounds like the same person for each one. Lots of "allegedly". Really? And they keep these on and don't verify?

The NHTSA process uses an on-line questionnaire from which information is extracted to create the reports you see. My guess is that the NHTSA inserts the word "allegedly" into complaints since they don't verify each complaint individually.

From http://www-odi.nhtsa.dot.gov/recalls/recallprocess.cfm

Information you provide on the questionnaire is entered into the NHTSA consumer complaint automated database, and catalogued according to vehicle make, model, model year, manufacturer, and the affected part, assembly, or system. These reports, with the consumer’s personal identifiers removed, are listed on www-odi.nhtsa.dot.gov/complaints and updated weekly. Citizen and consumer reports help NHTSA and manufacturers to determine if a safety recall is warranted, and also provide motorists with valuable information about potential safety problems currently under review.

Will I be contacted?

In some cases, an investigator from the Office of Defects Investigation (ODI) may call to clarify or verify information from your report. Unfortunately, the large volume of reports received by the agency does not permit a return call for each report filed. Questions about whether your concern involves an investigation or recall are best answered by contacting the DOT Vehicle Safety Hotline or by viewing our Web site.

NHTSA technical staff conducts a continuous analysis of these reports to determine whether an unusual number of complaints of potential safety-related problems have been received on any specific line of vehicles, tires, or equipment (e.g., child safety seats, jacks, trailer hitches, etc.). The number of reported complaints and the severity of the consequences are carefully reviewed by technical staff and measured against the number of vehicles (or items of equipment) manufactured, and how many years the vehicles or equipment have been in service.

This ongoing evaluation process allows NHTSA technical staff to determine whether complaints represent isolated reports or a trend. If a trend is suspected and a problem has a potential for causing a risk to safety, the agency will open an investigation for more detailed analysis of the problem.

How many reports must be filed before NHTSA investigates an issue?

There is no established number. Agency technical experts review each and every call, letter, and online report of an alleged safety problem filed with NHTSA. Although NHTSA has no jurisdiction over defects that are not safety-related, it does review each report that suggests a potential safety defect involving groups of motor vehicles or vehicle equipment.

 

[smorgasbord]

You are implying that NHTSA found a safety defect and negotiated with Tesla to correct it with the retrofit.

After more careful reading, that's not just what I'm implying, that's what I believe almost certainly happened. Let's look at the facts as we know them. There were two incidents, and the NHTSA covers them separately. No-one else is looking at the statement that way, but it's very instructive when you do.

First, the NHTSA investigation process includes frequent contact with the manufacturer:

3. Investigations:

Investigations are conducted in two phases: the Preliminary Evaluation and the Engineering Analysis.

Preliminary Evaluation (PE)
Most PEs are opened on the basis of information submitted by DAD, but they may be opened on the basis of other information as well. During the PE phase, ODI obtains information from the manufacturer (including, but not limited to, data on complaints, crashes, injuries, warranty claims, modifications, and part sales) and determines whether further analysis is warranted. At this stage, the manufacturer has an opportunity to present its views regarding the alleged defect. PEs are generally resolved within four months from the date they are opened. They are either closed on the basis that further investigation is not warranted, or because the manufacturer has decided to conduct a recall. In the event that ODI believes further analysis is warranted, the PE is upgraded to an Engineering Analysis.

Engineering Analysis (EA)
During an EA, ODI conducts a more detailed and complete analysis of the character and scope of the alleged defect. The EA builds on information collected during the PE and supplements it with appropriate inspections, tests, surveys, and additional information obtained from the manufacturer and suppliers. ODI attempts to resolve all EAs within one year from the date they are opened, but some complex investigations require more time. At the conclusion of the EA, the investigation may be closed if the manufacturer has notified the agency that it will conduct a safety recall or if the agency has not identified a safety-related defect. However, if ODI believes that the data developed indicates that a safety-related defect exists, the ODI investigator prepares a briefing to be presented to a panel of experts from throughout the agency for peer review. If the agency panel concurs with ODI’s recommendation that a recall should be conducted, ODI notifies the manufacturer of the panel’s concurrence and may, if appropriate, provide a final opportunity for the manufacturer to present new analysis or data. ODI then sends a Recall Request Letter to the manufacturer.

OK, so the process includes discussions with the manufacturer. Let's look at this specific case. The following paragraph from the NHTSA statement says alot:

Tesla performed a series of tests reconstructing this incident and determined that a similar shaped object contacting the forward edge of the HVB could be "tripped" and potentially penetrate the HVB case. As the object's opposite end digs into the pavement, vehicle momentum causes the object to impart upward force into the case, described by Tesla as a "piking effect". Tesla's testing reproduced damage similar to that seen in the Tennessee incident, and also showed that a change in ride height strategy, which was implemented in Nov. 2013 via a telematic software update to prevent the SVs lowering at legal roadway speeds, mitigates the risk of battery compartment penetration when a three-ball hitch is struck.

1. The NHTSA knew about the tests Tesla conducted before Tesla published some of them in their blog (NHTSA statement came out first).
2. The NHTSA agreed that the change in ride height already conducted (5.8 I believe) mitigated the fire risk.

So, there are movies we haven't seen that show that running over the trailer hitch with a low ride height will indeed damage the battery enough to cause a fire. And there are movies we haven't seen that show that increasing the ride height reduces or eliminates that damage.

Those are facts. The logical conclusion from the wording is that since Tesla already issued what was effectively a recall that reduces the risk to acceptable levels, the NHTSA did not feel it necessary to insist on making what was already done a new official recall. You may come to a different conclusion.


Let's look at the investigation into the second incident:

The object struck in the Washington incident was not identified. More severe damage to the incident vehicle and the unknown shape of the object raised concerns regarding the effectiveness of raising the ride height for objects other than a three-ball hitch. In a Mar. 10, 2014 meeting with ODI, Tesla stated it would conduct a free-of-charge service campaign to modify the SVs by adding three new components to the vehicle's undercarriage to protect the HVB. A low-hanging transverse member and an additional underbody plate would be mounted forward of the HVB case and a third plate would overlap the leading edge of the case. Testing conducted by Tesla demonstrated that these modifications improved protection from debris impacts.

The NHTSA saw the results of testing performed by Tesla before those were made public, so again they were in communication. Matter of fact, they conducted a meeting on March 10! And during that meeting Tesla said it would conduct a free-of-charge service campaign to modify the SVs. And, Tesla showed the results of their testing that these worked.

Neither of us were at the meeting between between Tesla and the NHTSA, but it seems obvious to me that what was happening can logically be described as a negotiation. Maybe the NHTSA said they would move the investigation into the EA phase. Maybe Tesla was simply being proactive in conducting the EA for the NHTSA, and providing a solution, in the hope that the NHTSA would not characterize the result as a recall. We weren't there, so we can't say for sure, but it seems far more likely that the discussion was more of a negotiation than a "well, here's what we've done, now we'll shut up so you can decide what you want to" kind of thing.


Now, time for the way-back machine. Remember back in Jan when Elon said that their over the air update and swapping out of 14-50 adapters was not a recall because "no Tesla vehicles are physically being recalled" () or Jerome Guillen saying: "calling Tesla’s fix a “recall” was outdated because it did represent a physical repair but rather a software update conducted wirelessly while the owners’ cars were parked in their garages. “We are going to be having some discussions about the appropriateness of using a word that has no physical sense,” Guillen said.

And now we have physical vehicles coming in to be modified and somehow that's not a recall? Well, turns out the NHTSA isn't making a unique exception for Tesla. As this Bloomberg story says:

Earlier this year, the agency and Chrysler Group LLC clashed over fixing 2.7 million Jeep Grand Cherokees for gasoline tanks that can leak in a crash. The agency says 51 people have died in post-crash fires.

Chrysler initially challenged a government-requested recall, which would have set up a rare public hearing and court case, before agreeing to a “voluntary campaign” to provide more protection in rear-end collisions.

So, whether the terminology is "voluntary campaign" or "retro-fit", the NHTSA only cares that the manufacturer is doing something that appropriately reduces the risk without charging the customer directly for the fix.

Kudos to Tesla for getting in front of the situation, especially after Musk himself said that the car did not need to modified.

 

[gpetti]

Tesla deploying titanium shield.

This is the most extreme of possible solutions Tesla could have done and shows that NHTSA did determine there was an elevated risk of fire and Tesla was forced to respond.

I had hoped that Tesla would have be able to get away with a minor software change such as raising the car. A titanium shield is the most expensive solution other than changing the battery chemistry which would have been a death knell.

This should alleviate the risk.

Interesting that the stock is not moving this morning. Recalling all existing cars and retrofitting with titanium should be relatively expensive.

No, the most extreme of possible solutions would have been mandating the battery be relocated which would have destroyed the car as it's whole design revolves around the battery. I'm not sure if tesla had been warned of any negative outcome but clearly they have been working on this seemingly ingenious design for a while. It seems more like they have preempted the NTHSA's findings and as a result, no further analysis was required. My only concern is how much weight this adds; however, the batteries could get lighter in the future and having this weight low down in the car is part of its great handling.

 

[brianman]

When the battery-puncture fires occurred last year, there was lots of discussion at TMC about possible solutions. Several people suggested cowcatcher-like deflectors for road debris. But as I recall, no one suggested Tesla could build a shield that simply smashes debris to pieces before it reaches the battery. Several engineers read this forum, but none of them thought of that.

This is incorrect. Options like this were discussed. In fact, some of us were talking (unseriously) about lasers.

 

[smorgasbord]

There have been no fires since the fluke two impalements (I discount the Mexico high speed drunk driving accident as it was willful negligence), despite the fact there have been more miles driven by Model S since the fires than before. This to me bolsters the stastical safety argument. But that's not enough for some people.

You're forgetting that Tesla increased the ride height after those. And the NHTSA statement specifically said that increasing the ride height reduced the risk of fires from hitting something like the trailer ball & hitch.

But, the NHTSA also said that increasing the ride height wasn't enough:

The object struck in the Washington incident was not identified. More severe damage to the incident vehicle and the unknown shape of the object raised concerns regarding the effectiveness of raising the ride height for objects other than a three-ball hitch.

That's why Tesla is doing the voluntary retrofit. As noted in my previous response, this appears to be typical: as long as the manufacturer is making the change for everyone free, the NHTSA doesn't seem to care what label is used to characterize the change.

Tesla said they conducted over 150 tests. We've seen movies from only 3 of those tests. The NHTSA refers to Tesla tests that reproduced the fire damage from the Tenn. incident with the 3 ball hitch. Anyone not preforming the retrofit on their car and nevertheless running at a low ride height is taking on an unnecessary risk.

And, it'll be interesting to see if Tesla actually lets you get away with them not retrofitting your car next time it's in for service. My guess is they'll practically insist on it.

 

[kenliles]

Anyone not preforming the retrofit on their car and nevertheless running at a low ride height is taking on an unnecessary risk.

While true- I'd argue it's unnecessary due to Tesla's offer. The actual risk however (unnecessary or not) is less the risk taken everyday by every ICE car on the road. you might argue, given the lack of comparative miles we don't actually know that- and I'll give you that. But it works both ways- we don't actually know, so it's not a given that there's more risk being taken by a ModS (unmodified low rider) vs any existing ICE driver.

 

[PeterJA]

Anyone not preforming the retrofit on their car and nevertheless running at a low ride height is taking on an unnecessary risk.

Risk of what?

I repeat: There was never any significant risk to human safety (because of the Model S's multiple safety systems). There was a minor risk of inconvenient damage to the car (which can now be eliminated with the retrofit).

Your very long posts obscure that essential difference, which you called "splitting hairs." A car is not a person. Risk to a car is not the same as risk to a person.

 

[Auzie]

While true- I'd argue it's unnecessary due to Tesla's offer. The actual risk however (unnecessary or not) is less the risk taken everyday by every ICE car on the road. you might argue, given the lack of comparative miles we don't actually know that- and I'll give you that. But it works both ways- we don't actually know, so it's not a given that there's more risk being taken by a ModS (unmodified low rider) vs any existing ICE driver.

Most people, especially people that frequent TMC, would agree with your evaluation of comparative risk taken by ice drivers vs Mod S drivers.

NHTSA are likely to evaluate each specific risk that they are investigating 'on its own merits', not in comparison to other cars. I would be surprised if a body like NHTSA decides to 'let a car off the hook' for a risk that is highly specific to that car (battery impact by road debris) just because it is safer than other cars that do not have that specific risk. Tesla's response of mitigating their specific risk is outstanding.

 

[kenliles]

NHTSA are likely to evaluate each specific risk that they are investigating 'on its own merits', not in comparison to other cars. I would be surprised if a body like NHTSA decides to 'let a car off the hook' for a risk that is highly specific to that car (battery impact by road debris) just because it is safer than other cars...

I think that's true. And consistent with their actions in this case as well. A customer makes a more wholistic decision than NHTSA, as it should be. My personal decision would be to not modify if there was a cost involved. I would see it as unnecessary. If I were Tesla I'd tie it to their warranties (fire and lease value programs), and suggest it to insurers, all to help force idiots like me to get it retroed.

 

[smorgasbord]

I repeat: There was never any significant risk to human safety (because of the Model S's multiple safety systems). There was a minor risk of inconvenient damage to the car (which can now be eliminated with the retrofit).

Your very long posts obscure that essential difference, which you called "splitting hairs." A car is not a person. Risk to a car is not the same as risk to a person.

The fact is that a risk of the car burning up is a risk to people inside or near the car. Suppose a disabled person was inside Model S - is there no risk that they couldn't get out in time, or if they could get out that they could get far enough away in time? What about children in child seats? Could the driver unbuckle and carry 4 children away from the scene in time?

I'm sorry you're having trouble reading more than a few paragraphs, as your take on my "very long posts" is quite incorrect.

Finally, despite his statement that no modification would be necessary, Musk did what I originally suggested - say that even the safest car can be made safer.

 

[RABaby]

The fact is that a risk of the car burning up is a risk to people inside or near the car. Suppose a disabled person was inside Model S - is there no risk that they couldn't get out in time, or if they could get out that they could get far enough away in time? What about children in child seats? Could the driver unbuckle and carry 4 children away from the scene in time?

This modification is more about perception than safety. The cost in money or performance is insignificant compared to the benefit the company earns in the eyes of everyone who envies the car, and the virtual elimination of concern about fire risks. No matter what is done, there will continue to be criticism by everyone who has an interest in seeing Tesla fail.

The risk of the car burning up with people inside has not been altered by this change. Elon stated in his letter:

"The onboard computer warned the occupants to exit the vehicles, which they did well before any fire was noticeable. However, even if the occupants had remained in the vehicle and the fire department had not arrived, they would still have been safely protected by the steel and ceramic firewall between the battery pack and the passenger compartment."

Tesla Adds Titanium Underbody Shield and Aluminum Deflector Plates to Model S | Blog | Tesla Motors

So whether there were disabled or young children as occupants who were incapable of walking out on their own and were forced to sit inside and watch the vehicle burn, I have to accept Elon's word that they would still be safe. And if that is true, then there is no risk to human safety, and risk to safety would be the reason for a recall. What Elon did was to come up with a simple, elegant solution that enhances value to owners and shareholders without the need to qualify any previous statement he has made about safety. Since the modification is optional, I don't see how it can be considered safety related.

I will certainly have the modification performed because I would prefer to damage debris rather than be inconvenienced when out on a trip. That is the more significant enhancement. Elon has made the safest car on the road more durable.

 

[kenliles]

The fact is that a risk of the car burning up is a risk to people inside or near the car....
Musk did what I originally suggested - say that even the safest car can be made safer.

But aren't those facts always true,
before, now and forever?
Even now the ModS can be made much safer than it is even with the plate mod. Especially for the examples mentioned (kids, disabled etc.). I think the harder equation is safety relative to risk and function. Otherwise I think we'll just go in circular discussion. And If I'm going to be dizzy anyway, I'd rather it be from the Cabernet I'm sitting next to (fortunately, safely while not in one of those flamin' ModSs- hang on, just checking my lounge chair for one of them new-fangled titanium plates under my ass.... nope could be safer. Think I'll have another pour)

 

[Auzie]

And If I'm going to be dizzy anyway, I'd rather it be from the Cabernet I'm sitting next to (fortunately, safely while not in one of those flamin' ModSs- hang on, just checking my lounge chair for one of them new-fangled titanium plates under my ass.... nope could be safer. Think I'll have another pour)

Good that market is closed now. It might be an idea to keep that Cabernet safely locked up during market hours.:wink:

 

[kenliles]

Good that market is closed now. It might be an idea to keep that Cabernet safely locked up during market hours.:wink:

Now that's a good safety tip. Will do !

 

[PeterJA]

Since the modification is optional, I don't see how it can be considered safety related.... Elon has made the safest car on the road more durable.

Exactly.

Maybe it's a small point, but I don't like to see people sneering that Elon reversed himself (with the implication that he can't be trusted) when he really didn't.

 

[smorgasbord]

Maybe it's a small point, but I don't like to see people sneering that Elon reversed himself (with the implication that he can't be trusted) when he really didn't.

You infer something I did not imply.

But, let's face the facts: Elon did reverse himself. I'm glad he did - he should not have said what he did at the time. If he had adopted this path from beginning, which I did suggest at the time (not that he'd listen to me), Tesla would have been better off. Instead, Tesla has been in damage control mode for the past few months and he admitted it did affect demand. I'm glad it looks like things are going to work out, but let's not sugar coat the truth.

- - - Updated - - -

Since the modification is optional, I don't see how it can be considered safety related.

What makes you think it's optional? It's performed "upon request or as part of a normally scheduled service."

 

[RABaby]

You infer something I did not imply.


What makes you think it's optional? It's performed "upon request or as part of a normally scheduled service."

The Model S has no normally scheduled service. If you recall, the $600 service is not mandatory and the warranty remains intact. So without normal service, it isn't required.