In Model 3 Production Push, Tesla Adapts Vehicle Testing … Is It Unsafe?

AWD Tesla Model 3 being built in a tent


Due to multiple circumstances surrounding Tesla Model 3 production, the automaker changed a common testing procedure.

Almost all cars undergo a brake-and-roll test as part of the long list of procedures that must happen before the vehicle is deemed “ready.” Basically, the test is done within factory walls and meant to simulate real-world driving. It assures that the correct parameters are in place for accelerating, braking, the ABS system, etc. Tesla doesn’t seem to have the factory space or personnel to continue with such a test, so during the recent Model 3 production push, the test was changed.

One may immediately jump to the conclusion that Tesla is cutting corners and rushing cars out. We’re not here to say that this may or may not be true. However, we will report that – as many of you probably already know – Business Insider recently called Tesla out for avoiding this “critical test.” The publication states that CEO Elon Musk was ordering engineers around and demanded that they discontinue this important brake test on all Model 3 sedans. The story reports:

The test was apparently shut down before 3 a.m. on June 26, according to a person familiar with the matter. It’s unclear why this particular test was halted or for how long.

According to an industry expert, the brake-and-roll test is a critical part of the car manufacturing process, taking place during its final stages. The test ensures that the car’s wheels are perfectly aligned and checks the brakes and their function by taking the vehicle’s engine up to certain revolutions per minute and observing how they react on diagnostic machines.

According to CNET Roadshow, Business Insider may have gone a bit too far here. In fact, if every car can be driven on a track to gauge real-world performance, rather than put on a simulator inside the factory, this may even be an advantage. Sure, for most automakers, doing this test inside the factory just makes more sense and likely saves times and energy. But, this is not the case for Tesla.

Consider the fact that Tesla is building some of its cars in a “tent” due to the lack of room inside the factory. Also, the automaker is utilizing every extra employee it can in order to help build and prep these cars for delivery. Rather than trying to find additional factory space to run the traditional brake-and-roll test, Tesla not only relied on earlier tests that it explains are redundant, but also test drove every vehicle after it came off the assembly line. So, to say that the test was skipped or eliminated is not really the case. Instead, the test was adapted based on the situation.

Should people be concerned about this and other reported “corner-cutting” practices?

Of course, there’s always a concern that rushing through the building process to meets goals and appease stockholders and consumers could lead to catastrophic results. Early reports pointed to the Model 3’s subpar build quality, and that was during a time that production was at a snail’s pace. However, it appears that over time Tesla has improved and refined the production process and addressed many of the early issues.

In regards to the braking test, the automaker told Roadshow that the current on-track testing procedures are “identical” with the previous brake-and-roll tests. If this is true, current Model 3s should be no less safe than those that were tested in the traditional way.

Keep in mind that after Consumer Reports found issue with the Model 3 stopping distance, Tesla initiated an over-the-air update, which has solved the problem. We can’t guarantee that the Model 3 will not have any braking issues, just like we can’t guarantee that about any car, but it sure seems like this story was a premature attempt to shed more negative light on the automaker without researching the details. Check out Tesla’s description of its current Model 3 on-track testing process via Roadshow:

Tesla has confirmed to Roadshow that its brake-and-roll testing and on-track testing procedures are identical. Both tests begin by burnishing (or bedding-in) the brakes by accelerating to 60 miles per hour and then decelerating down to five mph. This is done three times in relatively quick succession and is fairly standard practice across the industry. Next, the vehicle’s anti-lock brake system is checked for functionality. This is done by running the vehicle up to speed and triggering the system by braking hard. If the ABS triggers, the vehicle is passed and if it doesn’t, it isn’t. Finally the vehicle is brought to 60 miles per hour and then coasted down to a stop — this measures drivetrain drag.

Tesla can’t really afford for the Model 3 to have critical problems – and we mean that literally. The automaker is pushing to become profitable and to prove to the world that it can pull this off. One can only hope that the production constraints and expectations don’t put the company in any position where a corner is cut, which eventually leads to turmoil. However, in the case of the adapted brake-and-roll test, it seems Telsa is covering all the right bases.

Source: CNET Roadshow, Business Insider

Categories: Tesla

Tags: , ,

Leave a Reply

36 Comments on "In Model 3 Production Push, Tesla Adapts Vehicle Testing … Is It Unsafe?"

newest oldest most voted

If they’re doing that track test for every car, it’s probably fine from a safety standpoint. Doing it on a dyno where a machine can exactly measure everything would be more accurate. That’s especially true for any alignment issues. But, not the end of the world either way.

The Model 3 is filled with sensors – I bet they use on board sensors to confirm performance.

I still can’t imagine that a bunch of 60-0 tests are faster or easier on a test track than on a dyno, especially once you factor in the amount of track space needed to test a high volume of vehicles.

Tesla does the alignment in an entirely different step, using an alignment machine. It is actually more accurate than trying to get alignment data from a “rolling road”. Checking the alignment again on the brake-and-roll machine actually provides less accurate information than they were already getting in previous steps before the brake-and-roll test.

Tesla actually goes well above and beyond what the vast majority of car makers do when it comes to alignment. For example, Wheel alignment check and adjustment is include in the maintenance plan as a standard service every 12,500 miles. Typically that is not at all included in most car makers scheduled maintenance, and is almost always a substantial upcharge.

Absolutely true… Also the machine records the data by Vin, so it is in that cars record in case there ever is a problem its easier to track down. A driver on a track cannot record precise data points, and the car does not have sensors accurate enough to record small variations.

Model 3’s know their own VIN, and can collect data and transfer it OTA back to the factory. As long as the sensors are within the required accuracy of the test parameters, there is no reason not to use the car’s own sensors. They certainly have plenty of data on braking onboard the vehicle.


So, in summary:

– Tesla has a history of braking issues; two guys from Consumer Reports had to find their problems for them, for example, by running extremely basic tests.
– Tesla performed this test up until the “crazy all-nighter cram” at the end of the quarter.
– All other automakers still perform this test. All other automakers produce higher quality cars than the 3 (mine’s been in for service twice now).
– At that time, Tesla began skipping this test for the 3, but still presumably performs it for the X and S.

In light of Tesla’s history of quality control failures, especially around brakes, removing additional tests does not seem wise, and it is mighty suspicious that they waited to throw them out the window until they had a crazy deadline. Similarly, I don’t trust Elon’s decision making over the engineers on the line as to what to cut. This is the fellow who has shown time and time again he is divorced from reality.

Rhetorical question: did Seven Electrics read the article?

Doesn’t matter. Eleventy Pretend Electrics has his FUD talking points which he’s going to copy and paste here, regardless of facts or truth or even relevance.

A troll’s gotta do what a troll’s gotta do! 🙄

“Eleventy” — my favorite made up number, along with “Eleventeen”.

Thirteen Electrics, In the middle of ramp-up is EXACTLY when manufacturers remove redundant tests. You start production with an over-abundance of tests, including duplicate tests.

Redundant duplicate testing early in the process confirms that each test itself is accurate, with one test confirming the results of the other test. Once testing accuracy is proven during the ramp-up, manufacturers remove redundancies to as part of their normal plan to speed up production.

They also decrease sampling at the same time. For example, if they were pulling every 100’th part for destructive testing to ensure quality, they would move to every 500th part.

Sorry your fake ownership of zero Tesla’s is damaging you ability to understand manufacturing controls.

(⌐■_■) Trollnonymous

“(mine’s been in for service twice now).”

I think that applies to every automotive manufacturer.
I can tell you my old Ford explorer needed to go in 4 times in 2 years for repairs (waranty covered) and one was a tranny replacement!!!

So by all your bull$h1t criteria, Tesla is on it’s way to becoming just like “All other automakers”……..LMAO!!!!”

For some odd reason, I doubt that TwentySeven Electrics has even driven a Model 3, let alone owns one.

Or any other Tesla car, either.

I personally do not know if Model 3’s are safer or not, when you skip the Brake and Roll test, but I am skeptical when normal procedures are changed at crunch time to meet a goal… Normally you set up a production and validation system and then follow it to a tee for consistency. In Tesla’s case an employee reported this change to the media as happening at 3AM leading up to a big deadline, so I am not sure if the proper scientific validation was done to make sure this test was redundant and not needed. I have no opinion about the change in testing, but think the timing, and circumstances of the change are sketchy.

Wrong. For ramp-up you start with redundant testing steps and higher sampling rates, and as ramp-up matures you remove redundant testing and reduce sampling rates.

You start with a lower production rate, then as you remove redundancies this ALLOWS ramp up to go faster!! The act of removing redundant tests is EXACTLY the time when manufacturing companies increase rates. All of this is planned out as part of the ramp up plan to happen exactly when it happened and to have these exact results.

Why do you keep endlessly trying to compare a product that you know is in the middle of ramp up, or factories that are in the middle of construction to double their size, to products and factories that have already gone through all of this in their own ramp-ups and build-outs long ago?

It’s also exactly HOW you increase ramp-up.

It’s very common to create a set of interlocking tests that check each other, and as you get in further into manufacturing you eliminate certain tests. In the case of multiple assembly lines you can eliminate *different* tests from the different lines as long as the coverage is the same. This allows you to cross-check when things *do* go awry.

How do I know this? I was a test engineer for many years.

ding ding ding! We have a winner!

That is straight out of Lean/Agile Manufacturing 101 for optimizing production. It goes straight to the heart of JB Straubel talking about designing “the machine that makes the machine” by reexamining every single element for optimization through first principles thinking.

The more old ICE companies blindly stick to what they “know” just because that is how they have always done it, without re-examining things like test coverage, the further they will fall behind.

You get paid by the Merchants of Doubt to be skeptical .

@David Green

Why don’t you just change your screen name to “Concern Troll” and be done with it?

If the decision to stop the brake and roll test was indeed due to Tesla miraculously finding the test truly isn’t needed and is redundant, that means that Tesla should never perform the brake and roll test again for any future produced cars and will drop it from this point forward. Assuming it wasn’t a last minute, middle of the night decision to increase production to hit the ~5k/week mark.

🙄 Obviously neither logic nor critical thinking are your forte.

The FUD is strong in this one!

If Tesla goes back to doing the brake and roll tests, it’ll be proof they dropped the test purely to juice their unsustainable production rates.

This is what passes for “logic” and “proof” among serial Tesla bashers. 🙄

Too bad you have obviously have absolutely no knowledge of how manufacturing works.

On the other hand, I’ve done test engineering in manufacturing for many years and can pretty much guarantee that you’re full of it.

And with bro8008’s post we have the Full House of anti-Tesla posters all checking in to this story.

Jokers over Jacks.

Why wouldn’t Tesla use whatever test they want to spot check whatever they want whenever they want? Testing redundancy is common in manufacturing. I wouldn’t be surprised at all if 1 out of 1000 cars get pulled and tested on the brake and roll machine, even well into the future.

Why would they moronically limit themselves artificially, just because YOU are short-sighted?

It is quite covenient that they determined the test wasn’t needed during the month they were up against a deadline to hit the ~5k/week production mark, isn’t it?

It is quite “coincidental” that you raise these concern troll questions about Tesla’s manufacturing, but not General Motors’ manufacturing.

Which company was it, again, which deliberately hid evidence of a safety-critical ignition switch failure which cost dozens of lives?

Hint: It wasn’t Tesla!

Clean Technia gave an astonishingly scathing report about the Tesla-bashing Business Insider article. I’m not used to seeing a respected online news source accuse another established online news source of FUD!

Put on your flame retardant suit and read for yourself:

From the same team that brought you the unsubstantiated hype about Tesla’s demise… From the creators of the overblown stories of each and every Tesla that ever caught on fire, may have caught on fire, looked like it could have caught on fire or was potentially, allegedly parked in a location where the sunshine could, on at least one planet in the universe, result in a fire… We bring you a story from Business Insider that frames up Tesla as a terrible company because it ditched the “brake and roll” test for the Model 3.

[continued below]

[continued from above]

The BI article comes with a set of leaked photos of Tesla’s production management system for one of the Model 3 general assembly lines. BI claims that Tesla has abandoned the “critical” brake and roll test for the Model 3, which “ensures the car is correctly aligned.” The FUD is strong with BI this week, as it takes a single fact and extrapolates it to the nth degree just to pull in a few clicks…

From Clean Technica: “Business Insider Resorts To FUD As Tesla Streamlines Model 3 General Assembly Lines — #Pravduh”

Ah yes because the legacy OEM’s that do this testing didn’t even notice they forgot to install brake pads on some cars….

I doubt they test these 3’s on a track. In light of the high repair rates and lack of spare parts and repair facilities in Norway as reported by Electrek,Tesla should be testing more,untill quality improves, not less. They have to fix things under warranty, and it hurts their reputation and future sales, it would cost them less to do these brake and roll tests, rather than skipping them, also cars that missed they tests will have greater liability in future court cases.

They have their own track right next to the factory. They run every single car they build around their test track, and through suspension test sections.

Pls quitwith the tesla bad tesla brilliant bs. Read the cnet roadshow article which is balanced and realistic. This is not skrimping on safety and its not a disruptive genius play. They need to track them for general checks because everything is not nailed down yet. Since they have to do that they skip the brake and roll. They will go back to brake and roll as per x & s when things are settled. Its a workaround from production hell.

Domenick Yoney said much the same over on the InsideEVs’ forum: That the CNet article seems to be fair and balanced coverage of the issue, and pretty thoroughly refutes Business Insider‘s Tesla bashing claims, which arguably go beyond merely bad reporting to outright FUD.

The CNet article is linked at the end of the article above, but here it is again:

From CNet Road Show: “Tesla changed its testing regimen for the Model 3, and that’s OK”