That's laughably bad of Tesla - the thumb wide gaps in the body, the gluing rubber parts on glass seals because the original one was way out of tolerance, door rattle, one door having two ways to open but not the others - the screen that goes you have activated alternate door opening mechanism or some such thing - all pretty embarrassing. Never mind the insane difficulty to open the hood in an accident which looks like a real safety issue.
Also who in the 2018 world designs trunk doors that would take two hands to close for kids, smaller/older individuals? The non power operated trunk on my CR-V closes with a finger!
I do not understand how there is no manual mechanism for opening the rear doors from the inside, how does that even happen?
His comments about the outside door handles were at least prefaced with his own physical limitation but the state of that car is mind boggling. This is on top of the known promised features and oversight with regards to the programming of the display. I did notice on reddit there is a story of a III having cratered its engine. http://teslaweekly.com/first-model-3-motor-failure-reported-...
All in all 2017 wasn't their best year so hopefully 2018 will see a turn around but even max projections are well under 200k cars this year
>manual mechanism for opening the rear doors from the inside
Brother of my close friend literally burned to death in a Hyundai with no manual override for unlocking/opening doors. Granted he had modified the car's stereo system with custom speakers and the roadside mechanic did not do a good job with it which caused short circuit/fire. But had he been able to manually open the doors he would still be alive today. And to think Tesla is all batteries that have strong chance of starting a fire - ugh.
If car body construction tech was new you could cut Tesla some slack but given all of these issues have been fixed by pretty much all manufacturers years ago - I have no sympathy for Tesla there.
Yeah window winders too - they should have a knob that drops the window down someplace not accidentally accessible. That should be easy enough to do without adding any significant bulk.
They're extremely cheap and some of them also come with a blade for cutting seatbelts (inside a notch so you won't accidentally cut yourself with it). Definitely a life-saving tool worth keeping in your car.
Not a bad idea, but are those prongs hard enough? I thought window-breaking tools have a ceramic or tungsten (?) point on them to start the glass fracture.
I'm not going to test it, but I imagine myself desperately bouncing the metal prongs off the inside of a window.
Right, but the batteries are more vulnerable to starting a fire in collisions than gasoline. I think there were cases of Tesla car batteries starting fire on impact.
Gasoline still requires an ignition source to catch fire. Lithium cells can spontaneously ignite when they experience physical damage.
I'm not saying electric cars are safer or more dangerous in the event of a crash (because I don't know the answer to that), just that batteries can and will catch fire.
No, but it's fairly common knowledge that li-ion batteries are at fire risk on overheating and collision/impact(ref: Tesla fires). That's not to say Tesla didn't firewall the car enough to prevent fires from reaching in the cabin but it's still not a good idea to be locked in there when batteries are burning below.
It's pretty well known that gasoline is a fire risk... that's kind of how ICE engines work. Lithium-ion batteries come in a variety of different chemistries and some, e.g. LiFePo4, are not much of a fire risk at all.
So, yes, if you puncture the lithium cells you may put yourself at risk. You're also at risk if you puncture a gasoline tank (worse if you're talking something like hydrogen).
Note: Tesla does not, AFAIK, use LiFePo4 cells. Additionally one of the big challenges is that LiFePo4 cells are less energy dense than other lithium chemistries.
I agree, but if you are really safety minded you should have a metal pick/club to break the windows anyway. Can't rely on doors being operational during accident and can't wait for emergency response. Of course I don't have one :]
After a roll-over, it won't be there and you may not be able to reach the glove box. If you have a center console, I recommend keeping it there (that's where I keep mine).
If you're reading this and you don't have one in your car, buy one now. Here is a link for 2 for $10 (give the other one to your spouse/parent/kid/someone): https://www.amazon.com/dp/B01CQN0DDC/
Either place will find the tool potentially flying thru the cabin upon an accident. Strap it down. Same goes for fire extinguishers, jacks, and tools kept in the cabin. If your escape tool knovks you unconscious in the wreck, it does you no good.
Mine has velcro to keep it in place. Will it work in a roll-over? No idea, but yeah, definitely need it to be with in hand's reach for it to be effective.
Regarding that engine problem, crib death of complicated to manufacture systems is an expected phenomenon, though I'm not sure how car manufacturers have mostly sorted it out
That's a car where failure to open the hood quickly will not cause a delay in sawing through a door to save a life. That's the urgency with the Tesla, really: rescuers need to cut the cable that is inside.
I don't follow the car industry but I did happen to catch this 2015 video of Munro talking about the BMW i3. He says things like "this is the watershed," "they invented new science," "this is the Model T of our time," etc. I got a whole new respect for BMW after watching that video. Munro even followed the production chain to Morocco while investigating the costing!
That was a very informative video. My respect for the CEO shot up quickly. He wasn't really trying to sell anything but simply illustrating points of concern.
I can tell he very much enjoys doing what he does.
I know this is a place for software/web developers but good god, really? It was at best a minor tech article in a car magazine in the Wal-Mart magazine stands.
What exactly did you learn?
He kept saying the i3 "life module" (agile! AGILE!) was some magical achievement. Yeah, carbon fiber is neat. They also choose the grain direction of panels for optimum efficiency.
- We've been using grain direction on wood for what, thousands of years?
- We've been made cars "strong" for century. It's called a tube frame. And it's one of the main reasons dragsters/NASCAR drivers don't die on impacts that will kill most people in consumer cars.
What's actually important these days is not how "strong" a frame is. It's how it deforms to absorb maximum energy on impact so the passengers experience the least acceleration possible. "Strong" has been solved. And crush areas have been common for decades upon decades. The only new thing here is they're using carbon fiber. And the only reason they're doing that... is 1) "marketing" and 2) weight reduction because battery cars are very heavy AND very sensitive to weight for mileage.
Automotive companies aren't on the frontline of technology. They're on the frontline of CHEAP/mass-producible technology.
Simply putting some carbon fiber on a body is not magical to us practicing mechanical engineers. But I guess it may seem like "magic" to software people, the way mechanical people see IT/software as magic.
What's actually "Cool" about this company/video is how they can tear down entire vehicles and come up with accurate cost estimates, as well as the construction order. That takes a _lot_ of work and industry experience. But the technology in those cars? It's not even remotely new.
If these kinds of problems and design issues are still present when I'm ready to buy a 3, I'd probably reconsider.
Maybe they'll tighten their tolerances as they perfect their line, but some of those design decisions seem almost beyond explanation-- the double door handles (where one option shames the user for doing what should be the normal MO), the absence of mechanical releases in the back and in the frunk, and so on.
For me, in both of these videos, the audio is severely mangled (across different browsers). Have never experienced anything like this on youtube before.
Does anyone have any idea why this might be happening?
Edit: solved, some strange driver issue it seems.
BMW i3 has the worst designed interior of any car i’ve ever been in and the road noise is insane. I’d take a few annoyances that come with the first run of Tesla’s startup manufacturing over the i3 any damn day.
Is it that bad? The wheels look like ultra thin elastic band wheels which would transmit much less road noise than normal fat wheels, so is it just a lack of sound proofing in it?
I've seen similar comments everywhere I stumbled across this story. It's an easy takeaway, but that's not how automakers look at competitive benchmarking. Working with a company like Munro is an engineering decision. Munro breaks cars down, reverse engineers everything they possibly can, and draws as many conclusions about production methods as possible. Engineers with other automakers use this data to learn from how others approach problems.
The engineers don't give a damn about a Munro PR video about a competitor, period. It's a complete non-issue for them, and the same goes for their managers who are the ones who decide to work with Munro. They want to know every detail about how all the sausage is made (or what the sausage actually is). They don't care if the car is well made or poorly made; they're focused on the how; after all, even a car with severe quality control issues can offer useful insights. Besides, every peculiar (or even dumb) design or manufacturing decision they've made was previously laid bare by companies like Munro. Think of the process as something more like peer review or a morbidity and mortality conference for doctors. It's not a public process, despite something like this video.
Basically, the people who would care if Munro upset the company by speaking favorably about Tesla are the people who have no involvement in whether their company works with Munro in the first place. If someone from Ford marketing told Ford engineering to quit working with Munro because of a video, they'd be laughed at and promptly ignored.
The stuff that Munro is pointing out is so fundamental that I don't see how there's bias in it. The craziness that there is some hidden circle underneath the front hatch that you have to know to open and pull out a couple of leads to attach to a 12V battery is maddening. The fire department would have to do this to be able to cut the cable that's in the front hatch.
I can see there being a Model 3 recall over that kind of stuff. What's the point of over the air software updates if your hardware is unsafe.
The battery disconnect procedures there, using the high voltage service disconnects, start with lowering the back seat or jacking up the rear of the car.
Tesla provides the location of the little access panel on the quick response sheet here:
If you compare it to having a lever by the driver as in any other car, that's 2 seconds to do vs them having to search on the Internet / find documents in the car then find the hole then connect up a battery.
The car could be upside down or on a very steep hill. It all seems pretty frightening stuff. You're basically just praying you don't have a serious crash which is back to the 60s.
I don't think that emergency procedures should be matched to other electric vehicles they should be matched to existing ICE cars.
What’s up with the fixation on connecting to 12V? Don’t you need to DISconnect and not the 12V,but the high voltage one?
That’s not a difficult procedure: open the hood (forcefully if need be, latch is in the front), cut the wire in the place marked in bright orange and labeled accordingly - and immediately visible and accessible upon opening the hood.
I don't think Munro is in that business. Understanding how something is built is different from how to consistently and price effectively build it to specs/tolerances.
The counterfactual to this argument is that his clients would be much more likely to purchase his products if there was something to learn from a competitor like Tesla. They buy his services and reports to learn. If Tesla is vastly superior, and Munro understands how, that is critical to the Big three. Knowing or having some small boutique industry firm take pot shots at Tesla doesn't get them anything. This isn't being reported on NPR and CNN.
In reality, I think this has almost no impact on his business other than a little bit of publicity. Well run companies on the scale of GM are going to be just as interested in the data on a competitor whether their competitor is better or worse. There is always something to be learned. Ford is going to learn different things from Hyundai and from Mercedes...but they are going to learn.
> Plus, the engineers told me, the cost models are proven. Some manufactures, they said, actually hand their own parts to Munro asking them to use their model to see how close they can get to the actual cost of manufacture. Munro says their numbers always come very close, and that they’ve even been accused of having insider information. Munro isn’t just there to help engineers learn about how the competition designs and assembles its cars, it’s also there to help automakers—who really should be called “auto assemblers”—figure out if they’re getting ripped off by their parts suppliers. Knowing exactly how much it costs to build a wheel bearing, for example, can act as leverage for negotiation against a supplier of that part. At the same time, suppliers also go to Munro to learn. If, for example, one of their competitors is selling a part for what seems like impossibly too little, the supplier might ask Munro to tear it down and figure out how the competitor is saving cost.
I suspected that as well - but any hints of evidence?
To be fair though, Tesla's manufacturing quality does indeed falling behind their targeted segment majorly. Like Munro said, even Kia makes better assembly these days (without uneven gaps etc.) But yes, have a paid contract for product promotion doesn't hurt
From the title I was hoping they'd actually, well, "tear into" the Model 3. Instead I got seven minutes of them complaining about the door handles. They did raise a couple of legitimate concerns (dual ways to open the doors, no manual/mechanical way to open the rear doors, body fit and finish seemed poor) but it was hard to pick them out from all the "hurr durr never seen that before" and associated whinging about things that were probably on page one of the user manual.
I don't think "hurr durr never seen that before" is an accurate paraphrase; I'd say it was more along the lines of "I've never seen bodywork this shoddy on a modern luxury vehicle."
I would not doubt that the build quality of the Tesla is lacking in areas that are cosmetic. Ive seen numerous Model S and Model X vehicles with atrocious panel gaps and things that people coming from the likes of Audi, BMW, MB, and Lexus would notice. That said, it doesnt seem to be affecting their safety, as there is a pretty significant amount of crash data from tesla vehicles that shows that they are among the safest vehicles in crash tests.
With all due respect did you not watch the video? Much of the focus was on safety during an accident.
The 3 they were looking at had two points to cut the electrical system, both of which were hard to get to during an accident, and the rear doors had no mechanical backup at all. If power goes out rear passengers need to make their way to the front seats or trunk.
Maybe watch the actual video rather than imagining what the video was to say and then responding to that.
I did see the video, and I know the criticisms about the electrical system. I think that is a very real scenario that should be addressed. But not all crashes will cause fire and loss of power in this car. There have already been Model 3's that crashed, and the occupants were not hurt even after a head on collision with a utility pole that totaled the car. That was my point. It isn't a safety issue that the cars have panel gap issues and they look like beta versions. Saying that the car is built like a 90s Kia would imply that it is hazardous to drive the car and that it is unsafe at any speed. Its hyperbole. Its grandstanding, its not reality.
Im a fan of Tesla, I wish the best for them. They arent perfect, there is a ton of room for improvement, but it is largely cosmetic and not safety critical.
As a semi related aside, when I was younger I thought that 4 door vehicles were insanely ugly compared to their 2 door counterparts. Now that Im older I think about being pinned in the rear of a 2 door car after being in an accident and I instantly feel claustrophobic. On the other hand, the fewer holes you put in a car, the more rigid the structure.
My first car was a two door Honda hatchback, made in 1992, and when you pulled off the plastic panel over one of the rear seat speakers, underneath the speaker bracket was simply the sheet metal of the body. Literally nothing more than a cookie sheet between you and the outside world.
I guess I don't understand the surprise about this?
In the 1970s my father worked at a company that supplied parts to car manufacturers, and it was a normal thing - if Company X say they want a backlit switch with the clicky feel of a Company Y Model Z, they hired a Company Y Model Z, took the switch off and measured it.
According to my father, the car industry and the car rental industry all know this is commonplace.
This is the sort of thing that makes my eyes roll and my attention wander whenever politicos hammer on about the loss of our manufacturing base being America's imminent downfall.
The amount of institutional knowledge baked into our industries is simply mind-boggling. The only reason we can get away with outsourcing manufacturing in the first place is because this knowledge is so amazingly thorough.
We have transferred a TON of this institutional knowledge to China. Now companies like DJI can compete and beat us, and it’s all with knowledge that we transferred to them in the last 30 years.
DJI hires high quality Chinese engineers. Their software is better than what American consumer drone companies manage to push out, and they can push the latest tech to market insanely quick, as GoPro learned the hard way when they found out DJI was a half generation ahead of them on drone tech.
DJI has great customer service, an amazing insurance plan that provides piece of mind, and bullet proof build quality.
They do have a major advantage of being in China, so they can work with suppliers much more directly. It doesn't feel like they are using the grey technology market though. A lot of the lower priced drones are obviously put together from Shenzen part bins, but DJI just feels good all around.
That said, their Android mobile app is a bit janky at times. Then again, lots of stuff on Android is a bit janky...
That's exactly the parent's contention. Where did all the infrastructure, money, and knowledge that led to a world-class country emerging there come from?
Bootstrapping a tech economy gets a lot easier when people hand you the keys to the kingdom in exchange for short-term profits, without even requiring equal access to your domestic market in return.
This is patently untrue - DJI has some of the worst customer service I have ever experienced, especially for a 1000$ device. I spent two months calling every other day trying to resolve a hardware issue and ended up issuing a chargeback before getting any attention.
They're great for a toy or a hobbyist, but companies or anyone who relies on the drone tech should look elsewhere. I can relate to awful customer support and even strange software problems that are refused to exist by DJI.
Fair enough, I've seen mixed reports. I have the care program, and everything I've ever seen about that has consisted of people saying they got their drone replaced quickly and without fuss.
To be fair, from what I understand the no fly zone restrictions thing actually does require geolocation lookup and I think those no fly zones can change, so pinging a server and checking in makes sense.
From just a pure telemetry perspective, knowing where people are flying the drones at is useful for product development and marketing.
(In response, DJI did release a privacy mode that doesn't need internet.)
A sizable % of websites visited want geo-location data, which, at times, gets sold on the open market!
Then there are background phone apps that grab location data. Even when rate limited, once every half an hour is plenty enough to figure out where Military Secret is at.
> To be fair, from what I understand the no fly zone restrictions thing actually does require geolocation lookup and I think those no fly zones can change, so pinging a server and checking in makes sense.
Not exactly. Issue a map file with the no-fly zones as shapes (as easy as possible to save on space) and embed it with a creation timestamp. Force an update every 30 days - have the app download the file and upload it to the drone. Manipulating the drone time won't work as it gets its time directly from GPS.
Also these patch files should not clock in too much - the entirety of Europe's streets can be squeezed into ~3 GB (download size of full Navigon EU map set)... I'd assume a 512MB flash chip would be sufficient to store all the world's no fly zones.
A lot of that transfer was explicitly driven by government policy.
If you're ever around the Beijing airport, you'll notice a mushroom patch of well-known non-Chinese companies' facilities. The one constant? More security around the buildings than a military base.
I'm not one to beat the US / European nationalistic gong, but it's a fact that the Chinese government has specifically targeted foreign industries to force into tech transfer. And foreign governments generally haven't done anything meaningful about it.
Manufacturing is a large industry, but it's what China was targeting in the 1950s. They're aiming higher now. Most of the tech transfer is around next gen industries, such as batteries.
And as for who is forcing foreign companies to do this? Hilariously enough, capitalism. Multinationals can't afford to miss selling to the Chinese market, or their board / shareholder's vote them out. So they're left to strike the best deal they can get, given that the outcome is already known by both parties. (Exceptions like Google exist, but it does have an atypical stock structure)
"China’s quid pro quo policy also helps account for the unusual global investment flow. Chinese policy requires multinational firms that want access to China’s huge market to transfer partial property rights to their technology, often through R&D centers or joint ventures with Chinese firms (Holmes, McGrattan and Prescott 2015a, 2015b)."
> Then why didn't the multinationals come to share technology and set up manufacturing in India?
Because the indian government didn't demand it as a condition of entry like the chinese government did.
Have you never wondered why western companies have to engage in 50/50 ventures with chinese companies into china instead of just entering china like they do every other country?
Look up Hero Honda, Maruti Suzuki etc. What happened was that the MNCs weren't interested at all. India followed with liberalization and removal of such "license raj" in 1991. Economy has done much better since.
But the PRC can prevent them from selling in China. Shareholders care about the quarterly results. Nobody cares about the company a few years on.
The Chinese have a long view on this stuff. If you put out an engineering RFP, you can be sure that Chinese companies and government are probably the biggest web consumers of your content.
Munro & Associates wants to sell these automobile tear-down reports to China so much that they even translated their marketing materials into Chinese :)
With more and more functionality moving into software, is Munro or anyone else doing anything along these lines with the firmware and software in vehicles? Are manufacturers as interested in evaluating the process and design of their software as they are with the physical pieces of the vehicle? Do copyright and sometimes murky legalities around reverse engineering software prevent the equivalent teardown (disassembly) and analysis of software components?
Despite being known for kaizen processes implemented as Plan -> Do -> Check -> Act or 5 Whys, Toyota's issues with unintended acceleration were partly blamed on flaws and spaghetti code in the RTOS controlling their Electronic Throttle Control System in some models:
That's just one example. There's the CAN bus car hacking via OnStar and more examples out there. Would this kind of analysis help more than say, simply implementing better software development methodologies?
I'm amazed they sell all that info for just $150,000. I figure there's around ~125 major automakers in the world, maybe a dozen or two or so big enough to own their own labs. For each 23,000 page document, we're only taking ~15M, max. For a team of manufacturing experts, this just isn't that much cash.
I suspect a majority of their cash flow comes from those training seminars, and the $150K book is just to get you in the door and interested.
They don't just sell to major auto makers. Governments are a huge customer. OEM suppliers are another big one, they don't buy the comprehensive reports, just component reports.
I heard the owner/chief engineer give a talk a while ago, specifically about the i3, and he said the report was originally priced at $500k for the full thing. So it's possible that the price of these reports drop with age.
Edit: they also do perform costing on a contract basis. So an OEM will actually pay them to tear down a competitor vehicle, and Munro reserves the rights to sell that report too.
This is why the EU made disassembling software a right for certain use cases, because it's standard practice in all other industries, it is unreasonable to ban it in software.
Car companies use third parties to decap and analyse chips before negotiating the buying price. Source: I was the one buying the IT app to manage the results of the analysis.
I wonder if they acknowledge any responsibility if their tear-downs reveal recallable design defects? Not that I'd want them to be persecuted, but they'd be in a great position to notice that something may be under-spec'd for a given workload.
The article also didn't mention any reverse engineering of the software - I'm sure someone could clone the physical parts of an i3 and have a vastly substandard product because without a thorough understanding of the software...
> I wonder if they acknowledge any responsibility if their tear-downs reveal recallable design defects?
Why should anyone be held responsible to find a flaw in a 3rd party product and not report it? It might be considered unethical if a life-threatening issue, yes, but IMHO there is no way to attach legal responsibility.
I developed new appreciation for car manufacturing from this article. The way that one component depends on other common components reminded me of NPM packages.
I'm wondering what if they found one car infringing another car's design or engineering IP without fulfilling a license, or if they would find certain car makers "reinvent the wheel" simultaneously. A lot of car makers need to properly license certain design from other car makers. I wonder if this also helps facilitating and defending these licensable designs, or encouraging open sourcing design in cars.
I suspect the overall outcome of finding similar design and borrow existing design from one car maker into another is a good thing for bringing the cost down.
Most car components actually do not come from the car manufacturer. For example: Bosch (which is famous for its drilling machines) is also a common manufacturer of engine control units and other parts and got flak for it during the diesel scandal, Knorr Bremse is an industry supplier for brakes on trucks and buses and Takata a supplier of airbags.
It's kinda frustrating that reverse engineering in the physical space is not protected (as evidenced here), but stuff like the DMCA tends to do a lot more for protecting software. Seems ridiculous to me. Perhaps this is why these guys aren't also reverse engineering the control systems (as suggested by many comments, I'm not sure it is true)
Should not the government require manufacturers to extensively document all the parts and materials that go into a product? In this age, we should demand that products are easily (re/up/down)cycle-able, and this means third parties should be able to efficiently dismantle products.
If the regulatory burden for starting a new car company was that high, you would have never seen Tesla, which means you would never see any pressure on the big guys to go all electric. So you need to balance regulation with innovation.
All of that information is required knowledge for the business to operate and production to continue. You can't make something if you don't even know what materials you need to make it, nor if you don't know how to put it together.
In the case of extremely low production with hand-fitting the documented process may vary some from the actual process but it shouldn't be so significant that one product does not even resemble another.
What is the regulatory burden in forcing car manufacturers to publish the same manuals and offer the same tools for purchase they already have for their own service centers?
Besides many countries do require automobile mass market vendors to offer spare parts and docs on FRAND conditions to "unconnected" repair shops already.
Agreed. But for a product like a car, it would be fair to comply much later after-the-fact; it's not like the car is driven straight into the junkyard ;). For products like food (packaging) it seems different though.
what about software? I'm reminded of what farmers have to go through with John Deere tractors [https://www.youtube.com/watch?v=F8JCh0owT4w]. Isn't that going to start happening with cars (I thought this was already happening). Are they going to reverse engineer that too?
It's very interesting to look at the number and complexity of parts in even an extremely old and thus much simpler car, then contrast it to cars today with at least 2 orders of magnitude more; this one for 1929-32 Chevrolets, for example, is already over 300 pages:
I'd have loved to have seen some footage of them actually dismantling stuff, to see how they do it without damaging items, special techniques they use etc.
I did work experience for Rover many years ago and they had a department that took competitor's cars apart to see how they were built and how much they could build the cars for. Some cars were selling for much more than their assembly and parts cost.
I was under the impression all car MFGs did their own tear-downs as well to ensure or at least gauge how easy/difficult it is to recycle a car. The car "green" effort back in the mid 90s.
Munro on the BMW i3: https://www.youtube.com/watch?v=rqiBWfsDTAA
Munro on the Tesla Model 3: https://www.youtube.com/watch?v=QCIo8e12sBM