Automotive engineer and former Tesla employee [SuperfastMatt] takes at look at the notorious Tesla door handle design and how it’s changed over the years (see the video below the break). The original handle design consisted of many moving parts, switches and wires which were prone to failure. Strictly speaking, the door handle is located on the outside of the car’s interior. While it’s sheltered from direct exposure to the elements, it still experiences the extremes of temperature, humidity, and condensation. The handles were so prone to failure that a cottage industry sprang up to provide improved parts and replacements.
Tesla made various improvements over the years, culminating in the latest version which [Matt] reviews in this video. Nearly all the failure points have been eliminated, and the only moving parts, other than the handle itself, is a magnetic sensor to detect handle motion (previously this was sensed by microswitches). [Matt] indelicately opens up the control module, and discovers an NXP programmable angle sensor ( KMA215 ). This all-in-one sensor detects the angle of a magnetic field, and reports it over an automotive communications bus that’s become more and more common over the last ten years: Single Edge Nibble Transmission (SENT) aka SAE J2716. SENT is a low-cost, transmit-only protocol designed for sensors to send data to the ECU. Check out [Matt] decoding it on the oscilloscope and Raspberry Pi in the video — it looks pretty simple at first glance.
We agree with [Matt]’s conclusion that the door handle design has been significantly improved with this latest iteration, questions of whether one needs a retracting door handle aside. If you’d like to learn more about SENT, here is a tutorial written by IDT (now Renasas) applications engineer Tim White. This isn’t [Matt]’s first encounter with a Tesla door handle — back in 2012 we covered his project which used one to dispense beer. Thanks to [JohnU] for sending in this tip.
Liked the article & video, but kept replaying the transformer near the end as my inner boy wants one…NOW
Thats real nice, but considering each of those fethers is 2k retail iirc, could i please have a 8k cheaper car?
That’s not how that works.
Why? Because you don’t think car manufacturers would read this and respond? Or because it’s not “profitable” to sell cheaper, less complex machines?
Either way you can’t change reality that income inequality is on the rise and it’s reducing the number of potential car buyers.
Tesla managed to find ways to offer cheaper models in China, I’m sure they could do the same elsewhere if so compelled by the local governments.
“Or because it’s not “profitable” to sell cheaper, less complex machines?”
No, because car manufacturers don’t charge *themselves* for parts the same amount they charge customers. Markup on replacement parts is *ludicrous*.
Also, keep in mind the original post was talking about an “8K cheaper” vehicle. That’s like, a 20+% price reduction on a $35K car. Tesla’s margins are nice, but they’re not *that* nice.
All that being said, I won’t buy a Tesla literally *because* of the original poster’s point – massive markup on replacement parts means these cars won’t ever drop to “reasonable” values on the resale market. To me, the only way you can justify spending $30K+ on a vehicle is if it will last *at least* 20 years, and without cheap replacement parts, that’s just not possible.
“Tesla managed to find ways to offer cheaper models in China,”
Yeah, because consumer demand is different in China than the US, just like there are a bajillion tiny vehicles in Europe that you can’t buy in the US.
You’re making the fallacious assumption that market trends are the same as consumer demand. Literally after potential consumers voice their demands in this here comment section, you retort that you can’t just make demands like that. Well I don’t but that, if nobody is going to personally interview me to ask what I want in a car then they shouldn’t expect me to fall in love with whatever they produce. I’m allowed to hate it and tell them that it’s shit, it’s my opinion.
Tesla’s choices as a company do not reflect what consumers literally ask for, and suggesting that they can’t come around no matter what because it’s all baked into modern manufacturing just makes Tesla look like blithering morons that can’t handle change.
It’s not my problem to sort out Tesla’s issues with profitability. If they want my business they should build a car for me, and not solely the cars that Elon wants to drive.
Again: an 8K reduction is over a 20% reduction. Changing door handles won’t do that.
I can go and bark outside of SpaceX’s HQ that we demand warp drive, but they can’t provide the impossible, no matter how much the demand is. If you think literally *anyone* could provide that cheap an electric vehicle of comparable size… good luck.
One aspect left unresolved in the video is how far DC motor drivers have come as he wonders whether a tampered magnetic encoder would cause the motor to wreak havoc on the mechanism.
A glance at e.g. the Ti DRV8876 datasheet gives a really interesting insight that these chips are no longer mere H-bridges, integrating a set of features for protecting themselves, the mechanics and the user.
“Integrated current sensing allows for the driver to regulate the motor current during start up and high load events. A current limit can be set with an adjustable external voltage reference. Additionally, the devices provide an output current proportional to the motor load current. This can be used to detect motor stall or change in load conditions. The integrated current sensing uses an internal current mirror architecture, removing the need for a large power shunt resistor, saving board area and reducing system cost.”
Current sensing works well when there is a limited amount of gearing. We will have to see if the current sensor triggers too early with gearbox wear lubricant failures.
Or you need more intelligence than a fixed limit. E.g. you can adjust the limit slowly during normal operation when the sensor shows smooth motion at plausible speed.
Power windows have already quite sophisticated algorithms to avoid pinching limbs although body parts can be quite soft and some power is needed to lift the window at high speed (wind pressure). But if the battery was disconnected (or deep discharged) you have to hold the switch until it goes to the end position to re-learn the characteristics.
Why on hell would you complicate design? So much can go wrong in engine bay already.
Look up Trabant, you take a few components out from the engine and it still runs.
If you take out the entire Trabant engine, it pretty much runs better than before!
(Makes it easier to push to the mechanic)
B^)
my dad had a Trabant! He imported on in Canada and drove it daily for years!
Isn’t it amazing that nowadays people think a overengineered, failureprone, expensive and arguably not-up-to-the-task electronic door handle is more desirable than a cheap mechanical, lever-activated door handle which will probably last forever, and, in case of an untimely demise, is fixable on the side of the road with a paperclip and a piece of string?
Yepp. Lot’s of engineering goes into “designs” which look nice but are far from usable or let’s say friendy for human usage…
Look at all those touch UI’s in cars. Next to no physical buttons any more = no “blind” use of functions. Everything is hidden under a layer and another layer of complication, because it’s hip… Or on stove tops or other home appliances. I don’t want 1950’s tech back, but I want to USE stuff with my hands and fingers and have feedback.
Maybe because an electric door handle would fit nicely on a luxury electric vehicle?
Tesla Model S is not your tin can with wheels (or mine), so it have to have something unique. And it’s an iterative design. 10 years ago the handles were terrible, now they are good, maybe in 5 years they are great. There’s no easy way to jump from nothing to perfect, but it’s possible to crawl all the way from nothing to crappy to passable to good to great.
After reading the stories of people being trapped in burning Tesla’s and people outside the car can’t grab the door handle to rescue them (because they won’t retract if they’ve lost power or they think the car is running) – sorry this is just something I can’t see for the long term for cars…. It’s pretty, it’s cool, but it’s not safe, reliable or cost effective, I don’t see any long term plusses here….
Well. If that’s your information sources. You need to consider branching out beyond you oil soaked media.
This exact sentiment has been raised for every single automotive innovation ever. “It’s too complicated” “it costs too much” from windshield wipers to automatic choke to automatic transmission to cruise control to ABS brakes to CD changers to air bags to catalytic converters, it’s the same old hue and cry every time. You need a new argument.
This article reminds me of Bill Gate’s quip “If GM had kept up with technology like the computer industry has, we would all be driving $25.00 cars that got 1,000 miles to the gallon.”
This video seems to confirm my suspicion about Tesla cars being unreliable because they are designed by college grads with zero practical experience. Looks like each model year of the S is a science project kluge made with 3D printed PLA plastic parts.
DC motors and door latching mechanisms are hardly new. Putting them together isn’t new either.
This is just overly complicated crap thrown together to please the easily impressed
Now the question is, are there ultimate limitations to reliability, or does it all come down to engineering?
I’d play devil’s advocate and claim that it’s not the electronic door handle that’s the impediment to field repairs, but the lack of fallback mechanisms that cause a situation to become a no-fix. With them, there would be no need to complain.
Meh. You could also say that about power vs. crank windows but nobody want crank windows anymore as power is the standard.
Don’t get me started on those stupid, slow, heavy, motorised tailgates on modern vehicles! The ones that panic if they even _think_ there’s a flimsy bit of grocery bag or minor obstruction and refuse to close. Give me a simple mechanical latch and a gas strut.
“The handles were so prone to failure that a cottage industry sprang up to provide improved parts and replacements.”
Oh, and EXPENSIVE if you asked the dealer to fix it, no doubt.
You misspelled ‘replace’ there
They can keep polishing that dogturd another decade, and it will still be a dogturd.
The idea was never good, and it was poorly implemented with an overcomplicated design, it’s still a bad idea not to have a mechanical connection between the handle and the locking mechanism in case of an emergency.
Because mechanical mechanisms always behave perfectly, they never jam or corrode or lose alignment.
“poorly implemented with an overcomplicated design, it’s still a bad idea not to have a mechanical connection between the handle and the locking mechanism in case of an emergency.”
This.
In industry, we use lots of smart instrumentation but always have manual back up devices.
I love how the door handle functions but I’m 95% sure that if I had one in front of me I could simplify it. I’d try purely as i’d love them on my car!
As an aside, I had a Tesla for a day and hated driving it.
It’s a marketing product.
We all agree, there’s no need for motorised digital door handles, but differentiators are a big selling point to weak buyers.
Just remember that phrase ‘technology for technology’s sake’… or ‘form over function’.
The good news is that these innovations will level out over the years, and somebody will develop a product that is great, works and is affordable. Evolution.
I suppose there is a point where luxury gives a little extra leeway in terms of justifying the cost to value ratio, perhaps a little personal flair points in there as well. I’m a little disappointed in the fact that door handles are still an issue as far as durability goes, considering the hype behind the company in question, also considering I can go to a junkyard and yank on a 25 year old cars door handle, and it still works, on an Oldsmobile no less, which is where something like a door handle should be by now when thinking about durability. Leave the stuff that wears out to windshield wipers, cheap chrome plated plastic and such. I don’t know, I’m pretty sure we can collectively engineer better ‘stuff’ if we really wanted to. I would say if these things hadn’t had a history of costing customers unnecessary grievance and loss of income, it might not be as big a deal. The car should enter the junkyard with functional door handles at the least. Are these door handles more to replace than the cost of that over engineered juicer that made the news some time ago? Something else to consider I suppose.