zlacker

[parent] [thread] 51 comments
1. mirkul+(OP)[view] [source] 2023-10-23 22:09:39
I've worked for a large OEM, dealing with a large Japanese megacorp that is not Mazda for about two years (actually Mazda was one of our customers too, but I didn't get to work with them directly). This does not amaze me anymore.

We spent months agonizing over an interior temperature sensor, which was only used to display the information to the user on a smartphone app. We built both the hardware and software, and it was offered as an add-on at the dealerships. After months of negotiations, after the hardware was already built and the packages assembles, they decided temperature sensors were too inaccurate (+/- 5 degrees F) to use, and that it could present a legal liability. Again, this was nothing else but displaying the information on the app - and the user could then make a decision whether to remote start the car to cool it or heat it (no automatic process took place either).

This was at the height of "unintended accelerator" issue in Toyotas, so everyone was walking on egg shells playing it ultra safe to not invite any more lawsuits.

What surprises me is that this culture of "playing it safe" remained to this day, some 10 years later (but maybe it shouldn't).

replies(9): >>johnma+z2 >>hirund+Z2 >>mardif+f5 >>jdjdjd+B5 >>Aeolun+97 >>Affric+Yb >>wutwut+ld >>glimsh+Pa1 >>kayfox+Vg4
2. johnma+z2[view] [source] 2023-10-23 22:24:47
>>mirkul+(OP)
I wonder if they were concerned someone would use the data to make a decision about leaving a living creature in the vehicle.
replies(2): >>PvtGle+E4 >>mirkul+wJ2
3. hirund+Z2[view] [source] 2023-10-23 22:26:36
>>mirkul+(OP)
My cheap home thermostat has that frustrating +/- 5 degrees F accuracy. Is it very difficult to build an inexpensive 1 degree sensor?
replies(3): >>harris+m6 >>vGPU+P6 >>jacque+kb
◧◩
4. PvtGle+E4[view] [source] [discussion] 2023-10-23 22:36:57
>>johnma+z2
or medications or food. I can totally understand the legal liability angle 5° F is not a small margin of error.
5. mardif+f5[view] [source] 2023-10-23 22:41:04
>>mirkul+(OP)
Woah I'd have guessed that temperature sensors would be more accurate than that! Is it just an issue of cost, or are most affordable temp sensors that inaccurate and I've never realised it? That would explain a lot though!
replies(3): >>AlotOf+z6 >>junon+C6 >>throwa+a7
6. jdjdjd+B5[view] [source] 2023-10-23 22:43:16
>>mirkul+(OP)
It should remain, at least for mega corps
◧◩
7. harris+m6[view] [source] [discussion] 2023-10-23 22:48:23
>>hirund+Z2
I am curious what an "expensive" one would actually cost, too... It is a car so already a large purchase. I'd pay a bit more for an accurate thermostat.
replies(1): >>mirkul+XH2
◧◩
8. AlotOf+z6[view] [source] [discussion] 2023-10-23 22:50:05
>>mardif+f5
No, it's pretty easy to get better sensors than that. E.g. a cheap-ass SHT20 will do +-0.3C. In fact, my own automotive parts recommendations are the next grades up (SHT21/SHT3x) as standard for my employer's boards because the cost difference is justified.

Never underestimate the ability of a manufacturer to select subpar parts to save 25 cents on the BOM and spend 6 figures elsewhere trying to fix the resulting issues though.

replies(2): >>throwa+n7 >>_nalpl+D31
◧◩
9. junon+C6[view] [source] [discussion] 2023-10-23 22:50:21
>>mardif+f5
Depends on the sensor, depends on its calibration (or its lack of such functionality). Often a function of cost and/or size, as well as the means by which it measures temp.

Scientific sensors are highly accurate and can also be small, but you have a steep cost increase of course.

◧◩
10. vGPU+P6[view] [source] [discussion] 2023-10-23 22:51:09
>>hirund+Z2
Apparently +/- 2 degrees is fairly common.

One of the problems is the heat from the device itself, as well as limited airflow creating localized hotspots.

11. Aeolun+97[view] [source] 2023-10-23 22:53:18
>>mirkul+(OP)
Literally all of corporate Japan is built on playing it safe. Any innovation that comes out of it seems more like a happy little accident.
◧◩
12. throwa+a7[view] [source] [discussion] 2023-10-23 22:53:39
>>mardif+f5
If you just buy a bare sensor, yes, it's going to be +/-5. They also have a non-linear response which needs to be dealt with as well.

If you are only concerned about a 20 or so degree temperature range it's not an issue, but if you are trying to read over a 100 degree range, you'll want to account for non-linearity as well.

Also, accurate and precise to 10ths of a degree isn't really attainable unless you do fancy math as the sensor will heat each time you read it. The idea is to take multiple readings and average them but unless you are accounting for the heating of the sensor, your numbers will be garbage.

This is for consumer grade sub $50 sensors. Of course you can go fancier but you have to pay for it.

replies(2): >>maxeri+y8 >>Charon+bn
◧◩◪
13. throwa+n7[view] [source] [discussion] 2023-10-23 22:55:25
>>AlotOf+z6
No, you read the datasheet wrong. It's +/- 3C.
replies(1): >>AlotOf+38
◧◩◪◨
14. AlotOf+38[view] [source] [discussion] 2023-10-23 22:59:58
>>throwa+n7
I'm looking at the datasheet right now. Relative humidity accuracy tolerance is +-3.0, temperature is +-0.3. Different tables.
replies(1): >>chris_+0f
◧◩◪
15. maxeri+y8[view] [source] [discussion] 2023-10-23 23:04:45
>>throwa+a7
This isn't something I know anything about, but I know that 1-Wire exists and so on so am able to locate something like https://www.analog.com/en/products/max30207.html pretty easily. $2 in quantity, reports a temperature digitally, accurate to 0.3 C between 0 and 70 C.

What is it about that device (or similar) that would put it out of scope?

◧◩
16. jacque+kb[view] [source] [discussion] 2023-10-23 23:24:47
>>hirund+Z2
It is if you don't want a calibration step. If you do calibrate then it's no longer an inexpensive part...
17. Affric+Yb[view] [source] 2023-10-23 23:29:08
>>mirkul+(OP)
5 Rankine is 2.55555555... Kelvin for anyone wondering which is there or thereabouts in the range of an average air conditioner.
replies(1): >>rainbo+Bi
18. wutwut+ld[view] [source] 2023-10-23 23:39:27
>>mirkul+(OP)
Idk about everyone else but when it comes to anything running in my car, _anything_, there is no such thing as excessive "playing it safe". It's a 2 ton mass of steel barreling down the highway at 70+ mph next to other unpredictable 2 ton masses, please for the love of God, fight to maintain that culture of "playing it safe", regardless of what you're working on and for what purpose.
replies(4): >>ipaddr+5j >>lost_t+5D >>kazina+SM >>mirkul+nG2
◧◩◪◨⬒
19. chris_+0f[view] [source] [discussion] 2023-10-23 23:52:39
>>AlotOf+38
I imagine the temperature gradient across the car might be +/- 3degC, ignoring the actual sensor.
replies(3): >>HankB9+Dg >>mardif+vk >>AlotOf+fm
◧◩◪◨⬒⬓
20. HankB9+Dg[view] [source] [discussion] 2023-10-24 00:04:56
>>chris_+0f
Measuring temperature is not trivial. There's convective, conductive and radiative heat transfer and they all factor into the measurement. And that may not accurately reflect the "cabin temperature," particularly in a parked car.
◧◩
21. rainbo+Bi[view] [source] [discussion] 2023-10-24 00:20:11
>>Affric+Yb
Henceforth I will be making my fill of liquid helium in my car.

/s

replies(1): >>Affric+3B
◧◩
22. ipaddr+5j[view] [source] [discussion] 2023-10-24 00:23:43
>>wutwut+ld
Would you be willing to pay millions for your car to make it safer? The pope has bullet proof glass, different body materials can protect your life. How would you define excessive?
replies(3): >>gottor+Yj >>Spivak+Xk >>wutwut+dp
◧◩◪
23. gottor+Yj[view] [source] [discussion] 2023-10-24 00:30:46
>>ipaddr+5j
Yeah, everything is a tradeoff that should have a coherent risk-benefit analysis attached to it. "X at all costs" isn't realistic.
replies(1): >>wutwut+op
◧◩◪◨⬒⬓
24. mardif+vk[view] [source] [discussion] 2023-10-24 00:33:56
>>chris_+0f
But the weird part is that OEM seem to be fine with that variation when it comes to climate control. Do they use multiple temp sensors?
◧◩◪
25. Spivak+Xk[view] [source] [discussion] 2023-10-24 00:37:25
>>ipaddr+5j
Yes, I just don't want to pay all of it myself. I just want to pay the marginal cost of making it. Let the automaker invest said millions (really billions over all the individual components) into the design and manufacture.
◧◩◪◨⬒⬓
26. AlotOf+fm[view] [source] [discussion] 2023-10-24 00:48:52
>>chris_+0f
It's undoubtedly more than that, depending on where you measure. The gradient between e.g. the roof lining and the AC vent could easily be 20C+ degrees on a hot day. Most boards on a vehicle will have their own temperature sensors to measure enclosure temps, and there will be zonal sensors at various points in the cabin as well. The climate control loop will be defined in terms of those zonal sensors.
◧◩◪
27. Charon+bn[view] [source] [discussion] 2023-10-24 00:56:56
>>throwa+a7
It's probably not that the sensor is bad. It's the location of the sensor that is tucked away in the interior foam so that it's not reading the air inside the cabin
◧◩◪
28. wutwut+dp[view] [source] [discussion] 2023-10-24 01:16:09
>>ipaddr+5j
R&D is already baked into the final price of everything we buy so that has no argument.

And the bullet proof glass thing I shouldn't even respond to because of the ridiculous extreme you've had to go to, trying to argue against me saying the companies should play it safe, but I'll reply this one time. I'm not asking the car company to protect me from an assassin's bullets. That is not something they control. I'm asking them to "play it safe" when developing components for the car so the car doesn't kill me while I'm in the car. They are responsible for their domain and are not producing armored vehicles for war time. So ridiculous lol

replies(2): >>shiroi+zt >>ipaddr+JI
◧◩◪◨
29. wutwut+op[view] [source] [discussion] 2023-10-24 01:18:18
>>gottor+Yj
Never said "X at all costs" but thanks for trying to speak for me. Going forward, please note that my preference is to speak for myself, as should you.
replies(1): >>gottor+6w
◧◩◪◨
30. shiroi+zt[view] [source] [discussion] 2023-10-24 01:59:16
>>wutwut+dp
If you live in the US, getting shot while driving (or at school, or at work) at is something that happens from time to time, so if you value safety, you really do need armor and bullet-resistant glass on your car to drive in America.
replies(1): >>wutwut+Gy2
◧◩◪◨⬒
31. gottor+6w[view] [source] [discussion] 2023-10-24 02:27:23
>>wutwut+op
I'm not sure why you took such offense; it's a reasonable interpretation of the words you spoke for yourself:

> there is no such thing as excessive "playing it safe"

As you noted in your other comment:

> I'm asking them to "play it safe" when developing components for the car so the car doesn't kill me while I'm in the car.

As in the old adage in computing ("the only unhackable computer is one that isn't connected to anything"), there's no way to ensure that the components of a car don't fail, even while in routine use. There is only more or less likely that they won't fail, and of course, less and less likely to fail is more and more expensive.

We might say that the only uncrashable car is one that sits in the garage and never goes anywhere. Obviously, that would be playing it safe excessively, since it would defeat the purpose of having a car to begin with. But what about less obvious cases? Toyota recalled millions of cars for their "unintended acceleration" issue. The merits of that particular case aside, how much more would someone pay for a Corolla that would be progressively less likely to have safety issues? At some point before infinity, it would be considered excessive.

I think the sliding scale of how safe is playing it too safe is a discussion very much worth having.

◧◩◪
32. Affric+3B[view] [source] [discussion] 2023-10-24 03:26:03
>>rainbo+Bi
lol… it’s also a hint that google will evaluate the reported conversion as one of pure scaling without any of the transformation that occurs in converting 5 Fahrenheit to 5 Celsius.
◧◩
33. lost_t+5D[view] [source] [discussion] 2023-10-24 03:50:05
>>wutwut+ld
It should always be tempered with common sense. The older I get the more I just want a basic car. I just need gauges and a radio with bluetooth. I don't need to integrate everything into my car or have an 18" center console. A reliable car (electric or gas), decent interior space, with no surveillance and just basic features and I'll walk into your dealorship and pay cash. Closest I've gotten so far is my base model corolla with 4G antenna disabled (after market).
◧◩◪◨
34. ipaddr+JI[view] [source] [discussion] 2023-10-24 05:14:56
>>wutwut+dp
Your comment was nothing is too excessive. The truth is, everything has a level where we try to balance cost/safety. Having Mazda spend millions more puts the base stickier price up. It might be $100, $500, $5000, $50,000 $5,000,000. How much more are you willing to pay and if you really cared wouldn't you buy a Volvo over a Mazda?
replies(1): >>xctr94+SO
◧◩
35. kazina+SM[view] [source] [discussion] 2023-10-24 06:05:59
>>wutwut+ld
> It's a 2 ton mass of steel barreling down the highway at 70+ mph

So good thing it's connected to the internet and has four screens.

replies(1): >>wutwut+bU
◧◩◪◨⬒
36. xctr94+SO[view] [source] [discussion] 2023-10-24 06:30:04
>>ipaddr+JI
This really isn’t helping. Cars are very safe at the moment, with the driver being the key factor in accidents. They are very safe at the current price point. GP was arguing we should keep fighting to keep them safe. That means we keep doing whatever it is we’re doing, which is making cars safe, at a reasonable cost.
replies(1): >>kakwa_+Bj1
◧◩◪
37. wutwut+bU[view] [source] [discussion] 2023-10-24 07:21:47
>>kazina+SM
> good thing it's connected to the internet and has four screens

Not any of my cars that I've owned

◧◩◪
38. _nalpl+D31[view] [source] [discussion] 2023-10-24 08:41:22
>>AlotOf+z6
I think the problem is not the accuracy of the sensors themselves but that its difficult to have a placement inside the car such that the measurement is minimally influenced. Many factors influence temperature: heat from the motor, from heating and from the sun shining on the car and perhaps others.

And it also depends what exactly you want to measure: air, motor or inside temperature? People might get confused. And inside temperature might differ a lot: behind the windscreen it might be a lot hotter than at the floor.

39. glimsh+Pa1[view] [source] 2023-10-24 09:47:17
>>mirkul+(OP)
I'll be honest here... +/- 5F isn't useful to me as an add-on temperature sensor, that would have pissed me off as a consumer.
◧◩◪◨⬒⬓
40. kakwa_+Bj1[view] [source] [discussion] 2023-10-24 11:07:52
>>xctr94+SO
Are cars really that safe?

I mean, safer relative to what they used to be, yes.

But compared other modes of transportation, not so sure.

replies(2): >>xctr94+fP1 >>wutwut+Xx2
◧◩◪◨⬒⬓⬔
41. xctr94+fP1[view] [source] [discussion] 2023-10-24 14:00:30
>>kakwa_+Bj1
I’m guessing it’s arguable, but in the top 25 causes of accidents (in the US), I only found 2 that are linked to the car itself. Cars themselves seem fairly safe.
replies(1): >>kakwa_+CM2
◧◩◪◨⬒⬓⬔
42. wutwut+Xx2[view] [source] [discussion] 2023-10-24 17:07:34
>>kakwa_+Bj1
People roll their cars on the highway and walk away these days. That wasn't a thing that happened in the 60s, 70s, 80s, and even 90s. Some people don't survive, but a vast majority do. You're instantly surrounded by airbags from all sides these days and have a sort of forcefield to absorb the energy of the crash so you don't have to. The same concept has safely dropped rovers onto other planets. Engines nosedive downward into the ground instead of back into the firewall and then into your legs, crumple zones take the impact and absorb the energy instead of transferring it through the solid steel bumper, solid steel frame, solid steal dashboard, solid steal steering column, and solid steel steering where, where your skull is next in line to absorb that energy which up to that point has hardly dissipated. In the 60s a fender bender often messed up people's necks for life, yet people today can often flip their car and walk away with scratches, never to complain about any life long issues stemming from the accident. Automatic collision detechion systems can notice a stopped object and apply the brakes faster than our meat cpus can even process the eyeball input and notice what is going on, and then dealing with the latency of brain to muscle signals and muscle speed and accuracy. When you're about to hit a brick wall at highway speeds, 250ms more of brakes on full can shed an insane amount of speed/momentum/energy. And let's not forget about all the people who text and drive who would rear end or cross the lane and hit someone head on if it wasn't for collision detection stopping them or lane keep yanking the car back into the lane it should be in. Cars are safer than they have ever been.
◧◩◪◨⬒
43. wutwut+Gy2[view] [source] [discussion] 2023-10-24 17:10:35
>>shiroi+zt
So edgy
replies(1): >>shiroi+064
◧◩
44. mirkul+nG2[view] [source] [discussion] 2023-10-24 17:44:06
>>wutwut+ld
Just further proves my point. You should probably then take out any floor mats, cup holders, temperature controls, radio, or anything else that could potentially impede, obstruct, or otherwise distract from driving safely.

This comment was meant for the normal folks who spend a lot of time in our vehicles and are willing to accept a level of risk that comes along with having some sense of comfort.

replies(1): >>wutwut+fV2
◧◩◪
45. mirkul+XH2[view] [source] [discussion] 2023-10-24 17:50:41
>>harris+m6
It's not just material cost, probably different interface to the sensor so factor in some R&D, approvals, etc. Any time anything, no matter how small or innocuous (bracket, cable, screw, some piece of plastic cover, etc), was changed on a vehicle, it meant a different part number, which meant 6-12 months delay. This is because it has to go through all the testing - usability, fatigue, safety, etc all over again. This is why they pick cheap parts - not because they're cheap, but because they are old and got cheaper over the years, because old = already approved, which makes the lead time a lot less.
◧◩
46. mirkul+wJ2[view] [source] [discussion] 2023-10-24 17:57:28
>>johnma+z2
Now that I think back, I believe you are correct. The decision was based on making a previously-unavailable feature available, which meant people could potentially (be irresponsible and) leave pets/children in the vehicle while relying on an inaccurate sensor, which could open them up to lawsuits.
◧◩◪◨⬒⬓⬔⧯
47. kakwa_+CM2[view] [source] [discussion] 2023-10-24 18:10:00
>>xctr94+fP1
Indeed, in a car accident, the car itself is rarely to blame.

But that was not my point.

At the end of the day, no matter how well it's built, a car is a several tons lump of steel launched at significant speed. It's an inherently deadly machine.

Having a lapse of attention while driving a car? you might easily cause a someone to die.

Having the same lapse on a bike? you might cause some broken bones.

Having the same lapse while walking? you are good for some "Oh... I'm sorry".

◧◩◪
48. wutwut+fV2[view] [source] [discussion] 2023-10-24 18:50:40
>>mirkul+nG2
Yes because the obvious line to draw from me saying keep the playing it safe culture is that I wish car manufactures remove cup holders and floor mats because death is going to find me final destination style and use any object it can to facilitate my demise.

I forget the term for this, but it’s the same as me stating I like pancakes and you coming at me saying I hate waffles, when I wasn’t talking about waffles at any point. Those types of arguments are insane and I won’t engage with them. I wasn’t saying those things, I’m not defending against your claims that I did.

replies(1): >>mirkul+fY7
◧◩◪◨⬒⬓
49. shiroi+064[view] [source] [discussion] 2023-10-25 01:24:04
>>wutwut+Gy2
Are you denying that there's more guns than people in the US, or that tens of thousands of people are killed by guns there every year? Strange how Americans are so in denial of the war-like state of their country and daily life in it.
replies(1): >>wutwut+Mvj
50. kayfox+Vg4[view] [source] 2023-10-25 03:07:31
>>mirkul+(OP)
I work at a network hardware vendor who has a large Japanese customer base.

After Fukushima we were asked to provide specifications for acceptable operating environment radiation levels, after some negotiation they relented and gave us a figure we could test at, so engineers drove hardware to a spot with high background radiation and ran it for a couple of days off a generator to test.

The Japanese customers also insisted that each display on the hardware be the same shade of white, so they would all look nice in a datacenter, so our LCD displays have specifically graded white LEDs.

◧◩◪◨
51. mirkul+fY7[view] [source] [discussion] 2023-10-26 07:50:09
>>wutwut+fV2
If you are accusing me of being absurdist or reductionist, know this: floor mats were the official cause of "sudden acceleration", where they would slip off their pegs that were holding them to the floor (usually due to human error), and would jam the pedal to the floor. Or sometimes between the brake pedal and the floor preventing correct operation of the brake. In fact, Toyota and NHTSA issued an urgent recall in 2009 to remove all floor mats from vehicles due to this very issue: https://www.safetyresearch.net/toyota-and-nhtsa-issue-urgent...

So yes, the line was very obvious because these are events that happen in real life, risk that you say you wanted to eliminate by absolutely playing it safe: "_anything_, there is no such thing as excessive 'playing it safe'"

I can only assume that your original comment was reactionary and hyperbolic, but then got upset over where that kind of hyperbole lead in the past.

◧◩◪◨⬒⬓⬔
52. wutwut+Mvj[view] [source] [discussion] 2023-10-30 09:29:04
>>shiroi+064
Where did I deny anything? I called the comment edgy. I was stating my opinion regarding the edginess of the parent comment. My opinion, on the comment I replied to, was that it was edgy. That was my conclusion, regarding my opinion, on the topic of the parent comment.

Strange how HNs assume what country others reside in and apply their opinions, projecting them even, onto anything possible, whether or not the thing they are applying them to is at all related to what they reply with, and how they like to put words into the mouths of others with absolutely zero context to be able to make such assumptions, and are in denial about the ignorant-like state of their psyche and daily life as an idiot.

[go to top]