Hacker News Reader

Comment by codeTired | Hacker News Reader

Just be careful not to dent quarter panels on a rivian. It’s something like $40k to replace. Normally a $2k-$3k job on any other pick up truck.

These car companies suck when it becomes to repairability.

Edit: I’m not going to argue with naysayers below, this dude was quoted $41k but was able to PDR(https://www.thedrive.com/news/41000-rivian-fender-bender-act...). Tesla repairs with giga casting are also a lot more expensive than similar “regular” cars. This isn’t an issue of unibody design. My family owned a body shop and I painted cars for few years.

I will stick with main stream brands like Toyota.

View on HN

Probably closer to 8-14k if you’re not willing to leave the dent. The famed 42k bill was for a rear end collision that caused more damage than it looked like. Cars are much safer today but crumple zones mean minor fender benders result in huge repair bills. However, they are also a major part of the 90+% reduction in vehicle fatalities per mile over the last 100 years.

“They took the R1T to a certified Rivian body shop who said the entire rear quarter panel, as well as some tailgate internals, needed replacing. The total cost of replacing an entire body panel and some tailgate bits was reportedly just over $14,000.” https://www.thedrive.com/news/rivian-r1t-fender-bender-turns...

I was rear ended about 8 years ago. I was in a lane that is parking most of the day (like when I stopped) and no stopping/anti-congestion during rush hour. My Camaro was hit by a person traveling at least at the 45MPH speed limit. They were on their phone and thankfully were so angry they unintentionally confessed it all to the police office taking the report. I ended up not paying a cent to fix the car, but I did get a copy of the bill.

All in all the cost to repair it was in excess of $18,000 from a third party shop that didn’t do a great job on the finish. That was only $5k less than the Kelly Bluebook value for my car at that point in time.

In short, it costs a lot to repair crash damage. That’s something people probably don’t think about too deeply if they have comprehensive coverage and pay just their deductible.

A dent is much different than a 45mph crash. At that speed you have to replace half of the car. We have done those at our shop on an occasion. It’s rare as it’s a pricy repair.

It actually wasn’t far off from the “one panel” as the 5th Gen Camaro has a long panel that stretches from the bumper to the tip of the A Pillar. Otherwise it was a bumper cover, crash structure (which is disposable above 10MPH) and a trunk lid. The biggest cost was the panel.

> In short, it costs a lot to repair crash damage

...and I suspect the car will never be "right" again.

Mine was actually fine. GT3 race cars get straightened on the same rigs after crashes so it apparent works ok for a harder use case than daily driving. The computerized rigs are better than the old days.

In my case I did have a couple trim pieces that didn’t completely match.

The challenge with these repairs is that they only focus on the body part. However, nobody knows the damage to the other systems in the car when hit at 45mph. Imo, it would have been better for the insurance company to total the car and pay you to buy another one.

I had a VW Sirocco back in the 1980s and had to replace a rear taillight assembly; brake, turn, stop lights. I think it cost about $200 at the time which was a shocking amount. I probably only made $200 per a week in my job.

My uncle who knew a lot about cars gave me some advice. He said check how much an alternators costs before you buy a car. VWs seemed to be cheap to people but the parts were astoundingly expensive. But me as a teenage ignored all that. Kind of like people now buying Tesla for the looks, technology but pay dearly for any body damage.

I find it funny that all the Porsche guys try to find interchangeable parts with VW. Makes sense there is probably overlap between the manufacturing. I do the same thing with my old cars.

I drifted into the back of a car on snow and ice. My car was going 10-15mph. $15K of damage repaired at an independent shop... (they did an amazing job and fought the insurance company for OEM over pattern parts). Car was worth under $20K. Maybe $17K?

If I showed you a photo of the damage you would believe me even less...

(The morning I got the car back, someone drove into the front and caused another $12K of almost identical damage. That was a great day)

Vehicle fatalities for people inside the car but not for those outside. We’ve made the average vehicle much more dangerous for anyone not in another large vehicle.

You got a stat for this?

https://www.greencarcongress.com/2020/03/20200302-ghsa.html

It's unsurprising when you think about it: America is full of larger trucks which have to meet o on the laxest of regulations to minimize pedestrian damage in case of a collision. Where before the contact zone might be on the knees, many a modern truck will hit a pedestrian in the chest. Add to that how America sets up situations very unsafe street crossings, often 8 lanes wide without a median, and increased pedestrian deaths are unsurprising.

Note that this is a matter of most security freatures: Good crumple zones don't lead to tall trucks with large hoods. The only one that adds risk is the curtain airbags that have lead to wider A pillars, and with that somewhat lower visibility. But we could make cars safe and also safer for pedestrians: Just not by driving heavier, taller vehicles that look like 18 wheelers and have frontal dead zones that could hide multiple Miatas.

Walkability in the US is indeed dire, but:

> Add to that how America sets up situations very unsafe street crossings, often 8 lanes wide without a median

I don't think I've ever crossed an 8 line highway on foot (US native). I definitely wouldn't call it "often". Maybe in a place like LA or Houston? I do cross 4 lanes pretty frequently, though.

They are everywhere in suburban USA.

Arterial roads, such as those lined with big box stores and shopping plazas, are commonly 3 lanes in each direction at 40mph speed limits (which means cars are driven at 45mph+), plus turning lanes.

Your best case scenario is the crossing being ~56ft, but 70ft+ crossings are normal for busy intersections. Which are ones you would also want to walk across.

Even a two lane in each direction road will commonly have a turning lane or two, making the intersection a 5 to 6 lane wide crossing.

I spent time in a couple of midwestern cities this year and was shocked at how wide the streets were. It was mind-blowing to me, coming from a city in the northeastern US. The typical road was wider than the "highways" we have near me (not interstate, but they are numbered state routes).

It also threw me off when walking as I could see I'd have to go a couple of blocks but every road crossing took a few minutes. Invariably I'd have to wait at least one light cycle, perhaps 2 if I was crossing both angles. And those light cycles were long.

The worst part was that it was rare to see much vehicular traffic at all except during some key times. What a waste of space.

The insane part is that there are in many cases NO trees along the road. Lawns everywhere, concrete. Ok, maybe a few trees, think 1 tree every 20m, and barely any bushes. A chance to contain that madness and sound levels, but nah, gotta see the huge billboards clearly from kilometers away...

I had to walk down one of these roads in Denver, let me see if I can find it.

W Alameda Avenue. 7+ lane for a lot of its length. Super sad walk.

9 lane intersections are extremely common in Phoenix. 3 lanes in each direction, 2 lanes for a left turn, and 1 lane for a right turn. Often times there will be a small median where one could wait out a light, but not always. Scottsdale @ McDowell[1] is an example of one without. If you pan around to any of the major intersections (every 1.25 miles in each direction, generally), they are pretty much all this large.

1: https://maps.app.goo.gl/K4Qs5EtTp8Ms6DyC8

Pedestrian Safety: We Just ERASED 40 Years of Progress https://www.youtube.com/watch?v=fEj-pyjA2oo

I guess thats where AEB technologies come into play

[1]:https://twitter.com/remouherek/status/1716012670262931946

It will be a very long time before all cars will have systems like this (and I mean all cars on the road, not ones being sold).

This is why it’s important that we fix street design to lower speed limits, reduce conflicts at intersections, and improve safety of road users outside of cars.

Subaru had FCW since 1999, AEB since 2008...

To me this seems like a design failure (maybe intentional?)

I'd think a properly designed crumple zone would only crumple at a "fatality" speed and not at fender-bender speeds. I remember asking a friend why you almost never see Lexus in China - and they said they notoriously crumpled at the slightest collision (there are a lot of very low speed fender benders on city streets there). Other Japanese brands seemed to not have this reputation

There’s been an effort to minimize the damage at very low speed (under 5MPH), but it’s harder to optimize for fender benders than you might think. People tend to break before collision so crumple zones are optimized for lower than highway but still deadly speeds ~40 MPH.

New car buyers also have insurance so they rarely care much about collision repair costs. The fact fender benders don’t total the car is just about the only metric car buyers care about.

"just" over $14,000.

You can buy a new car with that amount.

Not in the US, you can't.

A new (used) car, perhaps.

Mitsubishi Mirage is ~$17k and gets incredible gas mileage. Certainly not $14k but not _that_ far off.

This has not been the case for years now.

> The famed 42k bill was for a rear end collision that caused more damage than it looked like.

This is the Paintless Dent Repair (PDR) for that Rivian that got the famously large bill. They go through what all would need to be removed to replace the panel. It seems extensive.

https://www.youtube.com/watch?v=DKPfy5djvLc

$40k is just incorrect. My buddy did exactly this and he's facing a bill of $14k. Which is still outrageous, but like three times less.

I bought my 20 year old infiniti for 14k, 10 years ago. I think most people will be in serious issues when they suffer a 14k bill for a fender. I think if do a large overhaul including a full repaint on my car it would cost less than 14k

Most people will just suffer the insurance deductible though.

If your vehicle costs that much for a "simple" repair you're going to be suffering from expensive insurance too.

So... pretty much everyone these days?

It's a third of $40k, "three times less" than $40k would be -$80k

"X times less" is actually a technically meaningless statement, but colloquially would mean 1/X. There's nothing to be gained by being pedantic over that

It's just annoying to see, like when people write "less then" or "should of"

Despite claims to the contrary, I haven't really seen evidence of this with Tesla. The giga casts have pretty conventional crush structures ahead and behind them. It seems like the kind of crashes that'd cause massive costs are big enough to be massively costly either way.

The Rivian case is a bit of a head scratcher. They made some truly odd choices that are making fairly simple repairs much harder than necessary. It is obvious they are new to this.

First to market, but with beginner mistakes. I'll wait for a more mature EV truck myself. I want a simple box frame, with a body on frame, nothing fancy. Just pure utilitarian EV truck like a base F150 but with a battery.

I've heard that EV trucks are mostly unibodies given the structure they need to protect and support the battery underneath. But I think the F-150 lightning is simple body on frame (as you said, although I think you are confused that this doesn't exist yet?), and the new Silverado EV will be neither unibody nor body on frame, whatever that means.

Unibody and body on frame have different pros and cons. Most people buying a truck want a unibody. The F150 lightening is only available in the crew cab version - you can get cars with as much cargo space. If you are actually hauling heavy things, or towing a trailer then a frame is strong and better able to handle that. However if you are only hauling people with some light luggage then a unibody is plenty strong and betters resists twisting forces making for a nicer ride.

I think F-150 Lightning is the closest to what you are looking for. But it has its own type of beginner's mistakes with the batteries. The charge curve isn't great, if that matters to you.

I'm in the same boat with you. I want a truck or SUV to replace my 2003 LX470 (Land Cruiser) but aside from the body on frame construction, I really need a 500 mile battery. The 320 mile EPA extended range lightning loses about 64 miles of range from the recommended battery boundaries (don't charge above 90% or below 10%) and probably at least another 10%-20% loss from the high torque required offroading and the fact you'd probably never use regenerative braking since there's a lot more technique involved leaves you with about 200 miles of actual range. Throw on a roof top tent and some camping gear in the back and you probably have less and you have to account for the lack of jerry cans that can help you in a pinch if your planned mileage is increased from a bridge that's out or closed trail. I really want an electric SUV or truck to replace my main rig but some of the locations I go to are just too dangerous to run out of juice with. I think in about 5 years a dream rig of mine will be on the horizon

I wanted a B2 bollinger, and still love the concept. As a work truck, this has some features that would save me hassle. Also like the simplicity. But last I checked the cost was over 100k, which blew me away.

It is fairly expensive to fix a Tesla panel too. I received a small dent on our trunk thanks to a driver nudging it while parked and it cost $8,800 to fix just that panel.

I’m a Rivian reservation holder, since June 2021. Likely going to pass on my reservation.

But, make no mistake, I’m pro-Rivian, and am on the Rivian subreddit.

Rivian owners and fans are quick to jump on any valid criticism of their beloved Rivian and downvote contrary thought into oblivion.

Not surprised you’re getting hate for valid feedback.

What does this have to do with them winning a contest? It comes across as argument baiting.

Good thing to keep in mind imo.

It's s truck, leave the dents alone : - )

If car bumpers are made immune to parking lot speed hits, will they necessarily be worse at crash safety? Looking at how 4m cars have more complex bodies than older Porsches and pack lights at each corner of the bumpers.

Does this mean they are incredibly expensive to insure?

A buddy close to underwriting heaves giant sighs and mutters darkly when anyone in his book of individual policies wants to insure one because yes.

Is that because after barely scraping by $80k or an $800/mth payment, they freak out why there’s another $300/mth?

Well at least the first two are paying for an asset, whereas the insurance payments are paying for something nebulous that you may or may not need. The ideal insurance premium is $0/month.

Er... I own one and it is maybe 10% more to insure than my previous Subaru.

How is that possible, just given the sheer cost difference? Or is most of your insurance policy covering liability (for if you hit other people), as opposed to the risk of damage to your own car?

It’s true. They are surprisingly cheap to insure. The current theory is that they are new and insurance companies don’t have enough data yet. Everyone expects rates to skyrocket as more $40k repair bills come in.

I kind of doubt rates will go up like that. It's not like BMWs, Audis, etc are much cheaper to repair (by "authorized repair shops" at least), and those similarly are not wildly more expensive to insure.

Because the cost of your vehicle is a small fraction of their total liability.

Suppose your insurance cost is 100$/month for a Subaru and 110$ for a R1T but only 90$/month for liability + personal injury. Now your total bill increased by 10% while they are charging you 2x to insure the vehicle itself.

They may also be getting this calculation wrong, but it’s unlikely to change that much.

For a while my new EV was less to insure than my older ICE both with full coverage and same driver listings. This last renewal the EV is now slightly more, by like $50/6mo

I'm curious how they hold up. Because, IRL, the Rivian is a beautiful vehicle, but the body has always struck me as 'precious' and for the kind of things I want a truck for, I'd be really hesitant to mess it up. By contrast, and old mid-90s pickup has the kinda body that a ding or two or ten isn't going to look all that terrible on.

Yeah, insurance will not pick that up if it was caused while off roading.

I’m guessing taking your Rivian off road will also void the warranty.

If you off road it’s good to know how to repair your rig.

It’s best to off-road in a reasonably built car, like a Land Cruiser or a bronco.

It won’t void your warranty. The Rivian was designed, and is, an incredibly competent off road vehicle.

It literally has an Off-Road drive mode with 4 different setups within it and a built in air compressor in the truck need for adjusting your tire pressure when you take it off road.

> It won’t void your warranty.

Wow, that is truly surprising. I’d be more worried about the driver if I were Rivian? What stops someone from driving through a creek and submerging their truck? Damaging their undercarriage on a rock? Sliding into a tree? etc.

> The Rivian was designed, and is, an incredibly competent off road vehicle.

It looks like it doesn’t have locking differentials but that’s usually overkill.

> It literally has an Off-Road drive mode with 4 different setups within it

I’ll need to read more about the “Off Road” drive modes. It sounds similar to the Tacoma’s “Crawl Control”.

> a built in air compressor in the truck need for adjusting your tire pressure when you take it off road.

An air compressor is nice but you can pick up a Tsunami Air Compressor and a tire deflator for like $130.

Off-road cars have limitations. I'll bet the Rivian just like other 4wds will have a maximum wading depth for water crossings.

If the door seals are good there's not a maximum wading depth, rather a maximum time paddle boating in water before seals give way | river current moves the 4x4 too far.

See: https://www.youtube.com/watch?v=LC5ld79joIA

Note that the Land Cruiser is bouyant in the water and not touching ground.

How did the driver turn??

Back wheels spinning in water create weak driving force forward.

Front wheels can turn and provide rudimentary front rudder effect wrt force from rear wheels pushing forward.

If 4xwheel drive was engaged the front wheels wouls also be driven and provide a "clawing forwards" directional force.

It's not the ideal boat, but it floats and works up to a point (as seen).

With the engine weight forward it's entirely possible the front wheels were occassionally touching down when over the pan of the crossing.

When the vehichle went to the left of the crossing it's unlikely the front wheels had any contact as the crossing (unflooded) is typically higher than the river bed on either side.

Looks to me like mostly momentum and current carried him to a spot where his front tires touched bottom at which point he was able to spin around. He was pretty adrift until he hit that spot.

https://www.youtube.com/watch?v=T21njN2HJgY

It doesn't have differentials at all ...

Each wheel has it's own motor.

I see so slips are impossible to prevent, this seems like a big limitation for serious off roading.

https://www.rivianforums.com/forum/threads/quad-motor-vs-loc...

The Tacoma Off Road Crawl Control is somewhat similar but uses ABS to prevent slips. It does come with a rear locker though.

I'm not an expert here, but what stops four wheels from all rotating at the same speed with independent motors?

From the link:

> If the axles are not physically tied, there is no way for any of the four wheels to know how much traction it actually has until it breaks free. Then, of course, it knows "too much" torque and it can apply traction control. But the damage is done. It has broken free, and any negative consequences have already happened prior to the computer being able to make any further calculations.

> Again, if physically locked together with large gears and drive shafts, there is no possible way for one wheel to break free in that manner, unless of course something is broken.

All four wheels rotating at the same speed is the normal case: straight line driving on good pavement.

The real question is can all the wheels go to zero traction at the same moment and thus the computer thinks all is well. I think the odds of that are low enough to not worry about.

Traction control systems also use accelerometers and gyroscopes, so the scenario you describe is not possible.

Well, yes, but you can also sense which ones are turning slower and just increase the torque to those. Locking diffs simply provide torque to the wheels with grip, which is exactly what one motor per wheel can do too. There’s no difference in the real world, and in fact performance is even better because there’s no turning circle penalty from having a locked diff.

> Well, yes, but you can also sense which ones are turning slower and just increase the torque to those.

From the link:

> Again, if physically locked together with large gears and drive shafts, there is no possible way for one wheel to break free in that manner, unless of course something is broken.

Some other interesting tests by an R1T owner.

https://www.rivianforums.com/forum/threads/quad-motor-vs-loc...

I mean, doesn't the set of computer controlled individual motors completely and totally contain all scenarios presented by a locked differential?

Specifically, wouldn't one line of code that says "never ever ever turn this wheel faster than this other wheel" be totally and completely indistinguishable from having the wheels on the same axle, as far as the physical world is concerned?

That's not how electric motors work.

You can measure the speed and adjust the power going to the motor if you go over or under that speed. The lower latency that this cycle has, the better, but it can never go to zero like with physically connected wheels.

Pedantry: The speed of sound in steel is about 5100m/s, so a solid 2m axel will still take about 4µs for interactions to propagate from one end to the other.

Maybe so but what does that have to do with anything?

A physical axle can have both wheels break free too can't it?

Comparing apples to apples the only scenario you're concerned with is where one wheel breaks traction and the other doesn't. Why is a logically driven exact speed match between wheels inferior to a physically linked speed match between the wheels. Explain the difference practically speaking.

You said "totally and completely indistinguishable".

Practically, electric motors can't generally transfer torque from one wheel to another like you can with locked hubs. (The only exception would be where your battery is underrated to provide sufficient power to both motor to maximize their torque)

It isn't surprising that off road is covered in warranty. They may put limits in, but look at truck ads: most of them are showing people towing a heavy load, hauling big things, or driving off road - often more than one of the above. Sure that isn't how people actually use them (most of the time), but that is how they sell them. I haven't looked at Rivian's ads, but if they show their trucks doing something in an ad, then a lawyer will have no problem convincing the courts that because it was in the ads the warranty fine print was a misprint and that is acceptable use.

Rivian insurance covers off-road use.

Toyota are also looking into switching to single casting bodies, so your comment both includes a fallacy and misinformation.

> These car companies suck when it becomes to repairability. > I will stick with main stream brands like Toyota.

Even if this is true (which is unlikely, the first reply says 8K - 12K, the truth is a lot more nuanced), the question we must ask is why do cars crash into each other? We must require all manufacturers to have Vehicle to Vehicle communication (beacons with chips and smarts are incredibly cheap) so they never crash into each other.

Car accidents are the biggest cause of death until late 50s (then cardio vascular, etc become the biggest cause). Modern emergency medicine is incredibly good, there are a huge number of people disabled and injured (but alive!) from car accidents. IN 2021, 42,939 deaths[1]. If we just add up this decade (2020 - 2030), we are looking at 400K preventable deaths.

Consider the injured and damaged vehicles. In 2019: 36,500 people, injured 4.5 million, and damaged 23 million vehicles[2]. In a decade (2020 - 2030), we are looking at 400K deaths, 45 million injuries and 230 million car repairs. And, this is just in US. Worldwide numbers are probably 10x! Body shop repairs are the costliest repairs even for minor scratches and dents.

A car is a machine, the safe operation and liability should be on the manufacturers. Legacy car manufacturers are criminally negligent, they don't know how to do software or electronics. New electric companies, they have a lot of young talent in software+electronics, can build the V2V capabilities to make a car safer.

[1] https://www.iihs.org/topics/fatality-statistics/detail/state...

[2] https://www.nhtsa.gov/press-releases/traffic-crashes-cost-am...

Reply on news.ycombinator.com