this post was submitted on 02 Dec 2023
712 points (93.0% liked)
Technology
57453 readers
4413 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
There's 2 significant inaccuracies in the article and 1 large oversight in the official video.
PS: differentials are irrelevant when the wheels aren't connected to each other. Individual-motor wheels, as shown in the video, don't need a diff. The non-drive wheels in a 2-wheel drive vehicle do not have a differential on the non-drive axle.
Cv joints are not specific to fwd as nearly all modern rwd cars with independent rear suspensions have CV joints. I don't know of any trucks still using U-joints either since big trucks are solid axle. Cv joints function the same as U joints. The difference is C.V. joints output constant velocity whereas U-joints (what you'll see often under trucks on the driveshaft, two square C shaft ends with an X link between) have lopey output that gets worse with greater deflection angle. If you own a u-joint bit for your socket wrench, I invite you to play with it. Instead of a solid pinned X between the U ends, CVs have free-rolling balls that can roll inboard and outboard to maintain the link between the shaft's cup and the wheel's cone.
The article is inaccurate but the video ignores this part, so I don't fault The writer. The CV joints are said to be a poor design, yet, it ignores the part where the video reinstalls them at 4:20 and 5:10 for the front wheels. This mechanism does not allow angular deflection between the motor and hub, as it's shown, without a CV joint. Lateral displacement, yes, but not angular - as in it can't steer. This may be an overall improvement by reducing how often it needs to bend (only when steering), but it doesn't eliminate it. And even then, the rear suspension is still designed to change camber as it changes ride height. Camber is the angle of the wheel as measured top to bottom, as in what you see from looking at the wheels from the front of the car. It keeps the wheels flat on the ground as you lean the car in a corner. You may see an overloaded car's rear wheels look like /---\ as viewed from the rear or ---/ when hanging free on a lift.
Look, I'm not an engineer at Hyundai (or even a competitor) but this doesn't quite pass the sniff test. Cool idea for sure, but it smells a little like marketing is clamoring for something edgy to display. Even as displayed, the motors and original reduces were already very compact and in close proximity to the wheels compared to a normal engine. The slightly reduced footprint of this uni wheel and slightly increased friction of a bunch of additional gears makes me think this is a fractional improvement in practice rather than a revolutionary improvement.
I'd be concerned with the amount of unsprung weight this adds, too. You're basically taking the transmission and adding that mass to the hub. Seems like it would be pretty crashy on rough surfaces.
I considered that but couldn't make any conclusions. The driveshaft and sun gear are not added to the unsprung. I'd guess only half the weight of planets and carriers is added. It definitely adds the weight of the ring gear to the unsprung mass.
I'm also curious how this affects rotational mass. So while every component spinning with the wheel from tire to motor shaft has rotational inertia, small-diameter components such as drive shafts have relatively little rotational inertia. Wheels and even brake discs have a lot more. I don't have numbers obviously but I'm curious if the rotational mass of the ring gear ends up being detrimental compared to a heavier-weight lower-inertia cv setup.
Isn't the ring gear the wheel body (or whatever it's called), that is, even a fixed axle would have weight there, the gearteeth even provide stiffness. CV joints also contain unsprung mass, I'd say there might be a bit of a difference but nothing drastic. With modern fancy biomimetic wheel body geometries and everything you'll probably definitely be lighter than 80s steel rims. What happened to spoked wheels, anyway.
The CV only contributes half it's mass to unsprung weight, and this system still requires a drive shaft and either a CV or u-joint. But with more mass at the hub.
Typical hubs are still lighter than this, because this setup still requires the hub structure, it's just adding gearing out there.
Half of the hub is unsprung, the transition between sprung and unsprung is at those fancy articulating planetary gears. The drive shaft is definitely sprung, it and the motor is completely static relative to the battery and everything.
Yep.
Trucks used these as far back as pre-WWII. It a great solution for off road vehicles to gain clearance. At low speeds, even universal joints work fine for this setup, because the shaft rotates at 1/3 wheel speed, like a drives haft does going into a differential.
This puts a diff at each wheel.
Edit: These are called Portal Gears
Not quite portal gears, that has the input shaft fixed at the top. This is like an adjustable portal gear.
Meh, it's still a portal hub, where you put the input shaft is a minor difference. It's still putting the gearset in the hub, increasing the total weight of the vehicle, and increasing unsprung weight.
Calling it a new thing is a lie. All they've done is switch it to planetary gears. I'd bet lots of money this was tried a long time ago, and was shelved in favor of an offset input shaft.
I noticed they conveniently didn't talk a lot about steering..
The claim of "one wheel drive" I think is meant to highlight what happens if traction is lost. It sounds like something I have heard on 4wd off-road forums. I agree the phrase "one wheel drive" is perhaps not a great way to explain the disadvantages of differentials vs limited slip differentials vs locking differentials vs individually driven wheels.
The idea of "one wheel drive" as I have seen it used, is that in a vehicle with one powered axle assembly (what we normally call 2wd-- either front or rear wheel drive) is that if you lose traction with either drive wheel, the vehicle no longer moves because all power is diverted to the slipping wheel.
If you have a limited slip differential, there is a limit to how much power is diverted to the slipping wheel. With a locking differential, you only stop moving if you lose traction to both drive wheels.
Anyway...
The design is really interesting.
You also bring up a good point about how camber changes with suspension position. Also the effective track width changes, such as with my 4Runner which has upper and lower control arms, a Double wishbone suspension. If the motor remains in a fixed position, the wheel will move onboard and outboard relative to the motor depending on suspension location.
I don't quite get how these two effects are addressed with this new design. Or are the suggesting a different suspension technology that they didn't discuss?
As for steering, I wonder if the design rotates the motor along with the wheel. In that case no CV is needed but I would guess there are some downsides to such a design.
I agree the video seems kind of... premature. The mechanism is cool but I don't get the sense that its applications haven't exactly been nailed down yet.
Individual motors on each wheel will still slip, just with half the power. So sure, it's an improvement by an unrelated mechanism, but not having the wheels connected with a limited slip means it'll still need a traction control system. And even still, the "half" power is a relative term because every car has a different output. That goes for not connecting left to right as much as it goes for front to back. So, not different than a traditional open diff or 2wd. There have been advances in brake-based traction control so they don't just cut power and apply single brakes like the 00s, they can properly modulate pressure to get equal propulsion.
That's a good point you've mentioned as well - the wheel will change distance to the motor as it goes through it's motions. The only way to avoid that is to place the motor at the effective pivot point of the suspension which is, in a properly design suspension, inside the other wheel to mimic the level dynamics of a solid axle. That of course defeats the short halfshaft design direction. So something has to allow variation in distance. In the non-steer wheels, maybe this could be as simple as a telescoping spline drive. However, the video shows a small black joint at the same time stamps above on the rear and still has those normal-looking cv boots on the fronts.
Or maybe they're ditching good handling and going with perfectly vertical suspension travel. Give it hard eco tires and it'll slide before the suspension shows it's flaws.
The steering wheels' hubs rotate in two directions. The steering action rotates through a vertical axis while the typical suspension rotates in a front-to-back axis pinned approximately through the other side of the car. So unless they abandoned common suspension design to let tires lean in turns more than a reasonable amount, there needs to be an allowance for angular deflection. A pinion gear arrangement sounds like it would take up a lot of the space they're trying to save but still not solve the multi-axis problem found at all 4 wheels. I'm trying to not take their video so literally but it's not like it's a dealership rep spouting incomplete info here
Unrelated comment, but holy shit I am a huge fan of the M8. So wild to see you on Lemmy, and this comment section is exactly what I'm here for.
But you need a CV or Universal on that drive shaft to accommodate suspension travel (or steering if needed on that wheel).
Now imagine there is a handle sticking out of the crankshaft that needs to rotate around the shaft with the hand crank as it gets turned. That is the motor.
The video never showed the motor moving along with wheels turning, only the up and down part. In fact, when they showed space savings they showed the motors as stationary with no clear way to turn.
Suspension movement still requires flex in the shaft. There's no way around it unless the engine moves with the suspension
With how small the motor is you could also pivot it alongside with the wheel, after all, something pivoting with the wheel doesn't mean that it has to be unsprung.
As to the rest- you're wrong. Sorry.
But the real reason this tech won't be very important is because it's a lot more complicated and expensive than a cheap ass cv joint and is minimally more efficient. I can buy both sides of my vehicle for like $80 and don't have to worry about em again for ages. I think this new hyundai stuff could be reliable, but it's going to be a lot more expensive.
Also, they look like they'd be noisy.
No u. Bam, same level argument right there. Are you going to explain why or just throw out contrarian comments?
Cost will not be a limiting factor. Just about every feature on a 2023 car already costs more than a 1993 car's version. Did adding a wheelspeed sensor, electronic 4-channel hydraulic brake actuator, and dedicated ecm programming cost too much to implement ABS? Did the complication of 40 sensors (100+ now) and a voodoo box of electronics cost too much to go efi instead of carbs? Did the price of disc brakes stop most cars from ditching rear drums? Did the cost of engineering and testing prevent manufacturers from implementing the following nearly-negligible aero improvements to eek out another 0.1% of fuel efficiency;
aero strakes into mirror shells (prius, escape)
relaminating roof spoilers into every hatchback/suv and even into every pickup bed
Vortex generators on the top surface of tail lights (sonata, chr)
Active grille shutters (fusion)
Full underbody trays
Chin spoilers (splitters) on just about every car to keep air out from underneath
Hood beak splitters to keep grille air off the canopy (Volvo, accord)
Short antennas/glass-embedded antennas to reduce antenna drag
Front fender outlet vents to create laminar flow over the wheels (f150 2015+)
No, it didn't.
And I'd be interested to hear why you think helical-cut gears will be "noisy". I'm guessing you don't know why reverse whines in certain cars but not the forward gears
Edit: also, seriously, go do some shopping. LSDs are on the decline. On top of never being common in the first place, manufacturers at removing to brake-based simulated LSD rather than discrete components. There are incredibly few Fwd cars that ever had LSDs and fwd obviously makes up the majority of North American sales. Even Miatas and Mustangs only get LSD with optional packages.
Yeah the amount of BS or confidently incorrect in tech articles about automotive engineering is crazy.
And while gear-sets are really efficient, adding 8 to each of the 4 wheels 32 total, in leu of 2-4 at each end of the car, plus 4 CVs (assuming it’s AWD) has to be close in efficiency. And as a DIY+ car restorer in New England. I wanna see how they seal this thing. And with steering wheels it’s either gonna have to swing the motor around or put the CV back in anyway.
NVH (noise vibration harshness) is also going to be a huge factor. That and cost and weight is why we don’t have gear driven camshafts except on a select few exotic motorcycle engines and F1 engines.
Helical gears are the cost effective way to reduce gear noise, but they impart axial loads which would be difficult to overcome in this hinged joint. Herringbone fears eliminate the axial load, but are much harder to make, unless you go with powdered metal sintering. Which I barely trust in a cordless drill, let alone a Kia.
Edit: And no accounting for wheel castor, or camber changes along the suspension travel (way more basic and important for handling than torque vectoring). This video is pure marketing wank.
Funny you mention motorcycles with gear-driven cams as being exotic. Generally correct, but I happen to own one of the cheapest out there: the Honda VFR. I sought out mine for being the last with the gears before going to a normal drive in 02. It's a glorious sound.
Anyway, that's true too, helical gear will need a beefy housing to resist that axial load. That's more weight. A pair of helicals fastened together could help bring down cost, probably, compared to herringbone, but that's added complexity. I can't claim the tech is marketing wank but the video absolutely is
Lol, what?
My RSX was made two decades ago in '03, so the newest that would be multiple decades old. It's also a Type-S, the sporty model. It's got an open diff.
My '93 Subaru Loyale, which is 3 decades old, has two open diffs, with a locking center diff. No limited slip.
My '04 (almost decades old) Crown Vic PI doesn't have an limited slip. It was an option on Interceptor that the city didn't opt for.
My '07 (not decades old) Volvo XC70 has no limited slip diffs. It uses the traction control to try to imitate them, but no actual limited slip differentials.
My partner's '07 (still not decades old) Kia Spectra5 has an open diff.
The only car in my fleet that has a limited slip is my '02 Subaru Legacy Outback, and it was an option that the person who bought it new opted for, and it's just the rear that's limited slip, the front is still an open diff. Apparently the limited slip isn't even that good either, you can still get stuck with two wheels spinning. I haven't tested that yet, I just got the car.
If you go out and buy most cars today they'll come with open differentials. The traction control system will likely try to compensate for this, but they do not have limited slip differentials.
Just how many cars do you have, exactly??
The RSX is on the chopping block. I want to get a more dedicated sports car, since I'm privelaged enough to own multiple vehicles.
The Outback and Volvo are wrecked title purchases from copart that I'll get salvage titles for, then enjoy for a bit before selling to break even, or even for a possible profit.
The crown vic is my beater. I drift it. I take it on logging roads. I take it on the Gambler 500. It's fucking awesome. I'll always have one, probably.
The Loyale is something I've wanted since highschool. I love the push button 4WD. I love how comically slow it is. I love the interior. I get quite a few compliments on it. It needs some work, but that's part of the fun. I dunno if I'll own it forever, but I really like the thing.
I, as of recently, own my grandpa's 1984 Oldsmobile Delta 88 Royale. I've gotten similar compliments to the Loyale on it. My grandpa is about two weeks away from dying of cancer. I'll own that car forever.
My partner owned a Spectra5 when I met her. It's our economy car, which every household needs. I'd like to upgrade it to a Volt at some point.
I definitely own too many vehicles, but I love the variety and unique characteristics of each of them.
Sort of related, but I'm not one of those dickish gearheads, I use and support public transport, and think that the world as a whole would be better off with fewer cars. If one doesn't want to drive, then they shouldn't fucking have to.
Hey don't let me take your fun away from you. I just think it's an unusual amount of cars, but if you enjoy it then that's great! Sounds pretty cool tbh
Yea, companies love to tout their "electronic differential" that's nothing more than an open diff and traction control via the brakes (Toyota Tundra comes to mind, as recent as 2017).
Yeah. I guess the person I replied to fell for their marketing? However, even traction control wasn't terribly common 20 years ago. I'm not sure what they're on about.
It surprises the hell out of me that ABS wasn't even mandatory in the US until 2008 I think. Retro mustangs were available without it. I think we got mandatory tire pressure monitoring around the same time, actually, although it was done by comparing wheel speeds over time. So yeah, traction control wasn't all that common indeed. I think GM was still running 3-channel ABS on their trucks then too
Me too, it was super late. However, holy crap it's different in quality between manufacturers.
My RSX has great ABS, even on gravel or in the snow. It does way better than I could in most situations. Coming from my '98 corolla (without ABS) to the RSX was a massive improvement in braking.
The ABS on my Crown Vic, which is one year newer than my RSX, just doesn't work. You don't lock up the wheels, so I guess by the definition of ABS, it works. However, the ABS also increases your stopping distance by a lot. I need to figure out the best way to disable it. Pulling a fuse doesn't work, as that's also the fuse for the dash instruments. When I do it I'll do a test to make sure that the stopping distance is improved by as much as I think it will be.
That's interesting, so it is pulsing under heavy braking? Is that a common problem? I wonder if either unplugging all the sensors or maybe the module itself would disable it. Just the sensors might freak it out and cause constant abs actuation each startup until it confirms the fault.
Yeah, the ABS "works" as far as I can tell. The wheels don't lock up. It pulses. It's just way too generous towards not locking up the brakes, and doesn't pulse with a high enough frequency, from what I can tell. My vic is lifted for offroading, so it's a real issue stopping on gravel.
I'm thinking pulling a wheel speed sensor would be the easiest, but I haven't looked into it yet.
Might have to pull all 4 or else it'll be 3-wheel braking for the first few miles. I assume the wheels are all the same size? Are the abs wires stained at all with the lift? Might be worth watching them all with a monitor to see if any are acting up. I think Forscan is available directly from their website. It's a pirated copy of Ford's IDS, basically, and I know it was recently pulled from the Google store. It works for ios and windows too and can be connected by usb or bluetooth/wifi obdii dongle.
In the RSX one came lose, and it disabled the entire system, so that's how I assumed it worked in the vic as well, but I guess I'm not sure.
All wires are sustained from when I bought the vehicle, some could've been compromised previously.
I have an OBDII BT scanner, I might check out Forscan, thanks!