Open vs Locked Differential - Torque Transfer - Explained 

“Well that’s not exactly true, In fact it’s not true at all” Lmao I loved that.

"Well im'onna use some math to try to make it clear" Me: *starts sweating intensely*

awesome videos. It's not easy to understand all the math and physics, write it up with neat diagrams, explain it to the RU-vid world, and do a good job at it. Great presentation, style, form, and function. I hope you are working at a university somewhere. I'm looking forward to watching your other videos.

Depending on the gearing of the differential (let's say 1:1 for simplicity, even though this is unlikely), then you would have 200 Nm to split between your two wheels. You could (with a locked diff) have up to 200 at one wheel (theoretically) or a 100Nm/100Nm split.

Yes, if both wheels have the same friction with the ground. But even a simple difference in temperature of the tires could mean only 1 wheel will spin.

Sorry, didn't see this. If the engines give more torque, the wheels will likely spin - the torque applied to the ground will remain what the frictional limit allows, which will be less than what the engine provides.

I've held multiple differentials in my hands, and read and studied them. I made them lock and see how they behave under different circumstances. I've set backlash etc, but honestly couldn't wrap my head around why or how it works, but I simply know that it does. Thanks for breaking this stuff down as I've just kind of gave up on it and decided to proclaim it witchcraft.

I'm fine with the fact that you unsubscribed, however, my logic is sound. The REASON you get out when you lock your diff is because you can now put more torque down, to the wheel that has grip. Find a single credible source that says otherwise. I've done my best to explain this as clearly as I can. I'm not trying to fool people with bogus info. I'm trying to help explain something that isn't intuitive.

Great question! Start with my channel page, the second playlist down (Learn how cars work). Here, I've ordered the playlists sequentially to best learn everything about how cars work. If you have any questions along the way feel free to ask!

3. The numbers were arbitrary, I just picked simple numbers for an example. You won't have a different input than output. You cannot put in 1000 lb-ft and only have one tire deliver 300 lb-ft. Your engine will only deliver the maximum possible. If the maximum torque you can put down is 300 lb-ft, that is what the engine will send if you allow it to. Any thing greater will cause the tire to spin, and rotate faster, until you reach an RPM limit.

The maximum amount of torque that the wheels could put down would be the amount of torque sent by the driveshaft, which in this example was 1300. Though it will be significantly smaller, even a spinning tire does apply some torque to the ground.

Can you do 3 and 3 and see if it makes a difference, or would it then be too loose?

Haha I certainly do have a copy of all of my videos. I've been considering whether or not it'd be something people would be interested in purchasing, and for me to set up a site to sell it. Thoughts?

You are correct on the tq split and wheel speed split. A difference in torque does not require a difference in wheel speed, which is what is somewhat difficult to grasp.

Not sure if this is the same person again and again, but you don't seem to be listening. A naturally aspirated engine does not have forced induction. Hence, ALL of my videos on engines are on naturally aspirated engines, unless I specifically say "this is turbo'd or supercharged." My video "engines - explained" explains naturally aspirated engines.

I find your videos very educational and interesting. What kind of education did you get to have such a great knowledge of cars?

The issues here are frame of reference and physics approaches to real-world situations. If you're looking internally at the stress to components, then your analysis is correct. However, it ignores the very real energy that is expended and wasted spinning the wheel that has less traction. So externally, it's visible *power* (not the "invisible" torque) that is being sent to the tire with less traction. In other words, what bothers people about your explanation is that you are explaining things they can't see while ignoring the things they can. There's a similar problem with the physics definition of "work" being "independent of path". Clearly, in the real world the path matters because a circuitous route with lots of elevation changes and stops and starts (and different modes such as on foot or bicycle or 18-wheeler) uses far more energy than a straight line from A to B. People think of "work" being the total amount of real-world energy they expend rather than the theoretical minimum amount of energy required under conditions that are not actually possible in the real universe.

I am going to take one more stab at this. something tells me this is going to be pointless. Okay forget about speed. Forget about math, Let's talk about how the locked diff actually works. When the Diff is locked. The axle is now one solid piece vs. an open diff that allows both side of the axle to turn independently. So all the torque that originates from the engine, through the drive shaft will not split at the diff when the diff is locked. Because the diff is turning both sides of the axle with the same amount of force or torque. So the amount of force however much it takes, is being used to turn the complete axle at the diff. So the diff is now forced to use whatever force needed to spin the complete axle. Now don't forget the axle is locked together. Again you are correct about the amount of force that might be applied to the ground when measured at the bottom of the tire. But the component itself (the locked diff) can not split torque because all the pieces that make up the component inside all move together as one. LOCKED. What wikki is saying is misleading and what they print should never be used as a source of end all. They can be wrong from time to time. Because people are behind those magical post. And people make mistakes,

No "so".....lost me a bit on this one, I had gotten it before this one....going to check out the next one and if still confused come back and watch a few more times until I do get it.

You have the locked down, but not the open. An open diff receives equal torque to both tires, always. So if one tire has no grip, then it requires hardly any power to rotate. Once rotating, the other wheel with traction is receiving the same force, but because the force is so small, it's not enough to move the vehicle.

Because of the grip/weight, as you mentioned, it looks like it's not getting any torque. Torque isn't something you can see, really. Twist your forearm as much as you can, and keep trying to twist. you are applying a torque at your elbow, even though your elbow is not rotating (hopefully, haha).

hey man, isn't the locked diff limited by the 1000 lbf force? in my mind, none of the wheels can spin unless you reach 1000lbf, the 300 lbf wheel can spin if it was free but it is locked in rotation by its fixity with the 1000lbf wheel and so once you reach 1000lbf it will spin. can you explain to me why the the force required to spin the wheels is additive? thanks

It may have driven fine but he's putting extra strain on the drivetrain, and will increase tire wear if he leaves it like that. Unless the diffs are open in 4WD, then it's not a problem.

Exactly, I do not want to make people pay to watch these, and I don't know how many would simply buy the set as a way of showing support. Though there are of course non-online areas which could benefit, or perhaps in a classroom situation, or parents who don't want their kids online.

Torque and speed do not necessarily correlate. You can have torque without motion. That doesn't mean if you put equal torque to two wheels one will rotate faster - this is only the case if one of the tires exceeds it's grip, and begins to spin, or if one of the tires doesn't have enough torque to rotate.

I wouldn't say one is best. Different applications different types probably work better. Also, I don't have experience driving with any so I can't really give a personal preference, and I think it would come down to the driver many times. An adjustable clutch type would probably be my preference, dunno other than that.

1. Not sure of your question, I will follow up. 2. Because only one side of the shaft is resisting. Don't necessarily think of it as only having torque on one side. The torque can only be delivered where there is resistance. If the other tire were to have resistance, torque would transfer to it as well. All of the torque must go somewhere, so it goes where there is a location to put it. 3. (continued below)

I know it seems unlikely, but it's more than possible. Due to the way the tire deforms, and the direction of travel, it's quite common for tires to be able to put down 1 g longitudinally. Laterally, it'll take some sticky tires or good downforce; see my video "traction circle" for clarity. NHRA dragsters can put down close to 5 g's, and yes, they're basically glued to the road. It has to do with how a tire sinks into the surface; they're not just simply rubbing two surfaces together.

I wouldn't say they don't make them. But most cars that plan on getting into situations where they may get stuck typically have power going to all 4 wheels. That said, I'm sure there are lockable diffs out there for 2WD setups. A diff is a diff. :)

Simply put, torque goes where there is resistance. When you press down on one of your fingers, torque passes to your knuckle. When a force is applied to the ring gear, it passes to what resists that force, aka the tires.

No need for all caps. If traction is equal both wheels will spin, if one wheel has significantly more traction the other tire will spin.

You're going the right direction with your last question. Once a wheel starts slipping, it doesn't take much torque to continue to spin it. Likewise, the other wheel is now receiving less torque, so you'll sit and spin.

Torque split does not require a difference in speed. This is what most people don't seem to grasp. I tried to explain as best as I could in the video above.

Understandable, I don't know the answer unfortunately.

Because it has grip. The friction between the tire and the road is high enough that the force applied is not great enough to rotate the tire.

Well, locked would be better than open - but it's not so hot on the street, so many go with a limited slip.

Hahaha, so tempting. Probably rare that you want someone dating your sister? I have a girlfriend though, sorry man!

Well they're really quite different. An LSD will transfer torque to the wheel with more grip, where as an open diff will not.

Your questions should be cleared up with my next video. If not, ask away and I will do my best.

I love thermodynamics. But what kind of videos were you thinking? I think many of my videos incorporate thermodynamic principles and theories without me explicitly saying "this is thermo."

Basically each brake (on newer vehicles with electronically controlled AWD via the brakes) is controlled individually. By applying the brake to the tire that is about to spin free, the car can keep torque up on the wheels that need it, simply by preventing wheel spin.

If friction is the same at both sides, the torque will be the same at both sides, regardless of the type of differential used. But it wouldn't be difficult for friction to be different. One tire could be slightly more worn, or at a different temperature, or have bits of dirt/dust on it.

This was very helpful. I got stuck with my FWD Open Diff in the snow, with the exact same conditions, and I was wondering why the pavement tire wasn't moving the car.

Improve any mechanical system. Reduce stresses, improve physical characteristics. Anything really.

A spinning tire in the air does no good, as you've mentioned. That's why locking differentials send the torque to the wheel on the ground, and both tires remain stationary until the torque is adequate enough to either spin the tire or move the car forward.

It should read 0.3 for the snow. Apologies if the video is a little unclear.

A live axle is a suspension set up, rather than a differential set-up. You can have different types of differentials within a live axle. I have a video on solid axles if you'd like to learn more.

You're welcome, thanks for watching!

Very welcome, glad you enjoy them!

Not to prevent the car from understeering, but to allow the car to put down the maximum amount of torque.

I believe the manual of the integra is pretty bulletproof; my old mechanic back in NC said they'd never die. Dunno about the autos - love the car. I'm Jason.

I know this is an old video but I've got a quick question about open diffs. If you lift one wheel off the ground to where it has no resistance to spin and left the other on pavement, wouldn't the only torque transfer to the wheel with traction be caused by the spider gears resistance to spin on their axis? Because if there was theoretically no friction/resistance to movement in the diff itself, then the spider gears would simply free spin around the driveshaft of the wheel with traction while trying to spin the one without traction.

It 0.3, but my decimal is a bit faint.

An STI and a bunch of track days.

Please don't use all caps, but thank you for your input.

The torque split depends on the frictional coefficients of the surface. So 77/23 on the surface I described, but it could be completely different on another surface.

Good question, upon googling I learned it's derive from the Latin word libra, which shortened is lb.

Answering your second question, yes. The first depends if it's enough torque to move the vehicle.

Isn't that what this is?

My bad. I upload in 1080p so hopefully it can be read. I'll try to remember to include my 0's.

I will cover Torsen which is basically an ATB diff, so yes.

great vid. I'd love to see you do a video on a centre diff with multi plate clutch, similar to Subaru's system on their 5 speed auto trans. I'm doing some reading on it, but would love to get your explanation

i am too dumb to understand these :O

LSD or locked.

I'm not sure what you're asking, could you elaborate?

my 89 camaro has an open diff..i wondered if the traction is limited by the spinning wheel then i guess my car had no power...but i guess thats only the case if they do happen to break loose...so ill just get supper sticky tires and make sure they don't ever spin

This is a good theoretical explanation, but I feel that it's lacking in practical application examples. You explain this in theory but leave it to the viewer to figure out how it all plays out in practice and how it would affect the behavior of a vehicle in different situations, i.e what are the pros and cons of either system in various hypothetical scenarios. I feel that is the missing piece of this video. I have no idea which one is better where and why.

Both tires get the same tq in a locked diff. If one tire has more traction then that tire is USING a higher percentage of the tq but that does not mean that it is getting more of the engines tq or that it has more available that the other tire.

Thanks! And I do agree, at first glance it is a bit counter intuitive.

Be honest, no it doesn't. :)

No, posi is an LSD.

You remind me of the guy from the movie "God is not dead"

0.3*

So essentially, if you think of the power formula (P=T*w), locked differentials maintain a constant angular velocity but the torque is variable b/w the tires. For open differentials, torque is constant and angular velocity is variable b/w the tires. Am I correct in that statement?

Sorry to comment this old video, but i am not sure it explains how torque can be split, say, continuous 60:40 rear axle vs front to give an awd car more of an "rwd car feel". I know you are talking single axle here, but the same principles should apply front vs rear? You seem to be using surface friction as the variable that defines how much torque can go to each wheel, but all things held constant it should be some characteristic of the diff that splits torque unevenly. Please feel free to point me to another video if this is explained elsewhere. Your videos are very informative, thanks!

I have a question. If say my sequoia with 4hi engaged and center diff unlocked is in a situation where 2 front tires are on the ice and 2 rear tires are on a pavement - am I benefiting from having it on 4hi or is it better to have it on 2hi? The reason I’m asking is as far as I understand if I’m on 4hi and the center diff is unlocked I’ll spin the 2 front tires on ice while rear 2 tires will not and if I just switch it back to 2hi at least 1 rear tire will spin. (I know I have a center diff lock button but assuming I don’t want to use it). I just want to know if I got the concept right:)

Yes.

Why does the torque delivered to the wheel on ice stop at 300 as it reaches the maximum friction of ice? Isn't it possible to have say 350lbf at the wheel with 200 overcoming the friction and another 150 speeding up the spinning wheel?

Ok I watched your video twice. I understand everything except how the locked diff is able to split the torque and how the open diff isn't. I looked on some pictures of open and locked differentials and they look identical to me. Maybe I'm missing something or maybe I'm just dumb (I hope not). As far as I know, both open and locked diffs have pinion gears, and on both diffs they can rotate around the axle they are fixed to. How is the torque split even with the open diff (when the fraction coefficient is different on one wheel) when the pinion gears can rotate?

sorry but your newons are wrong... one pound feet is more than one newton metre a pound feet is about 500grams at 30cm and a newton metre is about 100grams at 100cm... and yes i know grams are no measurment of force just mass but you get idea i hope :)

I love your videos! You've clearly done your research and thinking, your explanations are very clear, great pictures, you get to the point quickly and thoroughly in a short time, and I love your equations... Keep up the good work!!! And THANK YOU!!

hi there, you are saying that when we rev. the engine hard, torque distribution gets settled to 77:23 as in the illustration stated.but if it is having a single shaft for connecting wheel does not the shaft is subjected to shear stresses due to torsion,how that mechanism takes care of that.....looking forward for your esteem reply

so confused cuz it sounds like at first he is describing an open diff or limited slip.. I thought open diff transfers power and torque to whatever wheel isn't spinning .. but he said that a locked diff does that

do you mean you didnt understand my question? im referring to the part at 4:30 where you mention that the maximum the car can lay down is 1300, 1000 for the right plus 300 for the left. i would assume that once the car can move the tire with more resistance, the other tire would move too. but thats just my random guess and thats why im watching your videos!! id really like to understand more and why the engine would need to produce 1300 in order to get the car moving.

some fundamentals, if you would be so kind. 1. locked diff example, why do you add in the torque required to move the other tire with less grip? ie, why doesnt it move when it reaches 1000 instead of 1300? 2. how can there be a different torque on each end of a solid shaft? 3. in the open diff, why is 1000ft/lbs of torque insufficient to move both? if youre inputting 1000ft/lbs of torque but the diff is only utilizing 300 to spin the tire with less grip, what happens to the rest of the torque?

I disagree as well. If i weld my spiders together, there is no way it isn't a 50/50 split. Do you think the rear driveshaft will just magically spin the pinion which will just magically spin the ring gear and somehow more torque will go to one tire than the other? They're WELDED TOGETHER. It's the same split. It can't differ.

Congratulations, you are the only one in auto media that understands torque split is = with an open diff. I wanna have a discussion about the Torque Vectoring Diff as they don't really show how it works. And I said that on their ZF movie and they erased the comment. Problem is it cannot work with only one coupling per side. It's like low and high, or 1st - 2nd you need 2 couplings, which they don't show. And the ratio is fixed so there will be 50-50 in normal driving and a fixed 60-40 left-right in a right turn.

My FWD Volvo S60 has an open diff but when it gets stuck it will brake the wheel that is slipping, which is increasing the torque applied and allows the wheel that has grip to move the car. I hope i understood this correctly and i hope mr Stewart will get it soon. Thanks for the great vids!

So if I have a front wheel drive car and my only options are open differential or welded differential Open differential is better to track on and daily Drive but welded differential is better to drag race on?

Correction: 8:30 Think you mean 400lbs force with an open diff vs 1200lbs force with a locked diff, assuming the same lowered (dynamic?) friction coefficient...or 600lbs force vs 1300lbs if we assumed no decrease in friction...but point was taken never-the-less. Had to watch it a few times to get the difference, but I think I got it. Hard to get it out of your head that torque != velocity, especially when it comes to locked diffs.

Got open..but lockers: If I'm understanding right: If one tire is in the air, one can put down full 1000lbs, the system as a whole resists up to 1k lbs of torque, as that's what's being resisted by 1 wheel, after which both will start to spin at the same speed. This makes the velocity the same (which is what is experienced in a locker), but torque is 100% given to traction tire..but this is only 50% of the drivetrain's capability (as ideal conditions, 2 tires, 1k ft-lbs each, 2k lbs force)?

Locked differentials don't work well for normal driving on pavement or on any high friction surface on which turns are made. This is why old cars used to have one-wheel drive in pre-open-diff days! When you make a turn, your wheels must rotate at different speeds to maintain rolling contact with the road, because each tire has a slightly different distance to cover. For example, when you make a left turn, your right rear tire moves further than your left rear tire. Hope this helps.

502?! That's nuts. Very cool!

That was the stupidest comment ever mate. Why do you think new models of cars come out every year? Do you believe that everything is invented (is it impossible to invent new stuff)? There you go, with two simple questions you should have made yourself before posting the comment, I answered what you asked.

Glad you enjoyed the videos!

Stupid question - why don't they make lockable differentials for cars with 2WD? It would be a nice feature to have on a FWD or a RWD car, whenever you get stuck. Or is that basically a LSD?

Dude you are awesome, very interesting video. Thanks!

Thanks!

I love when car enthusiast try to prove a mechanical engineer wrong and he gives you a lesson on how you are wrong! Great videos, keep them coming!! I swear you have a video for all my questions about mechanical engineering.

Where did the frictional coefficient numbers come from? .3 for the ice side and 1 for the pavement side?