complex question for all you guru's
 

Hey VaDriven,

Wondering if anyone can explain how you can have torque split and the same angular velocity in an AWD car?

Re: complex question for all you guru's
 

Angular velocity of the car (yaw rate), or angular velocity of the wheels, or angular velocity of the prop shafts? Angular velocity of what?

Re: complex question for all you guru's
 

How exactly does a posi-trac rear-end on a Plymouth work? It just does.

Re: complex question for all you guru's
 

Yeah, more info is needed.

edit: I'm going to take a shot in the dark and say your are wondering how you can have torque split at the center differential if the front driveshaft RPM is the same as the rear driveshaft RPM?

Re: complex question for all you guru's
 

It just does. Somewhat crudely, but it works. But it's no mystery how it works. :D

Re: complex question for all you guru's
 

The answer is that torque doesn't care about RPM. See, easy.

Re: complex question for all you guru's
 

Duh, but I don't think that was the question. I think he is trying to ask how can a AWD car send more torque to the front or the rear of the car if all 4 tires are spinning at the same speed, i.e. the center diff is not "diffing".

Re: complex question for all you guru's
 

Well, that's an easy answer, but might be a difficult thing to wrap your head around. I know it tends to bend my brain a bit when I start to animate things in my head and visualize how it works.

Cars that have a center differential with a static torque split (under normal, non-slip conditions) have a different gear ratio for the front driveshaft and rear driveshaft, so with a planetary differential that would be different input/output gearing. You would likely say "well that would make the shafts run at different speeds" but that isn't correct. The difference in gearing changes the torque multiplication between the front and rear wheels, but the front and rear wheels are moving the same speed, so any differences in planetary gearset motion are internal to the differential.
This obviously changes when there is a difference in grip between the front and rear wheels, which means that there is now a slip condition if one set of wheels is overpowered by torque, and therefore a relative speed difference between the front and rear shafts. Then the torque split would change depending on the type of center differential and what it has for lockup characteristics, especially if you add clutches and other elements like that into the equation.

If it's not a planetary differential, so for example a crown gear differential, it gets much more brain bending because it becomes a case of gear diameter that gives the static torque split, and there is no change in gear ratio (tooth ratio). It's the application distance (moment) that is different between the front and rear shafts. If you see a picture of it, it's brain warping because it doesn't look like the torque split should actually work.

Re: complex question for all you guru's
 

The real question is can jet fuel melt steel beams? The answer is no.

Re: complex question for all you guru's
 

Why are boobs good?

Re: complex question for all you guru's
 

:D

Re: complex question for all you guru's
 

I looked up Audi's quattro crown gear center differential. I see what you mean by mind bending. It doesn't help that the input shaft and front output shaft are co-axial (one is hollow and the other passes through the center of it). That adds to the mind bendyness...haha.

Re: complex question for all you guru's
 

The coaxial part is fine, it's just two things on the same rotating axis that are different diameters so they don't interfere.

The bendy part is that you have two crown gears with the same tooth count, and are applying torque at a different radius on each gear while the two crown gears and the input shaft are all rotating the same speed (and therefore without any relative motion of the spider gears) and all three inputs/outputs share the same rotating axis. The shared rotating axis and difference in torque with no relative motion is what stalled me out the first time I saw it. It's easier to wrap your head around during wheel slip, when one of the crown gears is then rotating at a different speed than the other crown gear.

I think I remember that the key to the torque split is the ability of the spider gears (which can freely rotate about the cross shafts) to be in equilibrium on the two crown gears. I can't remember, I'll have to go look at it again. If they weren't in equilibrium, they'd rotate along the crown gears instead of just being a point of force application. So the spider gear transfers torque from the input shaft, but is also a gear that rotates at a distance from the center of the input shaft, and without that distance component you can't do the torque split. So the rotation of the spider gears make it possible to have different gear ratios from multiple shafts that share the same rotational axis.

You couldn't just have a flange on the input shaft (instead of the spider gears), and flanges on the two output shafts (instead of the crown gears) and bolt/rivet the flanges together at different diameters. You'd have no torque split if you did that, because you'd be transmitting torque at whatever flange attachment diameter into another flange with the same attachment diameter along the same axis. So you'd have a 1:1 ratio and the diameter therefore doesn't matter. You need an input gear (cross shaft and spider gears) driving two output gears for the concept to actually work at all.

Interestingly, just like an open rear diff in a RWD car, you have to have both output gears. If you don't have both, it will freewheel and you'll get no output torque at all. In this case though, if you break a front or rear shaft, the lockup ability from the clutch packs will still allow torque to be transmitted to the other shaft.

Re: complex question for all you guru's
 

Simply put: The angle of the dangle, needs to be in proportion to the heat of the beat.

Re: complex question for all you guru's
 

This is almost as fun as looking under a 5th Gen COPO this afternoon and pondering the multi-link strut (front) suspension.

Re: complex question for all you guru's
 

Too bad the OP doesn't seem to care about the thread he created...

Re: complex question for all you guru's
 

That's because the OP just wants to see the classifieds. You should probably reset his post count.

Re: complex question for all you guru's
 

He has 164 posts, so the classifieds shouldn't be an issue. I think....

Re: complex question for all you guru's
 

Oh, wow. No idea then.

Re: complex question for all you guru's
 

This is what helped me understand what as going on. Once I realized the input shaft and the front shaft were co-axial I was able to figure it out fairly quickly. I visually turned the input shaft 90 degrees and ran one of the crown gears (not by the same gear teeth that make the crown) with a piñon gear and then since you basically have spider gear running output gears on each side but the output gears are running on 2 different radii but with the same ratios because of the tooth count, it started to make sense. I think that works, but at any rate it helped me figure out what was going on.

Sometimes I wish I was a automotive engineer instead of a boat engineer...

Re: complex question for all you guru's
 

Lol! Lots of good answers here.. :D

Re: complex question for all you guru's
 

You asked.... :D

Re: complex question for all you guru's
 

That is exactly what I was looking for. So if it is a clutch type, the clutches are unlocked during strait line driving with no wheel slip? Can the tq biased be changed dynamically without wheel slip? Sorry, I'm not too familiar with how a planetary gear set is used as a tq biasing center diff.

Are there center diffs where the clutches lock the two outputs together unless they are "opened" or over powered? - like a clutch type rear diff?

I'll leave the crown types alone for now haha.

BTW - I care about the topic - all the useless comments in this thread and I get accused of trying to boost my post count? :rolleyes:

Re: complex question for all you guru's
 

No, the clutches are not unlocked. They are locked but aren't doing anything because there is not relative motion because there is no speed difference. A clutch-type center diff (there are viscous center diffs as well, etc.) works like the clutch type rear diffs you're talking about, where the outputs are locked together until there is a speed difference. So if you start with a 35/65 torque split (for example) in a straight line with no wheel slip, you'd add/subtract from that as the limited slip function shifts torque to the front or rear as you have a front/rear slip condition.

You can't change the torque split dynamically without wheel slip, because limited slip differentials don't function until there is a speed difference. Until then, the diff has the normal front/rear torque split from the difference in output gear ratio. So something like an STi has a 35/65 torque split until there is a slip condition. Then things get complex.

So if you wanted dynamic torque split without wheel slip, you'd need some variable gear ratio in the diff, or artificially create a speed difference between the two shafts without having a difference in wheel speed.

Re: complex question for all you guru's
 

http://en.wikipedia.org/wiki/Torque_vectoring

?

Re: complex question for all you guru's
 

Yes, they vector torque during wheel slip. During normal operation they use the mechanical torque split built into the differential, like I mentioned above. They are basically controlling the clutch operation (or other means based on the diff design) to transfer torque during slip conditions. There really isn't any need to dynamically transfer torque under normal conditions, until you lose traction or overpower the tires with engine torque.

Re: complex question for all you guru's
 

Yes I agree. Torque vectoring does allow for better handling while working with VSA. On SH-AWD I notice when climbing a steep hill under normal throttle, the system directs more torque to the rear wheels to better assist the climb. It is a very intelligent system if you ask me. On the RLX they also use torque vectoring to assist you through turns no matter what condition it is. It literally feels like you are floating(smooth) through a turn in that car. Maybe I can post a video later to show it's unique functions. On the ODMD, it allows you to monitor the SH-AWD system which is pretty cool to watch while driving especially in the snow/ice.

Re: complex question for all you guru's
 

This is possibly the poorest written article on a technical subject I have ever read on Wikipedia. It reads like some 11th grade Subaru fanboi wrote it for his class paper. The guy that wrote this thing sucks at technical writing. I would suggest reading this one instead:

http://en.wikipedia.org/wiki/Quattro...-drive_system)

Re: complex question for all you guru's
 

Yes. As someone who has had to do a lot of technical writing in the past, I agree.

Re: complex question for all you guru's
 

I couldn't even get through the second section. What a terrible fucking article.

Re: complex question for all you guru's
 

I too agree that it is poorly written but "Fabrik8" did you understand it? I understood it but I guess to the individuals with less automotive knowledge that article can be hard to grasp.

Re: complex question for all you guru's
 

Yes, but I have a decent knowledge of vehicle dynamics and stability control and can fill in the blanks that are all missing in the article. There really wasn't much actual information about how it works or the operational theory behind it.

Re: complex question for all you guru's
 

This. X2. I'm not claiming to know as much about vehicle dynamics or stability control as Fabrik8, but I know enough to realize that the article doesn't really give you any specifics about the theory or any examples of solutions other than a specific manufacturer's catch phrase for it.

Re: complex question for all you guru's
 

There's not an output speed difference but there has to be relative motion some where if the gear-determined-torque split is 35:65 right? I thought that's what you were explaining here:

Re: complex question for all you guru's
 

I believe he is talking about relative motion between the input and output of the clutches. If you are using a crown gear type center diff with a uneven torque split there is still no relative motion. I'm not sure about the planetary but I would think it is the same.

Re: complex question for all you guru's
 

Correct, the planetary gears and differential are moving, but there is no difference in speed between the front output shaft and the rear output shaft because there is no wheel slip. The clutches aren't doing anything because there is no difference in speed between the front output shaft and rear output shaft, so there is no limited slip action.

Re: complex question for all you guru's
 

Question: in a planetary differential that supplies something other than a 50/50 torque split, is the difference in size/ratio of the ring gears' the charactoristic that gives you the uneven torque split between front and rear outputs?

Re: complex question for all you guru's
 

The difference in ratio is between the gears that attach to each output shaft, so the front output and the rear output have different gear ratios with respect to each other. In a general sense, in a planetary gearset that can be done a few different ways, either by changing the tooth count of the gear that attaches to each output shaft, or changing the tooth count of the side gears (I can't remember the name for that gear).. There are multiple architectures of planetary diffs, with different ways of orienting the output gears and side gears in perpendicular or parallel configurations. So the front would have a different ratio of output gear to side gear than the rear would, but again there are some minor exceptions to that..

Re: complex question for all you guru's
 

Got it, I guess I need to try to find some pictures of some planetary style center diffs.

Re: complex question for all you guru's
 

Look up Torsen diff's, they're probably the most common type of center diff.