C4 Tech/Performance L98 Corvette and LT1 Corvette Technical Info, Internal Engine, External Engine

How a Differential REALLY works...

Old 12-16-2012, 01:13 AM
  #1  
Tom400CFI
Team Owner
Pro Mechanic
Thread Starter
 
Tom400CFI's Avatar
 
Member Since: Aug 2004
Location: Park City Utah
Posts: 21,544
Received 3,181 Likes on 2,322 Posts

Default How a Differential REALLY works...

I've seen things posted over the years that aren't "quite right" or out right wrong, so I thought I throw up a post to clarify, similar to my "function of a C-beam" thread, right HERE.

Some of the "lore":
*"The Differential sends power to (pick a wheel/side) first, then the other side gets it second."

*"Only the (pick a wheel/side) is powered, in normal driving."

*"The (pick a wheel/side) gets most of the power."

*"The carrier drives one axle, same as an open diff would. The posi or clutch pack drives the other axle."

^No, no NO.


There is no "driving one one wheel", "one side" or any thing like that going on. Tq is applied to the ring gear, which applies tq to the differential carrier/housing. This is supported by the two carrier bearings. The ring gear is physically on the left or drivers side of the car, but it's effect on the carrier is neutral/central. The ring gear could just as easily be central, or even be on the right side. Also important is that although the ring gear is physically on the left side of the carrier, it is in NO WAY, connected directly to the left axle shaft. The ring gear's orientation toward the left is incidental. The point is the function of the ring gear is singular; to apply tq to the carrier housing.

The carrier housing holds the spider/diff gears -the entire differential itself. The tq exerted on this housing is delivered to the spider gears through their locating pin. Since they are both connected with the differential side gears, the rotation of the spider gears must exert tq on both side gears. This tq is equal. This is important...this tq is equal. There is no favoring of one side over another. It is equal.

See video here for visual...
http://www.youtube.com/watch?v=gIGvhvOhLHU

The problem comes when you have differing coefficient of friction between the tire and the road, one side to the other; apply equal amounts of tq to both wheels and what happens? The one with less friction (resistance) spins. So we install some type or tq managing device -in the case of the C4, a clutch type limited slip. How does this work? It links the two axles together through the side gears, using a set of clutch disks and springs. The maximum static tq that this device can generate (based on spring pressure, and clutch area and material) = the maximum amount of tq that can be directed toward a wheel with traction and a wheel without. Usually no more than 200 lb-ft, which is really minimal when you consider max tq exerted on a C4's stock carrier can be up to 3500 lb-ft. But it's mostly enough to mitigate the differences in traction between tires in most situations.

Given the explanation for the C4's LSD above, when might one see the clutches slipping? ONLY (and always) when one rear wheel speed (not tq)is different than the other; going around corners/turns -a daily occurrence or spinning only one wheel on ice/sand etc. All other times, during ALL straight line driving, the LSD is effectively static, and not functioning in any meaningful way.

So where does this "lore" about power going to one wheel come from? I don't know but maybe it's from someone observing a stuck car with one wheel spinning. "Power is only going to that one wheel!" Or maybe it's from older drag racing platforms with solid rear axles; drive shaft tq is exerted on the rear axle housing itself (up to 1000 lb-ft in a stock C4). This would cause the right rear tire to "unweight" and the left rear tire to have added "weight", until the body rolls in the opposite direction from engine/trans tq, and equalizes on the springs. SO, the tendency in a solid rear/leaf spring car like that would be to see the right rear tire spin first, every time. The stuff "lore" is made of...

Last edited by Tom400CFI; 12-16-2012 at 01:20 AM.


Thread Tools
Search this Thread
Quick Reply: How a Differential REALLY works...



All times are GMT -4. The time now is 03:57 AM.