×

Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

# power distribution through a diff?

### Tech Articles

I'm wanting confirmation to see I understand this. Lets say for the sake of keeping things easy, a 200Nm engine... An open diff powers one wheel at a time so that 200Nm would be pushed through to that one wheel. (conveniently overlooking parasitic losses from the transmission and such)

If we have an LSD, at 100% lock, would that distribute that 200Nm to 100Nm per each wheel? So in theory... a diff with a LSD would have a less likely chance of snapping axles than an open diff. again overlooking other potential failure points.

The first line shows you're misunderstanding what happens.

While there are some excellent guides and tutorials on-line, the basic principle is...

First, the gear train, which a differential is, transfers torque, with the power being derived from the speed/rpm of the torque applied.

If there is, say, 200N.m of torque being passed into the differential (axle assembly), and the diff' ratio is 4:1, then there will be 800N.m of torque provided by the differential housing assembly, within the axle housing, to be distributed to the axles.

This torque from the housing is passed equally, through planetary gears, to the axles themselves - each receives 400N.m of torque. This torque is equal, regardless of how much there is entering, they get half each.

Where this gets confusing for many is when the torque to the axles exceeds the tyres' traction and the tyre starts to slip, or spin. If, we use the earlier values of 400M.m , but the tyre grip is limited to 350N.m, both axles will be transferring 350N.m - 700N.m total.

Further confusing people, "power" is NOT directly comparable to "torque". The engine torque curve is not the same thing as the power curve.

With the wheel slipping, the engine will increase rpm to the point where the torque curve falls to the point where only 175N.m is being provided to the diff', and the wasted 'power'is the difference between the 'road' speed of the spinning tyre and the tyre speed.

Consider a stationary 'burnout' - one axle/tyre is applying the 350N.m, but because it isn't moving, it has no 'power'passing through it, but the other, which is also getting 350N.m of torque, IS rotating, and therefor is producing power that is dissipated as heat, noise, etc.

With a conventional rear wheel drive, live beam axle car there is a torque reaction that tries to lift the rear right (compressing the spring, which is why that corner drops) and push down the rear left, and this means less potential grip to the right tyre, which is why it generally slips first. If you're turning hard left, there can be enough load/weight transfer to overcome this, and if the left's grip is low enough, it will spin instead. In between, if they're about equal, you may find the rear stepping out with little warning - epecially on damp roads!

There are different types of LSDs, with different characteristics, but they all have one main principle - some way of adding torque to the tyre with the better grip.

What you're describing is similar to a "spool", and under perfect conditions they will both have, using your figures, 100N.m. However, because they're fixed together, if there is any variation in grip that can't 'handle' the "100N.m" supplied, the other axle will receive the difference. Take an extreme condition where one tyre is completely off the ground, the other will receive ALL 200N.m. With a conventional open" diff', the maximum that can possibly be applied to the axle is half the torque.

It depends on the specific circumstances, but not only is the 'spool', or 100% locked axle more likely to break an axle, it's not uncommon to snap both at the same time because the second was over-loaded.

Worse, it's possible to actually have more than your 200N.m going through the inner loaded axle when turning under drive, because the outer can be reverse torque loading the axle as it's being forced to move faster by the chassis, and the inner is holding it back... I think I got that the right way round, it's been a while and may be the reverse, but principle applies.

what a response Gord! thank you.

I definitely over simplified my question, but I didn't want to go too far into the weeds. What technically brings me here is I'm in the process of rebuilding the rear axle on my car. Let's say that a set of axles claim to "support 300hp". Now at first I thought this was vague but Gord's mention of torque multiplication based on final drive ratios really got me thinking what a slight of hand trick that marketing is.

For this rebuild I was looking at a Kaaz or OS Giken LSD. Not even considering a spool or welding spider gears. does a clutch type LSD have the same risk of 100% of the load reaching one wheel under your one tyre off the road example?

Yes, you have to be very wary of people advertisings things using "power", especially when it's regarding torque sensitive components. Another example is with clutches, as 300hp at 2626 rpm is 600 lb.ft, at 5252 it's 300lb.ft, at 7878 it's 200lb.ft. - that's a heck of a lot of difference as may be found with a diesel, a large N/A, or a high rpm smaller engine, respectively.

No, most 'plate', or 'clutch' use a wedging action to apply a clamping force to clutch packs that are used to lock the axles together, but it will never get to that point.

The biggest stresses on axles are when the clutch is dumped from standstill, hard gear-shifts, and jumps/bumps where the wheel leaves the ground and the rpm rises before is hits the ground and drags the engine down again. When stickier tyres and/or more weight is added and/or close ratio gearkits are used the loads can be higher if before the tyre(s) lost traction as that limited the torque able to be transmitted, but now higher torque and/or shock loads can be put through the axles.

Depending on the vehicle, CVs/universal joints may be a weak point, others may have other issues, such as differential housings failing.

Again, depending on the vehicle, a broken axle may allow the wheel assembly to slip out of the housing, and if that's a serious risk there may be fully floating conversions available. For example, the infamous "C" clip.

We usually reply within 12hrs (often sooner)

#### Need Help?

Need help choosing a course?

Experiencing website difficulties?