How much torque across an axle would be = to locked.

Dave_w

Member
I have a clutch pack locking diff that I can preset 5 different levels of lock.

If I have an engine output of 600Nm and a diff ratio of 4.55:1 a low range of..., in 1st gear and 34" tyres with a grip level of...... Arhhhh. It is all too hard.

Would anyone know what would be a torque value across the diff that would act like a locked diff ? To reduce the stress on parts, I would like the least amount of preload that acted like a locked diff.

It will have 0-5 in the settings. #0 no lock. #5 locked. I thought I would start #1 pretty soft with around 100Nm and progress up to 5 in increasingly finer steps, but what should 5 be ?
 

Kippie

Well-Known Member
The main challenge you have is that the friction between your tyres and road surface will vary all the time. It's not constant whereas your clutch setting once setup is constant. It will slip on certain road surfaces or it will bind on others. So how long is a piece of string?
 

Dave_w

Member
The main challenge you have is that the friction between your tyres and road surface will vary all the time. It's not constant whereas your clutch setting once setup is constant. It will slip on certain road surfaces or it will bind on others. So how long is a piece of string?
I guess I need to match the maximum friction/loads i.e. where one wheel is off the ground and the other has maximum grip.

or

match the amount of torque needed to move the car up a rock climb where one wheel is off the ground, anything more is probably unnecessary and may lead to breaking stuff.

I was thinking around 1,250Nm ?

Too lose and it will generate a lot of heat.
 
Last edited:

Dave_w

Member
Is that switchable or permanently on?
Electrically switchable on the fly but I have to set the amount of lock for each of the 5 positions by down loading settings to the software that drives it, once downloaded I can choose 0-5 via a console switch. It will take 4-5 seconds to change from any position to the next and you can skip or not any position.

If the settings are wrong I can download a new set of 5 values but not on the fly.
 

Kippie

Well-Known Member
A locked diff makes the axle rigid and both tyres rotate at the same speed irrespective of whether or not they have grip. Shouldn't you therefore set your slip diff to its maximum setting so that it simulates a locked diff when switched on? Why would you set it at a lower value which may cause the tyre with grip to remain stationary whilst the one without grip spins freely?
 

Albynsw

Well-Known Member
I was thinking the same as Kippie. Locked is 100% locked with no slippage the same as an Air or E locker. I can’t see any advantage anything less
 

Dave_w

Member
Most locked diffs are mechanically engaged (dog gears or locked planetary gears) and are not under any or much load to stay locked. This one drives a motor that spreads two plates by ramping ball bearings between the plates. The spread plates put a lot of pressure on the clutch pack to stop them slipping. This pressure remains while the diff is locked. The mechanical sympathy in me says to have the least amount necessary to perform the task.
 

Chatty

Well-Known Member
A locked diff makes the axle rigid and both tyres rotate at the same speed irrespective of whether or not they have grip. Shouldn't you therefore set your slip diff to its maximum setting so that it simulates a locked diff when switched on? Why would you set it at a lower value which may cause the tyre with grip to remain stationary whilst the one without grip spins freely?
I was thinking the same as Kippie. Locked is 100% locked with no slippage the same as an Air or E locker. I can’t see any advantage anything less
I'm thinking (and I stand to be corrected) but this sounds like a fancy form of a Limited Slip Diff, rather than a locked diff.
If I recall properly the old Slippy Diff (LSD) allowed some rotation of the wheel without grip, while transferring most power to the wheel that wasn't slipping.

Reading between the lines, it looks like this electronic marvel allows you to set 5 values of allowable "slip", measured by torque settings.
Why you would want this on a normal 4WD eludes me, it sounds like something that's really meant for competition use.

And when I see the words "clutch pack" in relation to a diff, I really start to get cold sweats - clutch packs aren't something I would associate with a long life in a diff. We have enough trouble not blowing up big lumps of finely machined steel in a diff...
 

Chatty

Well-Known Member
@Albynsw and @Kippie - there's a great explanation here Plated LSD explained

For me the key words are right near the end of the article:
The chosen pressure angles depend on many factors, such as car weight, dimensions, engine torque, racing purpose and many other.
Unfortunately there is not such a thing as a basic formula, giving you the best setup possible. The setup is car, track and driver dependant.


The most important words being "racing purpose" and "track" - these diffs look like they are meant for motorsport use...
 

Petunia

Well-Known Member
To me, logic denotes 5 is totally locked as hard as designed to go, then 0 or 1 to 4 are of lesser values.

one wheel in the air and the other impossibly going up a rock step, the one in the air will take 0.8 frogs per square inch to turn, the one trying to climb will take 800 frogs per square inch, anything less than 100% lock will most likely still try to turn the easy one, and/or grind the bejeesus out the clutch pack on the hard side? [you trying to rotate/lock the hard one, not the easy one]

If the internals are designed to exert pressure onto the clutches, that should be designed to handle that pressure, yes mechanical sympathy is ok but on the fly just turning it on is not mechanically sympathetic unless there is no load.

If you go ahead, come back in a few years and tell me it did or did not work?:cool:
 

Kippie

Well-Known Member
@Albynsw and @Kippie - there's a great explanation here Plated LSD explained

For me the key words are right near the end of the article:
The chosen pressure angles depend on many factors, such as car weight, dimensions, engine torque, racing purpose and many other.
Unfortunately there is not such a thing as a basic formula, giving you the best setup possible. The setup is car, track and driver dependant.


The most important words being "racing purpose" and "track" - these diffs look like they are meant for motorsport use...
Years ago I used to race RC cars. It's standard practice to install an adjustable LSD in the rear to improve handling coming out of corners. The tighter the LSD the more throttle you could give without the inside rear wheel losing traction, but the harder it is to steer the car as the rear wheels push the car straight ahead.
 

Dave_w

Member
@Albynsw and @Kippie - there's a great explanation here Plated LSD explained

For me the key words are right near the end of the article:
The chosen pressure angles depend on many factors, such as car weight, dimensions, engine torque, racing purpose and many other.
Unfortunately there is not such a thing as a basic formula, giving you the best setup possible. The setup is car, track and driver dependant.


The most important words being "racing purpose" and "track" - these diffs look like they are meant for motorsport use...
Hi Chatty

There are clutching diffs used in many road cars and 4wds, they did use them racing as well. My off the showroom floor Subaru Impreza and Nissan Navara both have them. They can be noisy at slow speeds but can be assembled to be aggressive or gentle. The one I'm talking about here works on the same principal but is not activated by wheel speed differences. You can set the amount of torque needed before one wheel rotates at a different speed to the other by choosing a setting.

Aggressive ones used for competition do require regular resets of preload but for a daily driver, maintenance is very low.

All

For me, I will consider P5 to be locked if it resists the amount of torque that the engine can deliver not including inertia. Beyond this I want it to slip as this will reduce shock loading on the running gear through axel tramp etc..

I expect it will be so much better than traction control and better than a straight locker.

Also if I set P5 to whatever the maximum amount of lock thats possible it maybe that it is twice as much lock as needed, then what do I set P4 to?

Ideally P5 will be just enough lock and P1-4 will be progressions between P0 and P5.

I guess I'll adjust it so I can put one wheel in gravel and one on the bitumen and flatten it without spinning in the gravel add some % and work from there?
 
Last edited:

Lost1?

Well-Known Member
It sounds like he has an electronic LSD he wants to make into a Detroit locker style centre. There will always be a shock load as the clutch engages is my thought. How much depends on how progressive the clutch pack is as it engages or if you are building revs as the clutch pack begin to engage.
 
Last edited:

Albynsw

Well-Known Member
I expect it will be so much better than traction control and better than a straight locker.

It sounds like a compromise of both of those systems to me. Unless you have the setting just right for the conditions which sounds like it will always be a guessing game to get right it will be almost counterproductive
 

cam04

Well-Known Member
Engine output nm isn’t a factor. The torque is created at that end of the car by the tyres not slipping on any given surface and that equation is constantly variable. The maximum torque setting is equal to the breakout nm required to spin whatever tyre you have on a high traction surface. The engine can either do it or it can’t but once it’s spinning it doesn’t matter if you have 200 or 2000 hp. Unfortunately you only have 5 settings. Infinitely variable hydraulic and electric clutch packs are common as horseshit these days in diffs. Speed off road is better achieved by sticking your 5 setting diff into the centre diff position of a 4wd and vary the fore/aft torque split like rally cars. STi Imprezas have two centre diffs so that the driver can select the split. Cross axle torque management is much better sorted out automatically.
 

Dave_w

Member
Engine output nm isn’t a factor. The torque is created at that end of the car by the tyres not slipping on any given surface and that equation is constantly variable. The maximum torque setting is equal to the breakout nm required to spin whatever tyre you have on a high traction surface. The engine can either do it or it can’t but once it’s spinning it doesn’t matter if you have 200 or 2000 hp. Unfortunately you only have 5 settings. Infinitely variable hydraulic and electric clutch packs are common as horseshit these days in diffs. Speed off road is better achieved by sticking your 5 setting diff into the centre diff position of a 4wd and vary the fore/aft torque split like rally cars. STi Imprezas have two centre diffs so that the driver can select the split. Cross axle torque management is much better sorted out automatically.
Almost.

The engine has the potential to make torque, you can't have torque without resistance (push back), resistance comes from the grip and drag. So 100% right you can only have as much torque as you have grip. So logically I should have the clutch set up to match or close to the maximum level of grip as I said in post #3.

Rally cars, particularly the Subaru's have one centre diff, the earlier ones had an open centre diff with a torque split of 40/60 and a system almost exactly the same as this diff that allowed software to constantly vary the amount of slip, it wasn't a seperate diff it worked with the existing. The later cars had a helical centre diff with a 35/65 split and a similar clutch pack to support the existing centre diff. They went from open to helical because the street cars have a crap LSD in the front that didn't work, if you lifted a front wheel exiting a corner it spun up, the centre diff electronics saw a loss of traction across the front and removed the electronic lock. So there you are full noise exiting a corner and it suddenly switches to, what feels like rear wheel drive. The car instantly sends you bush.

To race these, a plated front diff with a preload solved this as does a Motec diff controller. Subaru did not consider a plated front diff was suitable for daily drivers so they put in a helical front which worked pretty well.

This rear diff works very like the Subaru's centre diff, is a remote selectable, variable slip, clutch plate differential from a Porsche Cayenne. The pressure on the clutch plates is achieved by a PWM electric motor. The electric motor rotates a plate that is separated by ball bearings from another plate on the same axis. The driven plate has freedom of movement around it's axis but not along the axis. The slave plate has freedom of movement along it's axis but not around it's axis.

The ball bearings between these two plates locate in ramped grooves. The groves are such that when the driven plate is rotated from rest the ball bearings roll up the ramps and force the two plates apart.

The slave plate backs onto the clutch plate assembly and as it is pushed apart from the driven plate it compresses the clutch assembly. The harder you drive the electric motor the greater the resistance to slip across the diff.

I can easily run it on/off but it will be better to have a variable level of lock.

I hope to run 6 options in one program.
0=Off
5=Locked or sufficiently tight it behaves like it was locked.
1-4=steps inbetween.
 

Dave_w

Member
It sounds like a compromise of both of those systems to me. Unless you have the setting just right for the conditions which sounds like it will always be a guessing game to get right it will be almost counterproductive

In sand, towing a couple of jetskis over a ton the TC steps in and brakes a wheel, not what I want. I also don't want full locked as it will churn the inside wheel. Optimal would be a % of lock, enough to stop one wheel spinning but enough slip. There are also times open and locked would be nice. I'm also sure it will drive better on a faster gravel road with a bit of lock.

I may be wrong, but I'm confident I'm not :)
 

Dave_w

Member
Why not tell us what diff lock you have so answers can be tailored to suit your particular application.


I cop enough shit now without telling people it's a Porsche :)

The thread has departed from my question so it's just an interesting discussion now about wether or not it's a good idea.

My question was, how much torque across an axle would simulate "Locked", I think we've all decided it's not a simple question. I'm just going to set XYZNm and see how it drives.
 
Top