Feral, glad you got the sketch. I always find it hard to describe mechanical assemblies on the web
1200rallycar - now you can officially give Feral heaps because he doubted you. Hahaha.........
Come to think of it, I used to think these things were a waste of time too. Hmmmm maybe I'm in for it too

Oh well, we all live and learn

Paul,
Thanks for the drawing you sent. Previously the way it was explained, it sounded like the springs were the only thing providing the force on the axle gears. This just would not work.

I see now they use the same principle as a normal clutch pack LSD. The cross pins run up the ramps in the block as torque is applied. This forces the block against the side gears.

I can see they would work with lower power applications.

I APOLOGISE PROFOUNDLY FOR DOUBTING YOU 1200 RALLY CAR!!!

its exerting constant pressure against the side gears, and that will help create a limited slip effect, but can't positively 'lock' anything together. Its totally dependant on the side gears gripping the carrier, there's no mechanical locking device in the unit.

I didn't mean to insult with the 'low power' remark. In the grand scheme of things, a high powered 1400cc engine is still a relatively low power car.
That Swift was a rocketship even when it was naturally aspirated, and spun to over 9000rpm. The PG worked well for him until he turboed it, with over 240bhp. All of a sudden the PG had absolutely no effect. If it had a method of mechanically locking the centre, it would still have worked - it might have broken, but it would have worked...

I'm not saying they're sh*!, they do work in low power applications, but they aren't a locker.

L18_B110,
The only time I have herd of these things not working as a locker/unlocker is when the production tollerances are not exactly suited to the diff centre or the torque they have to take is too great. They can sheer the diff centre drive pin if the torque loadings are too high resulting if complete diff failure.
I have a high powered 1397cc/stroked/efi/close ratio gearbox etc etc that is puts out heaps more grunt that a stock A12 but again I say that the limiting factor to the amount of grunt I can put to the ground is the road surface and tyres - in my case dunlops in the dirt. Things might be different if I were competing in hillclimbs etc.
I have herd of other failures like the diffs remaining locked and not unlocking or the diffs remaining unlocked and not locking but I believe it all has to do with accurately designing and fitting the spring loader locker plates to each and every individul diff centre.
I have spent a long time getting all this right and I'm glad I did or I might be one of those production customers that has nothing good to say about them. All I know is that I'm glad I spent the time doing mine right because it is doing a better job than I thought it would

i know the dmwdave referred to in the linked article, and Ive heard his account of these PG devices. He holds some records at MtCotton Hillclimb here in a couple of classes with his Swift. They do work OK with lower powered cars. But when he turboed his swift engine, the device was totally ineffective and he reverted to a locked diff (yes in a fwd).

It can't possibly act as a locker, but it can provide a LSD action.

All diffs (except detroit lockers) always provide a 50/50 torque split between both axles - even open centres. In an open centre, when one wheel looses traction, it limits the amount of torque the engine produces (equal and opposite forces and all that...). Both wheels are still getting the same torque applied to them, but its not enough to do any really useful work through the wheel that still has traction.

what LSDs do is artificially add resistance to a wheel loosing traction to increase the amount of torque the engine can produce and apply to the ground., so that some useful work can still be done to drive the car forward.

locked diffs (welded) provide a 50/50 torque split too, but their limiting factor is the wheel with the most available grip.

Im sure Im willing to risk a weekend or so and a few beers to make one up just for the heck of it just to see if it makes the same difference to my open diffs...a rsk Im willing to take

Feral,
I determined that it was locking and unlocking by simply sticking the car in the dirt, parking it, then dropping the clutch. Consequence was two wheelie marks in the dirt and not one. This characteristic remains right through the gears. Obviously I kept an eye on this action during my testing/rally and it remains consistant. This implies that it is locked.
I can tell that it is unlocking simply by turning the car. On tarmac with the old CIG locker one tyre would always cherp and chatter because it is in effect turning at a different radius. I checked this out on the dirt and tarmac and there is no longer any cherp or chatter. This characteristic is something that made the car more fun to drive at speed in the dirt because once you lift the throttle the rear axles unlock and the car immediately turns inn easier, something that doesnt always happen in the dirt with a CIG locker. It's real seat of the pants stuff and is very obvious when you are in the car in the dirt at speed. Please believe me when I say that it is locking and unlocking. I always try to be honest and have many years experience in Mech engineering.

I am certainely not calling anybody dumb.

I had to give all this considerable thought before taking on the task of designing and manufacturing one because I didnt want to waste my time and I also struggled with how they work/if they work.
How does it lock and unlock the axles - well if you imagine the normal plate type LSD. Inside thay have a drive pin arrangement, imagine that there is only one drive pin like inside the H165. Back to the LSD for a minute - inside there is a plate carrier that, when the drive pin exerts force on the plate carrier the plate carrier is forced away from the drive pin to exert pressure on the plates thus providing some driving force to the normally undriven axle. With the spring loaded locker it works in exactly the same way except there is no multi plate arrangement. Imagine a plate type LSD without the multi plates, when you plant your foot the plates of the spring loaded locker act in exactly the same way as the multi plate carrier inside the LSD. The centre drive pin of the diff exerts pressure on the spring loaded locker plates forcing them apart and exerting pressure on the normally undriven axle. It is the force being applied to the spider gears (gear that drives the axle) inside the H165 by the spring loaded locker plates that transfers drive to the normally undriven axle. That is why I originally thought the spring loaded locker was a waste of time, I didnt think there would be enough surface area on the face of the spider gears to apply enough load to lock and unlock the rear axles.
Keep in mind that in my application the amount of traction I need is limited because it is determined by the amount of grip my rear tyres give me on a dirt or loose surface.

Gees it's hard to explain over the web


I am not sure I understand your last question in your post so I wont speculate.

Sorry for the long post and I hope the above makes sense to you.

Paul.

Quote:

How did you determine this was happening?

Call me dumb but i cant see what makes the spring loaded plate lock and unlock?

How does it lock the axle?
More importantly, if it locks under load in one drive direction, how does it unlock in the opposite drive direction.

OK converted.
When I have the time I will post the "how to do" in the tech section. I am flat out at present and expect that I will be that way for a while, so just be patient and I'll get around to it.
It will be a bit hard to explain the machining processes over the web, and only those fimiliar with machining techniques will understand. I will try and keep my explanations simple for all to see.
There is always the option of going to "my photo's" and studying the photo's of the spring loaded locker plates that I have already posted.

Paul.

OK next question is can we have the method as to how to build one posted on the site in the tech section? As most of us use a H165 with varying ratios anyway....my application will be circuit racing.