PDA

View Full Version : [Magic inside!] Need Help With Differentials



DeathMetalRacer
25-09-2017, 21:17
Even with the help texts I'm confused about which ones I would want turned on/off and why. Could someone please explain. The diff setup in PC1 was so much easier to understand.

Jussi Karjalainen
25-09-2017, 23:19
EDIT: Clutch type differential calculator here: Project CARS 2 Differential Calculator v0.9 (https://docs.google.com/spreadsheets/d/1eFytaUotZfmSn_NR_YflMQkYJ3HxSiojbXgblJu7xYA/edit?usp=sharing) You can't directly edit this one, it's for sharing. To use it save a copy of it to your own drive and use that.



It was easier to understand in pC1 because it was vastly simplified from how they really do it. =)

OK so we have a few different differential types available to you, which you can (depending on the car) turn on and off at will. You can (and we have) also combined different differential types for more dynamic behavior and to better represent the behavior of some of the more active differentials found on modern cars:

1) Spool. This is basically equivalent to welding the driveshafts together, no differentiation possible, both driven wheels will always rotate at the same rate (within reason, there is some flex modeled in the driveshafts). If you turn Spool on, no other differentials will mean a thing, since it will solidly lock the two sides together.

2) Geared LSD. This is your torque sensitive type differential. It allows the wheels to rotate at different rates as long as both sides can handle the torque thrown at them. When one side can't handle the torque (i.e. threatens to slip) you start to get locking effect and more torque sent to the side that can handle more torque. The higher the bias ratio, the more locking you get, roughly following this curve (thanks to Torsen differentials for releasing them):

240561

So for example power bias of 2.0:1 equal roughly ~30% accel locking. This locking percentage is roughly equal to the old style of adjusting diffs in pCARS 1, though since the principle of the diff is different and our differentials work more accurately than before when transmitting forces, you'll get somewhat different results.

3) Clutch LSD. This is your traditional LSD that you'll find in other games and in pCARS 1, though the way you adjust it is closer to how they do it in real life. You have 4 controls:

Preload: Works exactly as in pCARS 1, this is the always on minimum amount of locking in the diff. If you find that the car is unstable between the transitions from throttle to off-throttle, or the other way around, you might want to increase this.
Clutches: The number of clutch faces in contact with each other. This is effectively a multiplier, the more clutch faces you have in contact with each other the more overall locking you will get for both power and coast side.
Power ramp: This is the ramp angle for the power/acceleration side of the diff. The higher the angle, the less locking you get. Lower angle leads to more locking.
Coast ramp: This is the ramp angle for the coast/deceleration side of the diff. Works just like the power side, but affects off throttle behavior.

The insides of the clutch differential look a bit like this (thanks to intothered.dk):

240562

You don't really need to fully understand the principles of how they work though. Basically you can use the ramp angles to select how much locking you're getting out of the clutches for each direction (20 degrees = maximum, 90 degrees = minimum = none) and the numbers of clutches will then increase or decrease the total locking force.

Summary: Low angle = high locking. High angle = low locking. More clutches = more locking. Less clutches = less locking.

A differential calculator for the clutch LSD will be included in my upcoming Project CARS 2 Suspension Calculator, which will convert the settings into numbers similar to what you saw in pCARS 1.

4) Viscous: A viscous type differential, that works on the speed difference between the two wheels. The higher the difference in wheelspeed, the more locking effect you get. Because it doesn't care about input torque at all and scales with speed, the effect is quite progressive and even subtle, but on the other hand it requires that one side slips before it offers any locking. We're using the industry standard measurements, showing how many Nm of locking torque you get for 50 rpm difference between the two axles. Higher number = more locking.

5) Ratcheting: This is a funky type of differential that was used way back in the early days of racing and in very limited capacity in more modern times, usually as a mandated differential rather than by free choice. It offers full lockup like a spool on acceleration, but is completely open on deceleration. So you get the possible clunkiness of a spool when accelerating and the instability of open diffs on deceleration. Trans-Am in the 90s was one series that mandated these differentials. Often known as the worst kind of differential for racing, apart from an open differential. =)


EDIT: On the center differential of some cars you also have a setting for rear power distribution, meaning you can adjust how much of the power is being sent to the rear wheels. Note that this might seem a bit high on some of the cars, for example the Loose and OEM setups on the Audi R8 V10+ are set to IIRC 90-95% rear torque distribution, instead of the 15/85 or 30/70 Audi has claimed at times. There's a reason for this though, and it's basically that front/rear torque distribution is not a simple thing that you can just put an easy number on, it's constantly variable and pretty much never static.

In reality most of these types of cars cars (including Audi, Lamborghini, Nissan, and others) have a normal RWD driveline, where all the power goes to the rear. Then they add a connection from the rear differential to the front differential, with either a viscous, electronic or mechanical differential in between. So what you have is a rear wheel drive car that then transfers a bit of torque to the front based on front and rear differential gearing differences and slip differences between the front and rear. The rear power balance value can be used to adjust for the different front and rear final drives (so that some torque is always sent to the front), while the locking settings will affect the dynamic behavior.

By adjusting the rear power balance setting as well as the differential properties of the center diff, you can simulate different kinds of "modes" that modern active systems have for different road conditions, or even properties of older static viscous based ones. 0% rear power balance makes the car a pure FWD, 100% makes it a pure RWD, assuming there's no other locking applied (locking will still transmit torque towards the other side). Yes, this means you can make an FWD Lambo if you really really want to... =)

EDIT2: To further explain the "front and rear differential" bit: If you make the front differential's ratio longer than the rear differential's ratio, when the tyres are rotating at the same speed the driveshaft to the front differential wants to rotate slower than the rear differential. If there's for example a viscous locking connection between the two, the faster spinning rear will cause the viscous LSD to transmit some torque to the front. The bigger the difference in ratios (or the higher the locking in the viscous LSD), the more torque will be sent to the front in all conditions.

MXR SMILER
25-09-2017, 23:52
Quality, thanks... Should be a sticky for this...

DeathMetalRacer
26-09-2017, 00:15
Jussi, can't thank you enough for your response!

I do have a couple more questions if you all don't mind.

1. If I'm using geared LSD and my car wants to spin out at corner exit do I want to increase or decrease power bias? What about spinning at corner entry? Would I increase or decrease coast bias?

2. In PC1 I learned that if my car felt unstable at corner entry or corner exit I could decrease preload to stabilize the car. Does this still apply in PC2?

Jussi Karjalainen
26-09-2017, 00:21
1. Generally less power bias means more stability on throttle (you're allowing the inside rear to spin more, transmitting less torque to the outside rear), generally more coast bias means more stability off throttle (you're locking up more when there's a grip difference).

2. Weird that corner entry would be more stable with less preload, usually more locking = more stability when you're not on the throttle and are turning, but preload effectively works like it did before, just more accurately physics simulation wise.

Aldo Zampatti
26-09-2017, 00:31
Quality, thanks... Should be a sticky for this...


Jussi's master class is too good to go unnoticed it. I concur.

Sticky

DeathMetalRacer
26-09-2017, 00:32
I wonder if it makes a difference that I race in game pad as to why lowering preload helped me gain stability out of the car.

Jussi Karjalainen
26-09-2017, 01:02
Possibly, and would of course depend on what it was to start with.

Cheers for the Sticky btw.

OperatorWay
26-09-2017, 08:38
Thanks for that!
This old "How Differential Gear Works" video is also a good easy-to-understand introduction to the basics:
http://www.youtube.com/watch?v=K4JhruinbWc

Roger Prynne
26-09-2017, 09:18
Another quality post from Jussi, I don't know how you do it mate.

Bealdor
26-09-2017, 09:28
Another quality post from Jussi, I don't know how you do it mate.

240594

blausven
26-09-2017, 09:28
So, Power Ramp with 90deg means spool, when accelerating?
And, 0deg in Coast Ramp means both wheels turn independently?

To give more grip when accelerating out of corners, set Power Ramp to e.g. 60deg?
To stabilise the car when going off-throttle, set Coast Ramp to e.g. 30deg?

Or is it reversed, not 60/30 but 30/60?

Ramjet
26-09-2017, 14:26
Great read and also that old educational movie is Gold, they just don't make them like that anymore ... makes it simple to understand by compare. cheers Jussi and OperatorWay ! :D

Jussi Karjalainen
26-09-2017, 18:13
So, Power Ramp with 90deg means spool, when accelerating?
And, 0deg in Coast Ramp means both wheels turn independently?

To give more grip when accelerating out of corners, set Power Ramp to e.g. 60deg?
To stabilise the car when going off-throttle, set Coast Ramp to e.g. 30deg?

Or is it reversed, not 60/30 but 30/60?Opposite, 90 degrees is fully open, 20 degrees is maximum locking (how much that is depends on how many clutches you have).

The engine's torque goes into the differential. For the clutches to press together, there needs to be a sideways force. 90 degrees has two flat surfaces pressing on each other, so all force goes straight together, not creating any sideways force at all. When you add angle, some of the force applied acts sideways, causing locking. 20 degrees is the minimum practical angle (any less and the shapes start getting too sharp), which gives you the most sideways force and gets the most locking out of the diff. Refer to the previous post showing the insides of a clutch LSD.


For more stability on off-throttle, go with a low angle to maximize locking. This will prevent the wheels from turning independently so easily, and makes turning overall more difficult.

For more stability on acceleration, go with a high angle to minimize locking, up to 90 degrees for fully open. This will allow the inside rear to spin and not transmit any extra torque on the outside rear, making it harder to get power oversteer. On the other hand you can't really get that much traction either, since you're limited by the inside rear.

Brado23
26-09-2017, 22:50
As a general statement, is the following correct whilst turning?

More locking = less stability on throttle, more stability off throttle
Less locking = more stability on throttle, less stability off throttle

Jussi Karjalainen
27-09-2017, 01:48
As a gross simplification, yes.

Jussi Karjalainen
27-09-2017, 02:03
Posted a link to the Project CARS 2 Differential Calculator v0.9 in the original post, check it out. It's Google Docs based, so everyone with a Google account can use it.

http://forum.projectcarsgame.com/showthread.php?52783-Magic-inside!-Need-Help-With-Differentials&p=1367069&viewfull=1#post1367069

senn
27-09-2017, 06:48
why do some AWD cars have the centre diff as Geared AND Viscous turned on with the default setup (trying to emulate a haldex maybe??), can they infact work in partnership in pcars2 or is this just going to causes glitchy weird handling?

Gix916
27-09-2017, 11:22
should the diffs only be on one at a time. ie. if you have Geared LSD on, then switch Clutch LSD on, should the Geared LSD turn off?

and should you be able to turn them all off?

Jussi Karjalainen
27-09-2017, 12:56
OK so we have a few different differential types available to you, which you can (depending on the car) turn on and off at will. You can (and we have) also combined different differential types for more dynamic behavior and to better represent the behavior of some of the more active differentials found on modern cars:

Some parts of the geared diff model are also used for the front/rear torque distribution, which is why they're active in many AWD cars, even when the bias ratios are set to 1.0 (= open).

Whether or not you should be able to turn all of them off depends a bit on the car, for example I don't think you can turn off the center geared diff on the cars with unequal torque distribution, at least not most of them.

Gix916
27-09-2017, 14:29
the only car that i've felt the need to tune so far has been the Renault Clio, on the tuning setup for that you can turn all the diffs on at the same time or all of them off. I wasn't sure if that was meant to happen or not.

so if a geared diff is driven by gears, a clutch diff uses plates and a viscous diff uses plates in fluid, a spool is a complete locked up diff, and ratcheting is diff fully on or off, how can a car have all of them? shouldn't this default to what type of diff the car actually uses?

sp3ctor
28-09-2017, 00:05
I wonder if it makes a difference that I race in game pad as to why lowering preload helped me gain stability out of the car.

I agree with this. The first thing I would do in PC1 when a car felt skatey was to lower the preload and viscous and, in most cases, turn them off, and the car would feel less unstable. Trying to figure out what to do now because I'm assuming that isn't optimal... however, I have noticed when I turn preload all the way down it has the same effect on feel in PC2 but I'm not sure it helps times.

Jussi Karjalainen
28-09-2017, 02:13
Hey, you do what works for you. =)

Aldo Zampatti
28-09-2017, 02:15
I wanted to take the opportunity to publicly thanks Jussi for spending so much time here since he (as every other Dev) are probably stuff up with work!.

I enjoy reading your technical post since 2011, so I hope everyone else does as I do. :)

DeathMetalRacer
28-09-2017, 03:00
Any chance someone could take the time to explain how to use Jussi's diff calculator? I downloaded the spreadsheet from the link but having never used a google doc spreadsheet before I'm not exactly sure how to use and read it correctly. Thanks!

Aldo Zampatti
28-09-2017, 03:08
Any chance someone could take the time to explain how to use Jussi's diff calculator? I downloaded the spreadsheet from the link but having never used a google doc spreadsheet before I'm not exactly sure how to use and read it correctly. Thanks!

Sorry buddy, Einstein died long time ago, and he was my only hope to understand that.... 2nd option is a PhD in standford but I can't afford it :P

/jk

DECATUR PLAYA
28-09-2017, 04:23
I wanted to take the opportunity to publicly thanks Jussi for spending so much time here since he (as every other Dev) are probably stuff up with work!.

I enjoy reading your technical post since 2011, so I hope everyone else does as I do. :)

^I 2nd this. Have always enjoyed Jussi's work. Was the perfect person to promote.

Jussi I have a couple questions. I'm good with the rest of the setup but right now I'm absolutely scared of the Differential and Dampers.

1) Does power ramp = limited slip acel
Does coast ramp = limited slip decel

2) Can bumps and rebounds be explained by feel. What should I feel by making the adjustment up or down. Most of my tuning is by feel.

Jussi Karjalainen
28-09-2017, 08:32
Any chance someone could take the time to explain how to use Jussi's diff calculator? I downloaded the spreadsheet from the link but having never used a google doc spreadsheet before I'm not exactly sure how to use and read it correctly. Thanks!It's fairly simple, at it's heart it's just a converter from clutch numbers and ramp angles to the old style locking effect percentages familiar from pCARS 1.

You just fill in the numbers from the game into the green fields and read the resulting locking torque percentage from the yellow fields. A and B are just for comparing two diffs, for example the change of adding more clutches. Then you can fiddle around with the numbers to find what you need.
^I 2nd this. Have always enjoyed Jussi's work. Was the perfect person to promote.

Jussi I have a couple questions. I'm good with the rest of the setup but right now I'm absolutely scared of the Differential and Dampers.

1) Does power ramp = limited slip acel
Does coast ramp = limited slip decel

2) Can bumps and rebounds be explained by feel. What should I feel by making the adjustment up or down. Most of my tuning is by feel.1. Yes, power ramp controls the limited slip acceleration locking, and coast ramp controls deceleration locking. However the numbers there aren't the same as the old percentages. Use the calculator to get exact numbers or adjust by feel, higher number = less locking, lower number = more locking.

2. To some extent. If the car seems to wobble or sway around a lot, your damping is probably on the low side. If the car feels kind of floaty (not boaty, that's different), like it's sort of skimming over the surface of the track, your bump damping might be too hard. Another common sign of having a bit too stiff is if it has very sudden reactions. Turn the steering wheel a bit too quickly and suddenly you lose front grip and start pushing. When the tail comes out it does so in a very sudden and violent way (can be a tyre issue as well). A good setup usually has the bump damping stiff enough that the car feels steady and not too soft, while still being soft enough that you get progressive responses from the tyres. Rebounds in racing situations are usually set quite hard and it's often not that big of a deal if they're a bit too hard even. Too soft though and again you have trouble controlling the body of the car.

Of course exceptions abound, and some modern cars go with super stiff bump damping and very soft rebound damping (so total damping is about the same, just the setup is reversed) to work the tyres hard, and some really high aero cars mostly just care about making absolutely sure the aero platform is stable, and essentially lock up the suspension with very stiff damping. There are many way to do things in the suspension world...

The problem though is that both too stiff and too soft damping can cause the same kind of instability issues on the car, though through different mechanics.

I'll be releasing a damper and spring calculator later this week (might be on the weekend) and in the future will also release some new videos on how to work with it.

Until then you can sort of get a feel for what's coming by looking at some of my old videos (from before I started working at SMS) as well:


https://www.youtube.com/watch?v=1S5kSrR3oM4

https://www.youtube.com/watch?v=MZ_t2cJcknM

https://www.youtube.com/watch?v=L5Asj0KtyNg


Note that these videos are old, and the new calculator will have a different look to it due to the various added features in pCARS 2. And even if you're not interested in using the calculator I do talk about some general level damper tuning stuff in them.

hagar147
28-09-2017, 10:30
Because of the lack of a 'garage/setups' section in this forum, I am posting my question here.

In the car setup menu in Project CARS 1, there was a deceleration/acceleration diff lock setting that I can't seem to find in Project CARS 2.
Could someone tell me which value(s) I need to adjust in PC2?
e.g. If I liked a high deceleration diff setting in PC1, what would I need now?

Kind regards,
hagar147

Bealdor
28-09-2017, 10:33
Threads merged.
Pleae check out Jussi's posts here in this thread.

Cholton82
28-09-2017, 12:04
Any chance someone could take the time to explain how to use Jussi's diff calculator? I downloaded the spreadsheet from the link but having never used a google doc spreadsheet before I'm not exactly sure how to use and read it correctly. Thanks!

I've used my iPhone to download and be able to edit it , basically you can use google drive app to save it to and then google sheet app lets you edit it . Then you can delete out the numbers that are in all ready and put your own in which automatically changes the numbers at the bottom . Hope that makes sense

thedarkjack
28-09-2017, 15:18
thanks jussi, quality post!

DECATUR PLAYA
28-09-2017, 15:25
It's fairly simple, at it's heart it's just a converter from clutch numbers and ramp angles to the old style locking effect percentages familiar from pCARS 1.

You just fill in the numbers from the game into the green fields and read the resulting locking torque percentage from the yellow fields. A and B are just for comparing two diffs, for example the change of adding more clutches. Then you can fiddle around with the numbers to find what you need.1. Yes, power ramp controls the limited slip acceleration locking, and coast ramp controls deceleration locking. However the numbers there aren't the same as the old percentages. Use the calculator to get exact numbers or adjust by feel, higher number = less locking, lower number = more locking.

2. To some extent. If the car seems to wobble or sway around a lot, your damping is probably on the low side. If the car feels kind of floaty (not boaty, that's different), like it's sort of skimming over the surface of the track, your bump damping might be too hard. Another common sign of having a bit too stiff is if it has very sudden reactions. Turn the steering wheel a bit too quickly and suddenly you lose front grip and start pushing. When the tail comes out it does so in a very sudden and violent way (can be a tyre issue as well). A good setup usually has the bump damping stiff enough that the car feels steady and not too soft, while still being soft enough that you get progressive responses from the tyres. Rebounds in racing situations are usually set quite hard and it's often not that big of a deal if they're a bit too hard even. Too soft though and again you have trouble controlling the body of the car.

Of course exceptions abound, and some modern cars go with super stiff bump damping and very soft rebound damping (so total damping is about the same, just the setup is reversed) to work the tyres hard, and some really high aero cars mostly just care about making absolutely sure the aero platform is stable, and essentially lock up the suspension with very stiff damping. There are many way to do things in the suspension world...

The problem though is that both too stiff and too soft damping can cause the same kind of instability issues on the car, though through different mechanics.

I'll be releasing a damper and spring calculator later this week (might be on the weekend) and in the future will also release some new videos on how to work with it.

Until then you can sort of get a feel for what's coming by looking at some of my old videos (from before I started working at SMS) as well:


https://www.youtube.com/watch?v=1S5kSrR3oM4

https://www.youtube.com/watch?v=MZ_t2cJcknM

https://www.youtube.com/watch?v=L5Asj0KtyNg


Note that these videos are old, and the new calculator will have a different look to it due to the various added features in pCARS 2. And even if you're not interested in using the calculator I do talk about some general level damper tuning stuff in them.

Thank you Jussi now I understand why the car feels so jerky as I enter a turn. Its probably why my stability control is kicking on so much. Answers lead to more questions. So how far off are the default damper and differential settings or is it a driver preference type of thing.

I miss the bars from PCARS 1. The bars made the numbers make sense for some reason.

DECATUR PLAYA
28-09-2017, 16:02
^Example:

Ford GT GTE is a awesome ride. That car is doing some special stuff in the turns.

NASCAR/Ford Fusion feels almost perfect at Daytona. Feel like with some damper tuning I can get to glide over that bumpy low groove better.

IndyCar is super quick makes the the racing feel like it looks on TV.

GT3 way off but as expected. Being the dominant class on PCARS 2 it should be harder to find balance in this class. Found balance in the SLS because of my familiarity with it not so much in other GT3's.

Jussi Karjalainen
01-10-2017, 20:53
I miss the bars from PCARS 1. The bars made the numbers make sense for some reason.
I can see that, though they could be very misleading. Maximum soft on a slider could still be hellishly stiff, like on the Ariel Atom V8, and maximum stiff might still be relatively soft on another car. The damper ranges on quite a few cars don't make that much sense even in real life unfortunately...

Ford GT GTE is a awesome ride. That car is doing some special stuff in the turns.Yeah, that's what you get when you build a prototype and call it a GTE car... =)

Renoldo1990
01-10-2017, 21:07
Hey Jussi :)

I just tried out your Differential Calculator. It makes everything a lot easier for me. Thanks for that!
Now i'm waiting for your new Suspension-Calculator. Any update on that?

Jussi Karjalainen
01-10-2017, 21:16
Hey Jussi :)

I just tried out your Differential Calculator. It makes everything a lot easier for me. Thanks for that!
Now i'm waiting for your new Suspension-Calculator. Any update on that?Working on getting it in a publishable condition. Google just doesn't allow the kind of protection on the sheets that I'd require so I must do it the old fashioned way via Excel/OpenOffice, which is annoying and time consuming. Should have it up before tomorrow is upon us.

Jussi Karjalainen
01-10-2017, 23:21
It's out now, made a new thread for it.

honespc
02-10-2017, 20:19
Looks like the Harvard Licence is now necessary in order to first understand and then setup diffs properly.

Well, first and foremost the question that should matter. Among that incomprehensible mare magnum the diff section has become now, what are the two options that should that equal old good and simple accel/decel diffs from pc1?

Bealdor
02-10-2017, 20:23
Looks like the Harvard Licence is now necessary in order to first understand and then setup diffs properly.

Well, first and foremost the question that should matter. Among that incomprehensible mare magnum the diff section has become now, what are the two options that should that equal old good and simple accel/decel diffs from pc1?

Ramp angle in the clutch LSD.
Check out post #2 for more info about this.

Jussi Karjalainen
02-10-2017, 21:32
Well, a combination of the ramp angles and the number of clutches.

DeathMetalRacer
03-10-2017, 03:40
Pardon me being a nub head lol. I understand how to add the values and when I do I get percentages in the yellow boxes. What do I do once I have these percentages?

Jussi Karjalainen
03-10-2017, 07:12
Pardon me being a nub head lol. I understand how to add the values and when I do I get percentages in the yellow boxes. What do I do once I have these percentages?The percentages are essentially the same as the percentages in Project CARS 1 or the locking percentage in many other sims.

rich1e I
06-10-2017, 12:35
So Mr. Jussi, not sure if I got this right. Let's take the most popular racing class. In GT3s we don't need spool, as it overrides all other diff types, and we don't need ratcheting as it's an old type of diff. Is that correct?

Rikirk
06-10-2017, 16:14
I'm actually wondering why we are able to modify the kind of differential in a car - specially a racing one - when most of us are just trying to simulate the handling behavior In Real Life. Jussi: huge thank you for the suspension guide! That alone makes PC2 stand out from some of the most acclaimed competition.

Jussi Karjalainen
06-10-2017, 18:47
So Mr. Jussi, not sure if I got this right. Let's take the most popular racing class. In GT3s we don't need spool, as it overrides all other diff types, and we don't need ratcheting as it's an old type of diff. Is that correct?More or less, though you're welcome to try them out. Putting a spool on a GT3 car would kinda make it a bit like a V8 Supercar with better aero... =)

rich1e I
06-10-2017, 18:59
More or less, though you're welcome to try them out. Putting a spool on a GT3 car would kinda make it a bit like a V8 Supercar with better aero... =)

No no, I prefer my GT3 car to behave like a GT3 car actually :p Driving a V8 supercar reminds me of that joke 'oversteer is when you hit the wall with the back end' ^^ Btw thanks for being around answering questions and helping people out. Really appreciate it!

F1_Racer68
06-10-2017, 19:13
No no, I prefer my GT3 car to behave like a GT3 car actually :p Whenever I drive a V8 supercar I recall that saying that explains that oversteer is when you hit the wall with the back end^^ Btw thanks for being around answering questions and helping people out. Really appreciate it!

You mean this explanation? :biggrin-new:


https://www.youtube.com/watch?v=EQWPumtDXk0

rich1e I
06-10-2017, 20:00
Haha, didn't know that one^^

Jussi Karjalainen
07-10-2017, 03:32
I'm actually wondering why we are able to modify the kind of differential in a car - specially a racing one - when most of us are just trying to simulate the handling behavior In Real Life.There are quite a few reasons.

It's not going to give you enough of an advantage to become an "exploit" so why not?

The cars in the game have often come out with different diffs available to them. Ginetta G55 GT3 and Audi R8 LMS are good examples.

Simulation is not just about what is done in real life, it's also about what would happen IF something was done. Hence why you can put any car on an RX or snowed in track, use long gearing suitable for circuit racing on an RX car, drive a low drag bodykit Indycar or Group C car on a small twisty circuit, etc.

And we also want to show off our tech and how versatile it is... =)

cpcdem
08-10-2017, 07:33
There are quite a few reasons.

It's not going to give you enough of an advantage to become an "exploit" so why not?

The cars in the game have often come out with different diffs available to them. Ginetta G55 GT3 and Audi R8 LMS are good examples.

Simulation is not just about what is done in real life, it's also about what would happen IF something was done. Hence why you can put any car on an RX or snowed in track, use long gearing suitable for circuit racing on an RX car, drive a low drag bodykit Indycar or Group C car on a small twisty circuit, etc.

And we also want to show off our tech and how versatile it is... =)

Hmm, ok, but following that argument, then we should be also allowed to put front and rear downforce to the escorts, would be interesting to see what would happen if that was done as well. Or handbrakes to formula 1 cars :). And so many other things, but no, the sim does not allow as to do such things and in my opinion very rightly so.

It also feels very strange to me being able to change differentials like that, even combining them all together, especially seems everything else seems to be done in a realistic way. I think you guys have no need to showcase the versatility of the system like that, I think it is already more than obvious how extremely versatile this game is, in so many different areas. Wouldn't it still be just fine and more realistic if only the "correct" - but different to each - differential was used in every car? We'd still need to learn how to deal with each one..

Finally, I think being able to mix differentials like that, is inviting people to search for exploits in this area indeed, and if some people find some, I think we will never hear about them..

Anyway, just my view on this, hope it is not taken as a rant or anything, it's meant as constructive criticism. In general having an absolute blast here with PC2.

rich1e I
08-10-2017, 10:31
Agreed. Would've been better if the diff types the cars don't have in real life were just greyed out or not available. I think 'time' is the key to understand why many decisions are made. I don't know if it's a licensing thing and manufacturers expect developers to add road cars as well for promotion purposes, but it would've saved the whole SMS team a lot of time just adding half the amount of road cars to the game and instead use the time to implement the above mentioned diff details and many other things like an appropriate sound to some cars. Sound is really bad in many cases. It takes away the enjoyment to drive these cars for me.

JyriK
10-10-2017, 11:21
There are quite a few reasons.

It's not going to give you enough of an advantage to become an "exploit" so why not?

The cars in the game have often come out with different diffs available to them. Ginetta G55 GT3 and Audi R8 LMS are good examples.

Simulation is not just about what is done in real life, it's also about what would happen IF something was done. Hence why you can put any car on an RX or snowed in track, use long gearing suitable for circuit racing on an RX car, drive a low drag bodykit Indycar or Group C car on a small twisty circuit, etc.

And we also want to show off our tech and how versatile it is... =)


Well, if you replace the spool in V8 Supercar for a fancy differential, it is not a V8 Supercar anymore. I mean what's the point?

Bealdor
10-10-2017, 11:44
Well, if you replace the spool in V8 Supercar for a fancy differential, it is not a V8 Supercar anymore. I mean what's the point?

This:


...it's also about what would happen IF something was done...

JyriK
10-10-2017, 11:59
This:

You are creating a cheater car, that is faster with a fancy differential, that's what it is.

Roger Prynne
10-10-2017, 12:51
^ Why are so concerned about this as it's just an option.

JyriK
10-10-2017, 14:04
^ Why are so concerned about this as it's just an option.

Race cars have regulations, and I would think people are more interested in the real thing than some Grand Turismo fantasy option. The latter, especially in the case of cars with spools, are going to be faster on many tracks.

Renoldo1990
11-10-2017, 11:40
It's just an option yes.
But if you drive for example in a Racing-League with the V8 Supercar you can use 'unrealistic' Differential-Settings to gain an advantage.
So probably you have two choices:

1.) Setting up a car with 'realistic' differential-settings by what you probably won't compete for the top spots (for example the V8 with a Spool)
2.) Setting up a car with 'unrealistic' differential-settings, gaining better results; BUT driving a car which feels really different than in reality (V8 with more modern Differential-System for example)

rosko
11-10-2017, 12:01
Is it possible to create a cam & pawl diff in this game?

JyriK
11-10-2017, 13:17
Is it possible to create a cam & pawl diff in this game?

Ratcheting is pretty close, but not quite since it gives full lock on acceleration, where as cam & pawl gives 50-75% lock on acceleration (if memory serves).

Rikirk
11-10-2017, 14:14
To my mind allowing different kinds of differential while racing a sport class under regulations is cheating. In fact, should be restricted for the same reason SMS decided to lock the gear ratio in GT3. I understand people would like to experiment, but at that point the game should allow to build and race a custom car instead of modeling the existing one with the highest possible grade of accuracy. This is a flaw that has to be fixed for fair MP racing or TT benchmark.

Roger Prynne
11-10-2017, 14:17
SMS decided to lock the gear ratio in GT3
Just like real life you mean.

JyriK
11-10-2017, 14:27
Just like real life you mean.

That is what he said. Quite baffling that the differentials are not restricted.

F1_Racer68
11-10-2017, 15:11
Just like real life you mean.

That is exactly the point Roger. Everything else about the car tuning is allowed or blocked according to real world specifications and regulations.

Why aren't the differentials?

I have to agree with those commenting that this is an oversight/mistake in particular for MP/online/community/esports drivers or events. It gives an unrealistic ability to the cars. I won't go so far as to say it is a "cheat", but it is certainly along the same vane as an unrealistic driver aid.

Roger Prynne
11-10-2017, 15:36
Yeah sorry I miss-read the post..... been a long day.

F1_Racer68
11-10-2017, 17:19
Yeah sorry I miss-read the post..... been a long day.

Been there. Fully understand....

Jussi Karjalainen
12-10-2017, 10:11
Cars like the V8 Supercar that were limited to only spools in real life were limited to only spools in the game as well (if we missed any let us know, Porsche 935 and 962C are grey areas because while Porsche themselves only ever ran spools on them, customer cars ran other diffs too). I said that we didn't want to give options that would give an unfair performance advantage, and I meant it. The other options can only change the feel to some extent and make the car more or less suitable to you, not make a massive outright difference.

I also feel that it's quite a different thing to limit gear ratio options, since those have a very direct effect on top speeds and acceleration and are used as a BoPping tool, like how the Merc SLS can barely reach 6th gear. Differentials are often barely mentioned and they're not used as a BoPping tool.

Your feedback though is noted.

JyriK
12-10-2017, 10:17
Cars like the V8 Supercar that were limited to only spools in real life were limited to only spools in the game as well...

Note to self: Always double check before making an argument.... Glad I was wrong!

Bealdor
12-10-2017, 16:40
Moved to car setup talk.

Hilazza
18-10-2017, 12:53
So if you have a car that is unstable of throttle you want to increase the locking off throttle ie. Lower the coast ramp angle for clutch LSD's.
If the car is unstable under on throttle acceleration you'll want to decrease the power rample angle as that gives higher locking which makes it more stable on throttle?

Jussi Karjalainen
19-10-2017, 16:37
If the car is unstable under on throttle acceleration you'll want to decrease the power rample angle as that gives higher locking which makes it more stable on throttle?Increasing the angle (= decreasing the locking) will mean the inside tyre is more free to spin without torque being transmitted to the outside rear. This makes getting power oversteer more difficult (with 90 degrees it's very difficult to get power oversteer) and helps stabilize the car when on throttle, but on the other hand you won't be able to put down as much power.

honespc
21-10-2017, 10:11
Hey, you do what works for you. =)yah. Problem is that many times it feels as if the settings would work right the opposite way to what is described in the paragraphs.

Let's see

A tight angle. Something lower than 50º (20º, 30º, etc) leads to more lock as told in the in-game description.
A greater angle (above 50º, right?, like 60, 70 80 etc) leads to less diff lock.

Same for the decel. diff which is now bellow the coast ramp, is it.

Ok. Does that work as described then?, because sometimes if feels the opposite.


Increasing the angle (= decreasing the locking)See?, this is right the opposite thing to what is described in game.

So what we do now?, what is correct and what is not

Jussi Karjalainen
21-10-2017, 11:18
I don't see the problem. Minimum angle is 20 degrees, which gives maximum locking, 90 degrees is the maximum angle which gives no locking. Increasing the angle towards 90 degrees leads to less locking effect. The in-game descriptions and my message are saying the same things, there's no issue there.

F1_Racer68
21-10-2017, 12:51
In game description is exactly correct.

A lower angle means less of a slope. A lower slope means less effort or energy is required to get from the bottom of the slope to the top of the slope. Less energy required means it's easier/faster to get to the top, which means MORE lock.

honespc
21-10-2017, 13:42
A lower (tighter, say 20 or 30º) angle should be meant precisely to avoid spinning wheels when accelerating off a corner, right?, then how come the car wants to enter better when the angle is like that, precisely, instead of understeering a bit more given the locking action. With a greater angle (60 or higher lets say) the car would turn better but at the expense of spinnig wheels when on throotle, wouldn't it?, because there's way less locking

I just dont get it, me guess. Sometimes one has no choice but to think all this comprehensive stuff comes to only overcomplicate things, and drive everybody8 mad at trying to figure things out.

Bealdor
21-10-2017, 14:13
A lower (tighter, say 20 or 30º) angle should be meant precisely to avoid spinning wheels when accelerating off a corner, right?, then how come the car wants to enter better when the angle is like that, precisely, instead of understeering a bit more given the locking action. With a greater angle (60 or higher lets say) the car would turn better but at the expense of spinnig wheels when on throotle, wouldn't it?, because there's way less locking

I just dont get it, me guess. Sometimes one has no choice but to think all this comprehensive stuff comes to only overcomplicate things, and drive everybody8 mad at trying to figure things out.

Acceleration (corner exit): Lower angle -> More Acc diff lock -> more power transferred to the inside wheel -> inside wheel more likely to lose traction -> power oversteer

Edit: Deisregard, I'm talking nonsense.

Coasting (corner entry): Lower angle -> More Dec diff lock -> inside wheel resisting to go slower than outside wheel -> understeer

bmanic
21-10-2017, 17:30
That is exactly the point Roger. Everything else about the car tuning is allowed or blocked according to real world specifications and regulations.

Why aren't the differentials?

I have to agree with those commenting that this is an oversight/mistake in particular for MP/online/community/esports drivers or events. It gives an unrealistic ability to the cars. I won't go so far as to say it is a "cheat", but it is certainly along the same vane as an unrealistic driver aid.

This is why I think the whole car setup system should be designed different.. the multiplayer admin should be able to go into the car setup and "lock" parts he doesn't want people to tweak, and also be able to put "default values" for things like air restrictor, turbo, gearing etc.

So basically the game should be 100% open with all possible setup options but then a "Realism" setting would automatically lock things depending on simulated car series.

honespc
21-10-2017, 20:59
How come that if there's more Acc diff lock then there's more power transferred to the wheel, and thus it will lose traction?. The game description says right the opposite thing, that if you apply a tighter accel diff it will obviously apply more block and thus wheel will spin less, and that if you go a larger angle, say 60, there will be less lock and thus the wheel will spin more on accel.

Same for dec diff.

I don't understand guys. Seriously. So you say that by a tighter angle = more lock (until here it's ok, but now: ), will cause wheel spin? shouldn't that be the other way with a greater angle because that offers less block?

This is seriously quite a headache. I don't understand anything

F1_Racer68
21-10-2017, 21:28
How come that if there's more Acc diff lock then there's more power transferred to the wheel, and thus it will lose traction?. The game description says right the opposite thing, that if you apply a tighter accel diff it will obviously apply more block and thus wheel will spin less, and that if you go a larger angle, say 60, there will be less lock and thus the wheel will spin more on accel.

Same for dec diff.

I don't understand guys. Seriously. So you say that by a tighter angle = more lock (until here it's ok, but now: ), will cause wheel spin? shouldn't that be the other way with a greater angle because that offers less block?

This is seriously quite a headache. I don't understand anything

This is a common area of confusion.

A more open diff allows the inside wheel to rotate more freely from the outside wheel. If all things are constant, then an open differential means a better turning car, whereas a more locked differential means both wheels closer to the same rate of rotation and therefore less willing to turn.

BUT (and this is a big but) there are multiple factors that come into play, because things are not always "constant" as described above.

Think of the power being sent into the differential as water. Water will always flow down the path of least resistance. In the case of a differential, the path of least resistance is the wheel with the least amount of grip.

With a more open differential, during cornering, there is less weight on the inside wheel, and therefore less grip. Because the power will want to follow the path of least resistance, more of it will go to the inside wheel causing the inside wheel to potentially break free and spin. This will not cause a big issue to the car though since there is little grip on this wheel anyways. All it will do is waste power and reduce the car's ability to accelerate.

With a more locked differential, the power is being forced more equally to both wheels. In this case, if power is applied more aggressively, both the inside and outside wheels will be able to overpower the amount of grip that they have while cornering. This will result in the car spinning out as the outside wheel loses grip due to the amount of power being forced to it. This is what is called "power on oversteer".

So yes, if the amount of power applied to an open diff and a locked diff is equal and constant, then a locked diff is more prone to understeer. If however the amount of power being applied is somehow manipuplated (by a heavy right foot), then that natural tendancy can be overcome.

Does that help it make sense, or did I just confuse you more?

EDIT: The key to tuning a differential is finding that "sweet spot" between open enough to not overpower the outside wheel, but locked enough to get maximum acceleration out of the corner by not causing excessive inside wheel spin.

honespc
21-10-2017, 21:37
lol

ok. After this is i see the only thing that matters is to get proper diff setups for each car, since literally its just too much of a gamble and make little sense (a is a, and b is b, but a can behave like b, and b like a ROFLMAO ok ) and I can pressume they vary depending on the car, making things even more laughably confusing.

So what's best choice after this mind****. Do I set both acel and dec (coast and whatever is called now) by around 50/50 then?

I think you sms have made the diff sections too overcomplicated, and I'm afraid prone to weird behaviours depending on each car too, because for starters, why are diffs tune-able in all cars?, that should be simply not possible.

Jussi Karjalainen
21-10-2017, 22:24
It's not that difficult:

If you want power oversteer, use a low power ramp angle. If you don't want power oversteer, use a high power ramp angle.

If you want lift-off oversteer, use a high coast ramp angle. If you don't want lift-off oversteer, use a low coast ramp angle.

Jussi Karjalainen
21-10-2017, 22:26
With a more locked differential, the power is being forced more equally to both wheels.The whole point of locking is to NOT apply power to the wheels unequally. An open differential will always send 50% of the torque to each wheel, it's always 50:50. When you add locking you start to change that, sending more torque to the wheel with more grip.

F1_Racer68
21-10-2017, 23:34
The whole point of locking is to NOT apply power to the wheels unequally. An open differential will always send 50% of the torque to each wheel, it's always 50:50. When you add locking you start to change that, sending more torque to the wheel with more grip.

Poor choice of wording on my part perhaps. My point was the same as the one made in the start of the video below. Perhaps an oversimplification.

For honespc, take a look at these videos. Perhaps that will help?

Limited Slip Differentials

https://www.youtube.com/watch?v=WeLm7wHvdxQ

This second video about LSDs may help clear up the "ramp angle" confusion. I believe what the game refers to as "Ramp Angle", is the angle on the "Cam Groove" in this video. As you can see, a smaller (shallower) angle on that grove would make it easier for the spider gear shaft to force the separation of the Pressure Rings.

https://www.youtube.com/watch?v=PEdnH7_7_yc

And for some more basics, here is a standard differential

https://www.youtube.com/watch?v=SOgoejxzF8c

honespc
22-10-2017, 08:55
why was my latest post deleted

Roger Prynne
22-10-2017, 10:13
I cant see any deleted posts by you, apart from the one you deleted yourself yesterday at 14:40.

honespc
22-10-2017, 16:08
It's not that difficult:

If you want power oversteer, use a low power ramp angle. If you don't want power oversteer, use a high power ramp angle.

If you want lift-off oversteer, use a high coast ramp angle. If you don't want lift-off oversteer, use a low coast ramp angle.Then it works the opposite way to described in the game. That was the point of my first post.

Otherwise man call me little stubborn because I don't understand this. Under more lock, a tighter angle (let's say 30º) according to games description, should lead to less spin, and what you say in this message is straight away the opposite thing to what is described, because wheel spin should be an effect induced by a more open (weaker locking) differential, right? (what in this game should be a greater angle according to description).

If I wanted a more locked the differential (in this game going for a tighter angle) then I should get less wheel spin and more understeer when throtling off corners.

JyriK
22-10-2017, 17:08
Look at it this way: You decrease inside tire spin by adding more lock, but when a treshold is reached the outside tire starts to spin, resulting in corner exit oversteer, which is usually snappy. So there's two types of spinning, the lock removes the first, but too much lock introduces the second.

tommysalami
23-10-2017, 00:55
Then it works the opposite way to described in the game. That was the point of my first post.

Otherwise man call me little stubborn because I don't understand this. Under more lock, a tighter angle (let's say 30º) according to games description, should lead to less spin, and what you say in this message is straight away the opposite thing to what is described, because wheel spin should be an effect induced by a more open (weaker locking) differential, right? (what in this game should be a greater angle according to description).

If I wanted a more locked the differential (in this game going for a tighter angle) then I should get less wheel spin and more understeer when throtling off corners.

Many drift cars have a fully locked spool differential to make the car drift easily. Think about that when considering locking. The more lock, the closer to a spool or welded diff it becomes, and more oversteer on power because it's more likely to spin the outside wheel.

Jussi Karjalainen
23-10-2017, 07:46
Then it works the opposite way to described in the game. That was the point of my first post.

Otherwise man call me little stubborn because I don't understand this. Under more lock, a tighter angle (let's say 30º) according to games description, should lead to less spin, and what you say in this message is straight away the opposite thing to what is described, because wheel spin should be an effect induced by a more open (weaker locking) differential, right? (what in this game should be a greater angle according to description).

If I wanted a more locked the differential (in this game going for a tighter angle) then I should get less wheel spin and more understeer when throtling off corners.When you have low or no amount of power locking, you're allowing the inside wheel (with less grip due to load transfer) to spin more freely. This however limits the amount of torque going to the outside wheel, so you end up with just the inside wheel spinning, which won't cause you power oversteer under most conditions.

However as you increase the amount of power locking, more and more torque is being sent to the outside wheel, the last driven wheel that still has grip. With enough locking you'll cause both driven wheels to lose grip, and end up with power oversteer. To add to that, sending more torque to the outside wheel acts partially like torque vectoring, creating an extra yaw moment that wants to rotate the car around.

So while it is true that with more locking you can use more of your available grip (since instead of being limited on the grip of the wheel with lowest amount of grip you're pushing more of the torque to the wheel with higher amount of grip), the dynamics of the car change more towards oversteer when using enough power.

F1_Racer68
23-10-2017, 11:02
When you have low or no amount of power locking, you're allowing the inside wheel (with less grip due to load transfer) to spin more freely. This however limits the amount of torque going to the outside wheel, so you end up with just the inside wheel spinning, which won't cause you power oversteer under most conditions.

However as you increase the amount of power locking, more and more torque is being sent to the outside wheel, the last driven wheel that still has grip. With enough locking you'll cause both driven wheels to lose grip, and end up with power oversteer. To add to that, sending more torque to the outside wheel acts partially like torque vectoring, creating an extra yaw moment that wants to rotate the car around.

So while it is true that with more locking you can use more of your available grip (since instead of being limited on the grip of the wheel with lowest amount of grip you're pushing more of the torque to the wheel with higher amount of grip), the dynamics of the car change more towards oversteer when using enough power.

Ha! So my explanation was right.... Sort of. And now I see where the issue was that you had with it. A simple difference of interpretation of the word "power" in my explanation.

Whereas I used the word "power" (little p) as a generic term, meaning a form of energy or a force being applied, you read it as "Power" (big p) meaning the actual mathematical value. :)

So had I said Torque in my original explanation rather than power it would have been virtually identical to your explanation above.

Whew...... At least now this confirms to me that I do understand this stuff.... I just need to be more accurate in the terminology :)

Thanks Jussi. Glad to have that sorted.

EDIT: This is also a perfect example why choice of words in everything we say and do is so critical, especially in a public, and international, space.

Atak Kat
29-10-2017, 06:59
It's not that difficult:

If you want power oversteer, use a low power ramp angle. If you don't want power oversteer, use a high power ramp angle.

If you want lift-off oversteer, use a high coast ramp angle. If you don't want lift-off oversteer, use a low coast ramp angle.

Your patience with everyone is golden. Thanks for that.
Can I test my comprehension (simplified, assuming suspension adjustments not a factor here...)?
- on corner exit, on power, if my car is understeering too much, I interpret that as the rear axle 'pushing' the car and therefore the diff is locking too much. So I would assume that I need to increase the power ramp angle? (if I'm correct, then I am shocked.)
- on corner entry, off power, if my car is understeering too much, I interpret that the rear axle deceleration forces from the engine are somehow not moving freely enough between the two rear wheels (causing one or the other wheel to skid and therefore pulling the car and feeling like an understeer?). So to fix that I need to allow the deceleration torque to move more freely between the rear wheels, which means to have a higher coast ramp angle? (I think that's wrong though... )

Thanks so much for your patience... I suspect I've understood the power on part, but not the coast part....

rich1e I
30-10-2017, 16:16
I think if your car understeers on corner exit you want more locking effect --> decrease power ramp angle. If you want more lift-off oversteer for better rotation on corner entry --> increase coast ramp angle.

eddo
03-11-2017, 08:06
I'm having trouble. cant turn spool off game wont let me is it a bug or is there something I'm missing

Bealdor
03-11-2017, 08:19
I'm having trouble. cant turn spool off game wont let me is it a bug or is there something I'm missing

Which car?

Racefancy
05-11-2017, 05:44
Quick question regarding clutch/friction plate modifications vs preload. In a situation where for example ramp angles and preload were unable to be changed, does reducing the clutch plates, for example 8 to 6 create a similar response to increasing ramp angles or more similar to reducing diff preload?

I assume even with less plates and with the same amount of preload applied that there won't be a benefit in mid corner rotation, however once the ramps are activated it will reduce the amount of friction in the system which would be similar to increasing ramp angles, but wanting to make sure.

sillib
08-11-2017, 13:03
Hi....

Question.....: ramp angle has effect during braking and at the just moment when lifting off the brakes for a second or less or at the moment that the lifting off the brakes takes place preload takes charge instantlly.....

In mid corner applying 5% throttle or so just to keep the speed stable is also under preload or even that small input activates power ramp.....?

Hope this thread is still active.....

Jussie thanks for your work..... amazing....!!!

hkraft300
08-11-2017, 14:25
Quick question regarding clutch/friction plate modifications vs preload. In a situation where for example ramp angles and preload were unable to be changed, does reducing the clutch plates, for example 8 to 6 create a similar response to increasing ramp angles or more similar to reducing diff preload?

Less clutches ~ Increasing ramp angles. Roughly.
The ramps push against the clutches to lock/ slip the differential.



Question.....: ramp angle has effect during braking and at the just moment when lifting off the brakes for a second or less or at the moment that the lifting off the brakes takes place preload takes charge instantlly.....

Jussie thanks for your work..... amazing....!!!

Whichever is greater I believe, but also would like clarification on this.
While there is decelerating/negative torque is available, if coast lock is greater than preload then I believe coast lock would be in effect while off-throttle. And vice versa.

Bealdor
08-11-2017, 15:08
Less clutches ~ Increasing ramp angles. Roughly.
The ramps push against the clutches to lock/ slip the differential.



Whichever is greater I believe, but also would like clarification on this.
While there is decelerating/negative torque is available, if coast lock is greater than preload then I believe coast lock would be in effect while off-throttle. And vice versa.

Preload is ALWAYS there and works during ACC and DEC.
Power lock and coast lock effect is added on top of it.

hkraft300
08-11-2017, 15:20
Preload is ALWAYS there and works during ACC and DEC.
Power lock and coast lock effect is added on top of it.

:hopelessness: if only I knew...

Bealdor
08-11-2017, 15:30
:hopelessness: if only I knew...

The preload is essentially just a spring that pushes the clutches together.
Since both, power and coast lock are working in the same direction you'll always add locking force to the already existing one (preload).

sillib
08-11-2017, 16:05
Thanks..... good to know......!!!! So..... when tuning better start with the preload ang then with the rest of the differential.... possiblly....

hkraft300
08-11-2017, 21:13
Thanks..... good to know......!!!! So..... when tuning better start with the preload ang then with the rest of the differential.... possiblly....

I’d done it backwards so far.
Adjusted the ramps, then clutches, then preload.
I might reverse that :rolleyes:

sillib
11-11-2017, 14:10
Well.... i think it is a situation where you start from the ramps and the clutches after all but you must have preload in mind in some off input situations to control oversteer or understeer that might come up.

Also another question if one might know...... :
Using the calculator if for instance i have lets say 40% for decell lock using different values for differrent combination of ramp or clutces that means i have the same results for both scenarios ex.. dec ramp 40 and clutch packs 4 gives 38% decel lock and dec ramp 50 and clutch packs 6 gives 40% decel lock... but lets assume that i had two cases with the exact same result lets say 40%..... does the car behave exactly the same or is there some hidden info in those different values.....?

I hope things wont get more complicated and the car behaves the same becsuse it sure feels very close.....

Jussi Karjalainen
01-12-2017, 01:15
Heya guys, been busy at work, sorry for no responding sooner.

Preload is indeed always active, and power/coast locking adds on top of that. When you're using the calculator you can see that in the graph. Whether you adjust the preload or the ramps first/last is up to personal preference.

Preload the vast majority of the time won't affect your total locking amount that much (low torque engines excepted) but it will affect the behavior when you're feathering the throttle, i.e. the behavior between engine braking and power on.


Well.... i think it is a situation where you start from the ramps and the clutches after all but you must have preload in mind in some off input situations to control oversteer or understeer that might come up.

Also another question if one might know...... :
Using the calculator if for instance i have lets say 40% for decell lock using different values for differrent combination of ramp or clutces that means i have the same results for both scenarios ex.. dec ramp 40 and clutch packs 4 gives 38% decel lock and dec ramp 50 and clutch packs 6 gives 40% decel lock... but lets assume that i had two cases with the exact same result lets say 40%..... does the car behave exactly the same or is there some hidden info in those different values.....?

I hope things wont get more complicated and the car behaves the same becsuse it sure feels very close.....The calculator is somewhat of a simplification of what really happens in the physics (I try but it's complicated), but different angles and clutch counts resulting in the same percentage should be pretty much the same end result.

poirqc
27-01-2018, 14:16
I've just accessed the Diff calculator and Google told me it's in the trash! Don't delete it! :p

blinkngone
30-01-2018, 20:41
I almost hate to ask this question. I have seen many setups with all the differential switched to off. They are running very fast times. How/why does turning all differentials to off work?
Here is an example of one guys front wheel drive LSD, he does the same on rwd, there are many Indy cars using the all diffs off LSD.
249301

lotzik
01-02-2018, 04:25
Hey Jussi, great information on this post. I need a quick tip on your guide.

Ok so I understand how different numbers affect locking and in what way. My question would be, if there is a general rule of thumb on the locking percentage we should be generally go after when tuning a car?

An example, I tried driving the La Ferrari and noticed that when releasing throttle entering a corner the car would throw it's tail out and the same when pressing throttle to exit the corner. Generally the driving felt very unstable. Also by default, I would even get spinning on a straight line with full throttle. Does this car need a 50% locking, 100% or maybe 75%? If the "ideal" (I understand that there can be a matter of preference here) is for example a 75% lock for the perfect balance, then would inducing a 100% create a very understeery car with opposite results?

So would it be correct if we said

25% = Oversteer generally
50% = Balanced Steering
75% = Understeer generally

This is the point I can't get out of this post, what is the percentage I should aim for when tuning?

Jussi Karjalainen
21-02-2018, 19:27
Hi guys, sorry for the silence. Not gonna sugarcoat it: I just kind of forgot about the whole forums (namely I forgot to read them personally, we do of course get information delivered to us regardless) for a while when I was sick over Dec and Jan, and didn't remember to come back after I got better. Trying to catch up with everything that I've missed now, better late than never I guess?
I've just accessed the Diff calculator and Google told me it's in the trash! Don't delete it! :pThat's alarming, it shouldn't be! I'll look into it.
I almost hate to ask this question. I have seen many setups with all the differential switched to off. They are running very fast times. How/why does turning all differentials to off work?
Here is an example of one guys front wheel drive LSD, he does the same on rwd, there are many Indy cars using the all diffs off LSD.
249301Well, at least an open diff prevents any kind of power oversteer for the most part, but it does also limit the maximum amount of power you can effectively utilize. Also it'd tend to make cars pretty unstable, so you'd either need to be super smooth on the controls or work the setup to be more stable otherwise. Some cars these days have so much grip in their tyres though that even in real life they run very little accel locking, LMP cars are a good example of that. They have so much grip and great traction control systems, that they don't really need much locking to put down fast laps. The driver might want some extra to get the behavior they want out of the car though.

None of us testing ever found that we were faster with an open diff than with a closed one, at least under the vast majority of conditions, but it's something to look into, definitely.


Hey Jussi, great information on this post. I need a quick tip on your guide.

Ok so I understand how different numbers affect locking and in what way. My question would be, if there is a general rule of thumb on the locking percentage we should be generally go after when tuning a car?

An example, I tried driving the La Ferrari and noticed that when releasing throttle entering a corner the car would throw it's tail out and the same when pressing throttle to exit the corner. Generally the driving felt very unstable. Also by default, I would even get spinning on a straight line with full throttle. Does this car need a 50% locking, 100% or maybe 75%? If the "ideal" (I understand that there can be a matter of preference here) is for example a 75% lock for the perfect balance, then would inducing a 100% create a very understeery car with opposite results?

So would it be correct if we said

25% = Oversteer generally
50% = Balanced Steering
75% = Understeer generally

This is the point I can't get out of this post, what is the percentage I should aim for when tuning?Ah, this is an interesting question. Honestly, it all depends (yay). The percentage refers to how much locking force is being generated by the torque that's coming into the differential. So if your driveline is putting in 1000 Nm of torque (remember that gearing ratios will multiply torque) and you have 50% locking, the locking torque is 500 Nm. So if you have a hugely powerful/torquey engine engine, you can get a lot of locking force even with relatively low locking percentage. Conversely if you have a low powered engine like lets say the Ford Escort or the Nissan Fairlady, to get enough locking force you might have to run the diffs much stiffer. So on a high power Formula car 10-15% might be enough while a low powered racer might work well with 70% or more (one of the reasons why the decel locking is sometimes so high, engine braking doesn't give you nearly as much negative torque as the engine's power gives positive torque). And of course what gear you're in affects things as well, you get more locking in lower gears than you get in higher gears due to the torque multiplication of the transmission changing.

As for the percentages you put out, are those for coast or power locking? 25% locking on coast would be very little and could lead to entry oversteer, while for acceleration locking it's not that much and might on some cars still have power understeer. And like I say above it's all quite dependent on the car anyway.

The best rule of thumb I can give is that if you want to avoid power oversteer try to stay under 30% accel locking, maybe even under 20% on high powered cars, and for entry oversteer you'll generally want more than 40%, and 70% is a-OK.

blinkngone
21-02-2018, 21:12
Hi Jussi, thanks for the reply. I guess they are getting away with it with the oval track Indycars(Honda and Chevrolet), Formula A and Formula X at Monza Historic Oval and the Daytona Tri Oval. If you try and activate any other LSD even minimally you are going to be way off the pace. But they can do it with some FWD/RWD cars as well. Some people are also able to run with all the differentials(except ratcheting) turned on at other tracks. So, I think you are saying that all differential types "off" is the same as an "open" differential then what would be the description of all differentials "on"?

Jussi Karjalainen
21-02-2018, 22:28
All of them on? That'd include spool, which trumps everything else (for hopefully obvious reasons), so that means it'd equal a spooled diff. =)

RJay
22-02-2018, 22:17
everything on = 100% locking, you basically have a drift car.
Everything off = 0% locking, you basically can manouver but you can't effectively put power down untill you're straight because you'll fire up the inner wheel and waste a lot of power :)

Although everything off can work in as you say ovals, because the steering angle and speed differences between the tires isn't great enough so that the power sent to the wheels overcomes the limit the inner tire can handle.

Jussi Karjalainen
22-02-2018, 23:09
100% locking, you basically have a drift car.This is a thing I kind of want to throw out there, make of it what you will: 100% locking, at least with the terminology used normally when talking about LSDs, is not the same thing as a spooled diff. Even with 100% locking the two axles can still rotate at different speeds.

100% locking means that the locking force (torque needed to make one axle rotate faster than the other) is 100% of the torque coming in to the differential. While it does mean that the engine can never rotate one axle faster than the other (the power of the engine will rotate the axles at the same speed even if they have a grip difference), other forces can still cause differentiation.

If you consider what happen when a car turns, when the inside tyre is trying to rotate slower than the outside tyre due to the curvature of the corner, the road is essentially applying a different amount of force to each side of the differential. If your tyres are grippy enough, the turn is tight enough and the car isn't super powerful, those forces can overcome the locking force in the diff and cause differentiation. A spooled diff will never allow the two axles to rotate at different speeds (though there is some flex in it).

So basically what I'm trying to say is that when you're turning into a parking space, there's potentially a big difference between having a diff with 100% locking and a spooled diff. It's a technicality but I like it. =)

RJay
22-02-2018, 23:31
This is a thing I kind of want to throw out there, make of it what you will: 100% locking, at least with the terminology used normally when talking about LSDs, is not the same thing as a spooled diff. Even with 100% locking the two axles can still rotate at different speeds.

100% locking means that the locking force (torque needed to make one axle rotate faster than the other) is 100% of the torque coming in to the differential. While it does mean that the engine can never rotate one axle faster than the other (the power of the engine will rotate the axles at the same speed even if they have a grip difference), other forces can still cause differentiation.

If you consider what happen when a car turns, when the inside tyre is trying to rotate slower than the outside tyre due to the curvature of the corner, the road is essentially applying a different amount of force to each side of the differential. If your tyres are grippy enough, the turn is tight enough and the car isn't super powerful, those forces can overcome the locking force in the diff and cause differentiation. A spooled diff will never allow the two axles to rotate at different speeds (though there is some flex in it).

So basically what I'm trying to say is that when you're turning into a parking space, there's potentially a big difference between having a diff with 100% locking and a spooled diff. It's a technicality but I like it. =)

Everything on means the spooled diff is on too ;) By a 100% locked diff I do mean a lsd in it's active locked state (at all times)

Or you can weld the diff shut as they say, that creates the same effect as a spool, or just a live axle. that's always fun.

Also I'm curious, for the Porsche GT3 users how do you set your diffs? I like it with the coast at at least 35% and the power ramp at 45-50 (depending on the track / rain).
With the stock coast at 20 (I believe it was) it would just try to kill me rolling into off camber corners, like for instance Rivage on Spa.

RJay
22-02-2018, 23:37
delete this post please

Jussi Karjalainen
23-02-2018, 16:54
That was more me pondering out loud the difference between what "100% locking" means compared to actually having a solid, spooled diff, that allows no differentiation at all. =)

Spooled diffs don't really have any sort of measurable "locking percentage", since they will prevent any and all slip always. The locking percentage is a measurement of the locking force between the two driven half axles compared to the torque coming into the differential. It's not a measurement of how much slip is allowed between the half axles, it tells you something about how much torque is needed to get them to differentiate, and how much extra torque can be exerted on the side with more grip. There aren't really any "locking states" as such, anytime there's torque going into the diff (engine power or engine braking basically) there's some amount of locking happening. More torque = more locking. And theoretically this force can always be overcome and slip generated, with really high locking percentages and power/torque levels it might not occur in any reasonable condition. But a spool will never allow differentiation without being twisted and warped, likely breaking in the process. So it's an apples and oranges thing to some extent. =)

This is just me rambling theoretical stuff, it doesn't really apply to any reasonable driving conditions. =)

On the Porsche GT3, which one do you mean, the GT3 RS road car or the GT3 R racing car? Checking the defaults for the GT3 R racing car, the Stable defaults to these values:

Number of clutches: 4
Power ramp: 65 degrees (~15% locking with 4 clutches)
Decel ramp: 25 degrees (~70% locking with 4 clutches)

On the Loose it's:

Number of clutches: 8
Power ramp: 55 degrees (~45% locking with 8 clutches)
Coast ramp: 45 degrees (~64% locking with 8 clutches)

The coast locking is similar on both and not that extreme all things considered. I also didn't really noticed it being excessively unstable at Rivage, but if you've been having such issues I'm happy that you managed to tune it out for yourself. =)

2010 Synergy Camaro
23-02-2018, 21:09
So I'm trying to simplify this in my fragile eggshell mind after reading through this thread. Higher Acc Lock in game setting = higher safety/less performance. Lower Dec lock in game setting = higher safety/less performance.

Bealdor
23-02-2018, 21:36
So I'm trying to simplify this in my fragile eggshell mind after reading through this thread. Higher Acc Lock in game setting = higher safety/less performance. Lower Dec lock in game setting = higher safety/less performance.

It's the other way around.

High acc lock = better acceleration possible but more prone to power oversteer
Low acc lock = less acceleration possible but safer at corner exit

High dec lock = more understeering / worse turn in ability but safer
Low dec lock = better turn in ability but makes the rear quite loose (oversteering)

2010 Synergy Camaro
23-02-2018, 22:41
It's the other way around.

High acc lock = better acceleration possible but more prone to power oversteer
Low acc lock = less acceleration possible but safer at corner exit

High dec lock = more understeering / worse turn in ability but safer
Low dec lock = better turn in ability but makes the rear quite loose (oversteering)

Thank You so much Bealdor. I'm writing this down. Trying to understand differential percentages and angles and all that stuff isn't my thing.

RJay
23-02-2018, 23:41
That was more me pondering out loud the difference between what "100% locking" means compared to actually having a solid, spooled diff, that allows no differentiation at all. =)

Spooled diffs don't really have any sort of measurable "locking percentage", since they will prevent any and all slip always. The locking percentage is a measurement of the locking force between the two driven half axles compared to the torque coming into the differential. It's not a measurement of how much slip is allowed between the half axles, it tells you something about how much torque is needed to get them to differentiate, and how much extra torque can be exerted on the side with more grip. There aren't really any "locking states" as such, anytime there's torque going into the diff (engine power or engine braking basically) there's some amount of locking happening. More torque = more locking. And theoretically this force can always be overcome and slip generated, with really high locking percentages and power/torque levels it might not occur in any reasonable condition. But a spool will never allow differentiation without being twisted and warped, likely breaking in the process. So it's an apples and oranges thing to some extent. =)

This is just me rambling theoretical stuff, it doesn't really apply to any reasonable driving conditions. =)

On the Porsche GT3, which one do you mean, the GT3 RS road car or the GT3 R racing car? Checking the defaults for the GT3 R racing car, the Stable defaults to these values:

Number of clutches: 4
Power ramp: 65 degrees (~15% locking with 4 clutches)
Decel ramp: 25 degrees (~70% locking with 4 clutches)

On the Loose it's:

Number of clutches: 8
Power ramp: 55 degrees (~45% locking with 8 clutches)
Coast ramp: 45 degrees (~64% locking with 8 clutches)

The coast locking is similar on both and not that extreme all things considered. I also didn't really noticed it being excessively unstable at Rivage, but if you've been having such issues I'm happy that you managed to tune it out for yourself. =)
Must be a different setting, because I tuned mine from that 20-25 about-ish coast ramp to the 40 degrees, and I'm on a "loose" setting.
Odd, could be a difference between patches?

Jussi Karjalainen
24-02-2018, 08:46
Thank You so much Bealdor. I'm writing this down. Trying to understand differential percentages and angles and all that stuff isn't my thing.Also good to know is that higher ramp angle = less locking, as the help text denotes. So in a very very simplified form if you want to avoid power oversteer, increase the power ramp angle, and if you want to avoid lift off oversteer, decrease coast ramp angle.

M4MKey
24-02-2018, 10:25
what ? So by less locking on the accel diff, you have less powersteering ? That' doesn't make any sense. I agree with the first and final part of your post. Higher angle = less locking. But then Higher angles on the accel diff makes more powersteering :o

RJay
24-02-2018, 14:10
what ? So by less locking on the accel diff, you have less powersteering ? That' doesn't make any sense. I agree with the first and final part of your post. Higher angle = less locking. But then Higher angles on the accel diff makes more powersteering :o

less locking = less power over is actually very logical, because the axle "locks" it self later, allowing for more power being sent to the unloaded wheel.
Power steering is an altogether different thing.

2010 Synergy Camaro
24-02-2018, 16:25
Also good to know is that higher ramp angle = less locking, as the help text denotes. So in a very very simplified form if you want to avoid power oversteer, increase the power ramp angle, and if you want to avoid lift off oversteer, decrease coast ramp angle.

Thank You Jussi. I also appreciate your suspension calculator very much. I am able to use and understand it just fine. Hopefully, after playing around with the differential calculator I will get the hang of it in time also.

Jussi Karjalainen
24-02-2018, 16:55
what ? So by less locking on the accel diff, you have less powersteering ? That' doesn't make any sense. I agree with the first and final part of your post. Higher angle = less locking. But then Higher angles on the accel diff makes more powersteering :oMore accel locking increases power oversteer for the same reason why drift cars use aggressive accel locking (or even spooled diffs) instead of open diffs. =)

If your diff is open, the torque distribution is 50:50, so the amount of power/torque you can put down is limited by the tyre with the lowest grip. If one tyre has only enough grip to manage 100 Nm of torque without slipping, you can't send more than 100 Nm to the other tyre either, for a total of 200 Nm going to the ground and doing useful work, even if it could handle 1000 Nm of torque.

What locking does is effectively increase the amount of torque you can send to the side with more grip. If you were in the situation above but had 100 Nm of preload, you could send 100 Nm to the tyre with less grip and 100 Nm to the tyre with more grip due to the 50:50 torque split, but also 100 Nm extra on the tyre with more grip due to the preload, for a total of 300 Nm of torque going to the ground. Accel locking works in the same way basically, the main difference between it and preload is that preload is always the same, whereas accel (and decel) locking varies by how much torque is coming to the diff (so on 100% throttle you get more locking than on 50% throttle, in low gears you get more locking than in high gears, etc.).

So, locking allows you to put extra torque/power on the tyre with more grip. Before you were limited by the tyre with less grip, but locking enables you to overpower both tyres at the same time. This causes oversteer. =)

And on top of that there's the potential for a yawing torque to be applied on the car with more locking. If you have lots of accel locking, maybe even a spooled diff, and you're cornering hard, your outside tyre will have way higher grip potential than your inside tyre. If you mash the throttle, most of the thrust pushing the car forwards will be coming from the outside rear tyre. That tyre is not in the center of the car, so as it pushes the car it will also try to rotate it (just like if you push a box forwards near it's edge it'll spin, but if you push it in the center it'll go straight). That can add an extra tendency towards oversteer. =)

Here's a showcase of a car with 90 degree accel ramp and no other locking, vs. the same car with 20 degree accel ramp:


https://www.youtube.com/watch?v=ke4zYqMgxL8

EDIT: At low loads where you're not really straining against the limitations of the tyres (for example turning into a parking spot, which requires lots of differentiation but isn't really putting the tyres at their limits of grip under normal conditions) lots of locking can indeed cause understeer, but when pushing to the edge it becomes about the difference between spinning one wheel vs. spinning both. With locking you'll always want to rotate both wheels, one can't turn independently, so if you're going to get wheelspin, you'll get it on both driven wheels.

PostBox981
24-02-2018, 18:49
Thanks Jussi for spending so much time in this thread. I am sure I still don´t get all my diff settings right but with every single post of yours the technology is getting a little clearer for me. Without all your help I guess some 95% of us would be completely lost in the dark. :victorious:

Max Torque
02-05-2018, 22:08
So I need low values to prevent over/understeer? Thought it was the other way around.

Jussi Karjalainen
02-05-2018, 23:41
So I need low values to prevent over/understeer? Thought it was the other way around.Low coast ramp angle = more deceleration locking = more understeer

Low power ramp angle = more acceleration locking = more power oversteer (when a car is capable of breaking traction at least)

Max Torque
03-05-2018, 09:39
Low coast ramp angle = more deceleration locking = more understeer

Low power ramp angle = more acceleration locking = more power oversteer (when a car is capable of breaking traction at least)

Ah, ok, I confused "more lock" with "higher value". But "higher value" means less lock.

Jezza819
14-05-2018, 15:49
I've just started to try and see if I can make diff changes to fix the cars that I'm having trouble with spinouts either off throttle or on. It doesn't seem to matter. I actually think I could spin one out if it were sitting totally still. But in looking at the diff settings the power and coast doesn't seem to have as big of a range as preload does. Does it make sense to see what the range of power and coast are and set them right in the middle, see what that does, then start going up or down on each and see what happens? I forget which car I was looking at but the low end of preload was -100 and it went all the way up to 400. Needless to say I put it right back where it started and didn't touch it.

El_Tim
21-05-2018, 09:20
As this is my first post in this forum I'd like to say hello to everybody first. Having read through different topics here I really appreciate the helpfullness among each other in general and especially the patient manner of Jussi's explanations...

I've been enjoying PC2 for some weeks now and found myself being best with lsd switched off in formula rookie. With lsd on I would spin out midcorners pretty much as Jezza describes it. Now that I've proceeded to Formula C I wanted to take advantage of the fully adjustable clutch lsd but again I was u-turning unintentionally rather often. Switching off the lsd I was fine again. So I tried to figure out which detail of the setup was causing me trouble starting with clutch lsd switched ON but power and coast ramp to 90 degrees and preload to zero Nm. At least this is what I thought would neutralize the lsd. There still were side-steps at the rear axle I didn't encounter with lsd off. Going all the way down too -100Nm preload finally made it feel like completely switched off. From that point I could start playing with the ramp angles and creating a setup making me confident to push harder (at least within my rather limited skills...).

So there appear some more questions to the experienced lsd tuners.

1) Do I need to calibrate that negative preload in order to reduce the base friction of the clutch plates? If yes: how much base lock does the clutch lsd have? -100Nm seems a bit much I have to admit.
2) especially at Jussi: as drivetrain torque changes sign in transitions from power to coast and vice versa does this also apply to preload?

Thanks in advance and have a nice week,

Tim

Jezza819
25-05-2018, 14:21
I tried to adjust engine braking on one of my most troublesome cars last night, the BMW 320TC, but the settings only went from 1 to 5 and I guess the default was already at 4. It didn't seem to help it much. At the shorter Hockenheim course where I was running it it spun out at turn 1 and also that medium speed right hander as you enter the arena. It's like whichever direction the turn is, the outside tire speeds up and pushes it further in the direction you're turning. Now if I'm going way too fast for the turn and spin it out, that's on me. And at least with me this car doesn't like kerbs because I've spun it out at turn 1 Hockenheim by clipping that left hand kerb, Long Beach on the second right after you go around the fountain, and numerous other places.

The V8 Supercar is the same way except more extreme I'm guessing because of all that torque it's very touchy on corner exit but I haven't looked at it's engine braking yet.

El_Tim
26-05-2018, 14:41
For me lowering preload to minimum really fixed it. No sudden spin-outs anymore and a way more predictable behaviour. Have you tried to switch off all diffs completely in order to see whether this changes your car's disadvantageous attitude towards cornering?

Jezza819
30-05-2018, 15:31
For me lowering preload to minimum really fixed it. No sudden spin-outs anymore and a way more predictable behaviour. Have you tried to switch off all diffs completely in order to see whether this changes your car's disadvantageous attitude towards cornering?

No not really. Since I've just started messing around with those settings I haven't done much. I didn't even know you could switch them off.

Oomph
02-06-2018, 22:36
I tried to adjust engine braking on one of my most troublesome cars last night, the BMW 320TC, but the settings only went from 1 to 5 and I guess the default was already at 4. It didn't seem to help it much. At the shorter Hockenheim course where I was running it it spun out at turn 1 and also that medium speed right hander as you enter the arena. It's like whichever direction the turn is, the outside tire speeds up and pushes it further in the direction you're turning. Now if I'm going way too fast for the turn and spin it out, that's on me. And at least with me this car doesn't like kerbs because I've spun it out at turn 1 Hockenheim by clipping that left hand kerb, Long Beach on the second right after you go around the fountain, and numerous other places.

The V8 Supercar is the same way except more extreme I'm guessing because of all that torque it's very touchy on corner exit but I haven't looked at it's engine braking yet.


Hi,
In regards to these two cars, try using the clutch + heel and toe if you are up for it, especially with the 320TC. I was having serious rear end lock ups as you are having and with a little patience these start to work well. i am actually using the H-gate and clutch instead of sequential with both these cars. i can also pm you my setups if you like.

Phil

David Wright
15-07-2018, 16:55
Jussi - I've been investigating the Clio Cup limited slip diff and found that when you enter the power side ramp angle and number of clutches you end up with more than 100% locking. This isn't possible. Is there an error in your spreadsheet?

Jussi Karjalainen
15-07-2018, 19:47
Jussi - I've been investigating the Clio Cup limited slip diff and found that when you enter the power side ramp angle and number of clutches you end up with more than 100% locking. This isn't possible. Is there an error in your spreadsheet?It is possible, nothing says you can't have more than 100% locking, and 100% locking doesn't mean that the two sides are "totally solidly locked together".

What the locking percentage means is that if there's for example 500 Nm of torque coming into the diff, the locking torque between the left and right side will be XX% of that. So 50% locking would be 250 Nm in that case. With suitable materials it's not tough to achieve a higher locking torque than that. As an EXTREME example, you could have something like this:

257714

When you put those two faces together you don't need to push them together hard at all to create a huge amount of locking force (force required to make them rotate at different speeds). With enough clutch plates that have enough friction, and aggressive ramp angles, you can get a diff that has over 100% locking force, which basically means that you need more torque than what the driveline is transmitting to make the left and right sides rotate at different speeds. 150% locking percentage would mean that with 500 Nm of torque going in to the diff, you'd need 750 Nm of torque difference. You wouldn't be able to get one side to rotate faster than the other side with engine power in that case, but theoretically you could still use external forces to create differentiation.

David Wright
15-07-2018, 21:01
Thanks for your reply. I think we must be using different definitions of locking %. In post 2 you include a diagram which shows how locking % relates to torque bias.

locking % = (torque to high grip side - torque to low grip side)/input torque

so for example with a torque bias of 3:1 = (3-1)/4 = 50%

using this definition you can't get more than 100% locking.

The problem is I think the above definition is used by limited slip diff manufacturers, so its not possible to use your spreadsheet to convert manufacturers figures to ramp angles and clutches.

Jussi Karjalainen
17-07-2018, 14:37
Good point there, but after conversing with others we think the calculations for both should be very comparable where locking percentage/effect is concerned. It's more of a difference of how do you approach the situation, do you try to measure the torque differences at the wheels and then work backwards to the input torque, or do you take the input torque and use the diff specs to calculate how much extra torque could be sent to one side. You should get very close to the same results either way, it's just approaching it from another direction in a manner of speaking.

Don't worry too much about the over 100% locking though, it's mostly theoretical. In practice, since the engine can't put out more than 100% of its torque it basically means that from the driveline's point of view 100% locking behaves the same as over 100% locking. It's more of a theoretical question of "would this amount of input torque also be able to fight off external causes for slip as well, like the torque caused by the two sides trying to roll at different rates?" Feel free to limit the top end to 100% in your mind. =)

EDIT: One reason why I like the "over 100% possible locking" is that it kind of reminds me every time that 100% locking percentage/effect is not the same thing as "fully locked" or the same effect as having a spooled diff. A spooled diff will never allow the two sides to rotate at different rates, beyond the material flexing at least, while even a 100% locking percentage clutch based diff is only just pushing friction faces together and doesn't solidly lock the two sides to one. It basically just means that the engine torque can never make one side rotate faster than the other.

Zeratall
18-07-2018, 01:29
clutch based diff is only just pushing friction faces together and doesn't solidly lock the two sides to one

I think the statement above summarize the distinction quite well.

Zeratall
25-07-2018, 02:02
Here is some data taken with PC2Tuner to demonstrate what happens when adjust the power ramp, the signals are wheel slip, the rear axle are green and orange, notice how too high of power ramp, the inside tire slips first and more than the outside wheel, as you go down in power ramp, they slip the same amount, but if you go too low the spin % of the outside wheel spins more and sooner. From the graphs below I'd probally set my power ramp somewhere between 70 and 50 degs.

258062

Jussi Karjalainen
25-07-2018, 19:55
Nice analysis, that's the kind of good stuff you can do with telemetry. =)

Also remember that the number of clutches is important, not just the angle.

Zeratall
25-07-2018, 21:50
Nice analysis, that's the kind of good stuff you can do with telemetry. =)

Also remember that the number of clutches is important, not just the angle.

Yeah exactly, I kept everything constant except the power ramp, t just to illustrate to people. Once you have data everything becomes pretty clear.

senna94f1
30-07-2018, 01:57
I tried to adjust engine braking on one of my most troublesome cars last night, the BMW 320TC, but the settings only went from 1 to 5 and I guess the default was already at 4. It didn't seem to help it much. At the shorter Hockenheim course where I was running it it spun out at turn 1 and also that medium speed right hander as you enter the arena. It's like whichever direction the turn is, the outside tire speeds up and pushes it further in the direction you're turning. Now if I'm going way too fast for the turn and spin it out, that's on me. And at least with me this car doesn't like kerbs because I've spun it out at turn 1 Hockenheim by clipping that left hand kerb, Long Beach on the second right after you go around the fountain, and numerous other places.

The V8 Supercar is the same way except more extreme I'm guessing because of all that torque it's very touchy on corner exit but I haven't looked at it's engine braking yet.

I actually won a online league championship with the said car , one of the 1st things was lower or decease rear toe in and increase front toe in along with the tow bars and it made the car so stable while everyone else was struggling with spinning or oversteer or understeer ,

This is on ps4 using t300 wheel , once I'd done a few tweaks to get the car to point into the corner I was pretty much ready to race, other then that played around with camber and tyre pressure as to not overheat tyres ,

This was in the dry and rain ,night and day different tracks hot and cold ,so quite a few variables but I mostly over 6 weeks and 9 tracks kept to same set up, just changing fuel loads required for league required pit stops and final gear ratio for slow and fast tracks, ,

Hope this helps as its a difficult car lol

alonsomoslie
29-09-2018, 16:27
Hello Jussi Karjalainen,

I have a little problem. I cant update your Diff Calculator, MS Excel says: "We cant update some of the links in your workbook" etc. And if I click on button Edit Links, I honestly dont know what to do here.
Please help me a little bit in this.
Thank you
Alonso