Feb 13, 2006 10:35:08 PM
- Ulf Karlsson
- Newbie
-
- Loc: Sweden
- Posts: 7, Gallery
- Registered on: Oct 14, 2003
Crown
Board: Porsche - 911 - 996
Language: English
Region: Worldwide
Close
CloseForum - Thread
Re: Limited slip differentials
The first percentage is the locking ratio while accelerating and the second is locking ratio under braking. 911's use a higher locking ratio for braking than for power (many people confuse this). The first asymetrical LSD's were used by Porsche on the 964 Turbo where it had 20% locking on the power and 80% locking on braking. This helps to steady the car while approaching and entering a corner while slowing. This benefit is at least as important (to me) as the reduction in wheelspin it provides at corner exit while on the power. 911's have such a strong rear-weight bias that not much locking is required for good traction under power. And any more than necessary just causes excessive understeer in tight corners.
What do the percentages mean - percent of what? This is harder to answer. The LSD has clutches and the number of clutches as well as the preload and ramps all contribute to the amount of torque that is required to "break" the axle loose (allow the wheels to turn independently). Not all 40% LSD's break loose at the same torque. This number can be somewhat misleading. Maybe the percentage is meant to reference the amount of torque the particular engine can provide. I'm not sure...
What do the percentages mean - percent of what? This is harder to answer. The LSD has clutches and the number of clutches as well as the preload and ramps all contribute to the amount of torque that is required to "break" the axle loose (allow the wheels to turn independently). Not all 40% LSD's break loose at the same torque. This number can be somewhat misleading. Maybe the percentage is meant to reference the amount of torque the particular engine can provide. I'm not sure...
Re: Limited slip differentials
On the last page of the following document has some writing about LSD for GT3.
https://techinfo.porsche.com/techinfo/sit/en/996Carrera/2004/3_2004_996_gt3.pdf
https://techinfo.porsche.com/techinfo/sit/en/996Carrera/2004/3_2004_996_gt3.pdf
Re: Limited slip differentials
Quote:
Grant said:
The first percentage is the locking ratio while accelerating and the second is locking ratio under braking. 911's use a higher locking ratio for braking than for power (many people confuse this). The first asymetrical LSD's were used by Porsche on the 964 Turbo where it had 20% locking on the power and 80% locking on braking. This helps to steady the car while approaching and entering a corner while slowing. This benefit is at least as important (to me) as the reduction in wheelspin it provides at corner exit while on the power. 911's have such a strong rear-weight bias that not much locking is required for good traction under power. And any more than necessary just causes excessive understeer in tight corners.
What do the percentages mean - percent of what? This is harder to answer. The LSD has clutches and the number of clutches as well as the preload and ramps all contribute to the amount of torque that is required to "break" the axle loose (allow the wheels to turn independently). Not all 40% LSD's break loose at the same torque. This number can be somewhat misleading. Maybe the percentage is meant to reference the amount of torque the particular engine can provide. I'm not sure...
So grant,
An LSD, while being able to lock both wheels, are not always locked?! As in, able to act as a open differential under a normal U-turn, and when you hit a certain pressure coefficient - BAM - LSD???
Hope you understand what im trying to say
Re: Limited slip differentials
Quote:
TopherV said:So grant,
An LSD, while being able to lock both wheels, are not always locked?! As in, able to act as a open differential under a normal U-turn, and when you hit a certain pressure coefficient - BAM - LSD???
Hope you understand what im trying to say
Yes, the rear wheels are always locked together like a solid axle, until the differing torque between left and right is enough to break them apart (causes the internal clutches to slip). This does usually happen in U-turns and you can hear/feel the cogging effect as the clutches break free. Any GT3 owner can tell you this...
Most of the time you're driving around with LSD (no slipping) and when you take a U-turn, BAM - Open differential (left and right can independently turn freely, not the opposite).
Re: Limited slip differentials
Quote:
Grant said:
Most of the time you're driving around with LSD (no slipping) and when you take a U-turn, BAM - Open differential (left and right can independently turn freely, not the opposite).
Don't the ramps modify this situation slightly? What you describe is just the initial effect of the clutches -- i.e. the preload.
Do you know anything about the procedure they use to test the functionality of an LSD?
Stephen
Mar 1, 2006 3:38:12 PM
- FixedWing
- Veteran
-
- Loc:
- Posts: 870, Gallery
- Registered on: May 25, 2002
- Reply to: Ulf Karlsson
Re: Limited slip differentials
Greetings Ulf. 
There has been a lot of discussion recently of LSD's.
Look here:
http://www.rennteam.com/showflat.php?Cat=0&Number=190826
And here:
http://forums.rennlist.com/rennforums/showthread.php?t=252396
Stephen
There has been a lot of discussion recently of LSD's.
Look here:
http://www.rennteam.com/showflat.php?Cat=0&Number=190826
And here:
http://forums.rennlist.com/rennforums/showthread.php?t=252396
Stephen
Re: Limited slip differentials
Quote:
Grant said:
Most of the time you're driving around with LSD (no slipping) and when you take a U-turn, BAM - Open differential (left and right can independently turn freely, not the opposite).
I am trying to understand how asymmetrical locking LSD's work:
In a straight line under either acceleration or overrun, the asymmetrical LSD is 100% locked (similar to solid axle) which means 50% of the engine torque entering (through the input pinion shaft) goes to the left rear wheel and 50% to the right, correct?
As you turn left in a U-turn and let completely off the accelerator, engine braking/engine drag/overrun takes over and a 964 Turbo's 20/80 (20% acceleration/80% overrun) LSD is now only 80% locked and transfers 80% of the torque to the slower, inside left wheel and retains 20% torque for the faster, outside right wheel, correct? And if overrun is still present as the car enters and exits the apex, the differential stays locked in 80% mode, correct?
http://auto.howstuffworks.com/differential3.htm
http://members.aol.com/agspeed/LSD-lock.htm
Re: Limited slip differentials
Yes, you're right about the 50/50 torque split when driving straight, but I think you may be a tad confused when turning.
Try not to assign torque value to the inside vs. the outside wheel. The natural tendency of an open diff is to apply all the power to the spinning wheel without traction. 20% LSD on the power is enough in a 911 to deliver power to the outside wheel too through locking the two rears together. Some slipping occurs to allow for the outside wheel to travel a longer path (and not create too much understeer).
The 20% and 80% numbers I find are really only useful as relative numbers (4 times as much locking on overrun and where 100% would mean a solid rear axle). Just consider how strongly the left and right rear wheels are locked together. The diff would also be locked, even if the car were in neutral (no torque from the motor in either direction).
On overrun, the higher locking factor will keep the car like it has a solid rear axle for a longer period as it encounters differential torque (one rear wheel wants to turn faster than the other due to a turn or differences in traction). It doesn't send 80% of the torque to one wheel or another. Think of it as behaving like 80% of a solid axle (where it can only slip if it encounters enough differential torque), whereas it only behaves like 20% of a solid axle when under power (more easily behaves like an open diff as it encounters differential torque).
Try not to assign torque value to the inside vs. the outside wheel. The natural tendency of an open diff is to apply all the power to the spinning wheel without traction. 20% LSD on the power is enough in a 911 to deliver power to the outside wheel too through locking the two rears together. Some slipping occurs to allow for the outside wheel to travel a longer path (and not create too much understeer).
The 20% and 80% numbers I find are really only useful as relative numbers (4 times as much locking on overrun and where 100% would mean a solid rear axle). Just consider how strongly the left and right rear wheels are locked together. The diff would also be locked, even if the car were in neutral (no torque from the motor in either direction).
On overrun, the higher locking factor will keep the car like it has a solid rear axle for a longer period as it encounters differential torque (one rear wheel wants to turn faster than the other due to a turn or differences in traction). It doesn't send 80% of the torque to one wheel or another. Think of it as behaving like 80% of a solid axle (where it can only slip if it encounters enough differential torque), whereas it only behaves like 20% of a solid axle when under power (more easily behaves like an open diff as it encounters differential torque).
Re: Limited slip differentials
Quote:
Avoe said:
Is it possible to retrofit LSD.
I have a 996 Cabrio C2, without PSM!, so actually it schould be possible, right?
Is Porsche doing this?
Probably best to buy a used transmission (98 and 99 are only model years for Carrera - ROW) with LSD and sell yours...
Re: Limited slip differentials
1) If it is true the 997 GT3's locking numbers are 28/40 as opposed to the 996 GT3's 40/60 and 964 Turbo's 20/80, is it safe to infer the 911 suspension design and tuning over time is progressing towards the goal of having a symmetrical LSD like the CGT's 32/32? Or has the 28/40 been selected because this is the first GT3 model with TC?
2) Why does the 2005 motorsport season 997 GT3 Cup use a 40/60 LSD (without TC) when the 997 GT3 uses 28/40? Does this marked difference mean the 996 GT3 with its 40/60 is closer to its race and rally counterparts than the 997 GT3? Is the better to have TC and a 28/40 or a 40/60 without TC?
Sources: http://www.worldcarfans.com/news.cfm/new...t3-more-details
http://www.rennteam.com/showflat.php?Cat=0&Number=49287&an=0&page=1#49287
http://www.rennteam.com/showflat.php?Cat=0&Number=48366&an=0&page=1#48366
http://www11.porsche.com/usa/eventsandracing/motorsport/racingcars/996gt3cup/technologyandconcept/
2) Why does the 2005 motorsport season 997 GT3 Cup use a 40/60 LSD (without TC) when the 997 GT3 uses 28/40? Does this marked difference mean the 996 GT3 with its 40/60 is closer to its race and rally counterparts than the 997 GT3? Is the better to have TC and a 28/40 or a 40/60 without TC?
Sources: http://www.worldcarfans.com/news.cfm/new...t3-more-details
http://www.rennteam.com/showflat.php?Cat=0&Number=49287&an=0&page=1#49287
http://www.rennteam.com/showflat.php?Cat=0&Number=48366&an=0&page=1#48366
http://www11.porsche.com/usa/eventsandracing/motorsport/racingcars/996gt3cup/technologyandconcept/
Re: Limited slip differentials
Quote:
MAVERICK said:
1) If it is true the 997 GT3's locking numbers are 28/40 as opposed to the 996 GT3's 40/60 and 964 Turbo's 20/80, is it safe to infer the 911 suspension design and tuning over time is progressing towards the goal of having a symmetrical LSD like the CGT's 32/32? Or has the 28/40 been selected because this is the first GT3 model with TC?
2) Why does the 2005 motorsport season 997 GT3 Cup use a 40/60 LSD (without TC) when the 997 GT3 uses 28/40? Does this marked difference mean the 996 GT3 with its 40/60 is closer to its race and rally counterparts than the 997 GT3? Is the better to have TC and a 28/40 or a 40/60 without TC?
Yes, I think you are right. I think the lower locking percentages in the 997GT3 are because of Traction Control (and its additional features that make it sort of half way to a PSM system - it has some function under braking, unlike any TC before it).
I believe that the locking numbers are symetrical (didn't know this) on the CGT would probably be due to its weight distribution being mid-engined, more than its advanced suspension...
Re: Limited slip differentials
Quote:
Grant said:
I believe that the locking numbers are symetrical (didn't know this) on the CGT would probably be due to its weight distribution being mid-engined, more than its advanced suspension...
If we assume it's probably more due to it being mid-engined, why does the Enzo get a 30/55 and the 360 a 25/45?
Source: http://www.ferraricars.org/ferrari-enzo/transmission.html
Re: Limited slip differentials
Quote:
MAVERICK said:Quote:
Grant said:
I believe that the locking numbers are symetrical (didn't know this) on the CGT would probably be due to its weight distribution being mid-engined, more than its advanced suspension...
If we assume it's probably more due to it being mid-engined, why does the Enzo get a 30/55 and the 360 a 25/45?
Source: http://www.ferraricars.org/ferrari-enzo/transmission.html
Good question! Although I wouldn't say that the Enzo's suspension is any less advanced than the CGT's, would you?
I guess I'd like to change my answer and say that LSD is part of the suspension and it's likely that the CGT's suspension was tuned with this symetrical LSD to achieve the desired handling balance. Changing the locking ratios will affect under/oversteering behavior and the LSD and suspension spring (and swaybar) rates need to be developed together as a whole system. I guess the engineers at Weissach felt this was their preference...
Re: Limited slip differentials
Quote:
Grant said:
Good question! Although I wouldn't say that the Enzo's suspension is any less advanced than the CGT's, would you?
No, sir, I would not.
Quote:
Grant said:
I guess I'd like to change my answer and say that LSD is part of the suspension and it's likely that the CGT's suspension was tuned with this symetrical LSD to achieve the desired handling balance. Changing the locking ratios will affect under/oversteering behavior and the LSD and suspension spring (and swaybar) rates need to be developed together as a whole system. I guess the engineers at Weissach felt this was their preference...
Excellent point, Grant. I never thought of the LSD being part of the suspension and suspension tuning. And may I say I am enjoying picking your brain.
Now, about cone/clutch-type LSD's versus torque sensitive/torsen LSD's:
Why would Koenigsegg use a torque sensitive differential when Porsche, Ferrari, Pagani and racing teams usually do not? Yet, the CCR is still able to generate excellent lap times. Are not torque sensitive LSD's inferior to cone/clutch LSD's under conditions where the inside wheel lifts off the ground?
Re: Limited slip differentials
Yes, torque sensing diffs are inferior to clutch diffs for tight tracks. They provide no stabilizing effect under braking (especially useful for trail-braking into corners). I'm not sure what happens to a torsen diff with one wheel in the air...
Perhaps the CCR would be even faster with a clutch LSD on a tight course. Noble and Lotus also use torsen LSD's.
Perhaps the CCR (and the Brits) use a Torsen diff due to its two main advantages (less understeer in tight corners and totally maintenance-free - no clutches to replace)
Perhaps the CCR would be even faster with a clutch LSD on a tight course. Noble and Lotus also use torsen LSD's.
Perhaps the CCR (and the Brits) use a Torsen diff due to its two main advantages (less understeer in tight corners and totally maintenance-free - no clutches to replace)
Re: Limited slip differentials
Quote:
Grant said:
I'm not sure what happens to a torsen diff with one wheel in the air...
The one downside of the Torsen is that if one wheel is completely off the ground (zero traction) then the characteristics of the Torsen will make it basically an open diff, both wheels have to have at least some traction for the unit to function properly.
What about Porsche 959 PSK (pioneer) vs. BMW M Variable Differential Lock vs. Ferrari E-Diff vs. Honda's SH-4WD system vs. Mercedes 4-Matic vs. Mitsubishi ACD/AYC vs. Nissan Skyline GT-R's ATTESA E-TS PRO vs. VW/Audi Haldex, would you agree with the following quote regarding the "E-Diff" active, electrohydraulic differential:
Ferrari claims this technology was transferred from its F1 cars and it is the world's first for production car. In fact, there is no fundamental difference between E-Diff and Mitsubishi's AYC or BMW's Variable M-differential. Well, it's a good technology, just don't mention "world's first" or "F1 technology" in every sentence. The real F1 technology is "Manettino", a steering wheel-mounted control interface for various systems.
Reference Sources:
http://www.miata.net/garage/vlsd.html
http://www.autozine.org/html/Ferrari/F430.html
http://www.italiaspeed.com/new_models/new_models_2004/f430/diff.html
http://www.autoworld.com/news/bmw/m_difflock.asp
http://www.canadiandriver.com/articles/jk/at_011219.htm
http://www.autozine.org/technical_school/traction/tech_traction_4wd_21.htm
http://www.autozine.org/html/Mitsubishi/Lancer.html#E8
Re: Limited slip differentials
Yeah, all of those systems are electronically (rather than purely mechanical) LSD's that are modulated by integrating them into an array of sensors (accelerometers) to maximize traction and stability under a variety of conditions.
The one in the Mitsubishi is the most advanced (complicated) as it operates as part of an AWD system with 3 differentials. The BMW one is the most basic (no AWD and only one setting), while the Ferrari system is adjustable for different levels via the Manettino. But, they are all more similar than different.
If Porsche ever uses this technology again in their GT cars, it will allow integration with PSM for the first time.
I have very little seat time with any of these electronic LSD's, so I don't know if it's truly better for the driver or not...
The one in the Mitsubishi is the most advanced (complicated) as it operates as part of an AWD system with 3 differentials. The BMW one is the most basic (no AWD and only one setting), while the Ferrari system is adjustable for different levels via the Manettino. But, they are all more similar than different.
If Porsche ever uses this technology again in their GT cars, it will allow integration with PSM for the first time.
I have very little seat time with any of these electronic LSD's, so I don't know if it's truly better for the driver or not...
Forum
About Rennteam | Help | Terms of use | Privacy Policy | Support | Contact | Imprint | Archive
Powered by
Community
and
