Why not Sti What's the differences

[Samiam1017]

It says in the what you need section 02-07 impreza or wrx (except Sti). Why is this. What's the differences between the two?

[fateo66]

As of right now, FFR is not supporting the STI's 6 speed transmission because its slightly longer and might interfere with the body. Also the front hubs on any 05+ STI are a different design and wont work with the new upper control arm that FFR's designing. But other then those two thing you could still use nearly everything else from the STI; motor, steering rack, control arms... etc

[BipDBo]

The problem with the STI six speed was more than the size. There was some issue with converting it to 2WD because of the way the center differential works. There was a lot of discussion about this in the forum early on, but finding it wouldn't be easy.

They eventually made a custom transaxle for the GTM so I'd imagine that it's only a matter of time before someone finds a workaround for this issue.

[Nuul]

Yeah, I imagine someone will find a way to make it work. In the meantime if you're worried about your gears breaking you can always use a few of the lower STI gears in your WRX trans. I want to use gears 1-3 from an STI but it's cost prohibitive right now.

[StatGSR]

The problem with the STI six speed was more than the size. There was some issue with converting it to 2WD because of the way the center differential works. There was a lot of discussion about this in the forum early on, but finding it wouldn't be easy.


that's just not true, there are conversion kits available for the 6spd already, not sure why so many people think its impossible, it has been covered several times on this forum already....

but as mentioned above, the STI is currently not a Single Donor, that's why its not "approved"

[Niburu]

that's just not true, there are conversion kits available for the 6spd already, not sure why so many people think its impossible,

please post some links to said conversion kits

[fateo66]

please post some links to said conversion kits


They already exist and are around $400 from what I hear


http://www.possumbourne.co.nz/trans_diff_parts.html

http://www.possumbourne.co.nz/images/products/Center-Diff-locking-tube1.jpg

Here you go

[BipDBo]

I vaguely remember Dave saying something to the effect early on. I don't understand the issue, but I really doubt that they would be limiting the donor list if the only reason was the length of the transmission. That could be fixed with a sawsall.

[Etos]

Weight is also an issue. The 6speed is EXTREMELY heavy. It's also on the far end of the car, that's alot of weight on the very tip of a mid engine car.

An 06+ WRX trans will have no problem handling most peoples needs. You're gonna need well over 400hp to start needing more then the 5 speed. Even then I'd take a beefed up 5 speed over the 6 based on weight alone unless it's a drag only car.

[StatGSR]

besides the weight difference, the transmission mount points are also different for the STI (same as a subaru automatic). in any case, you should be able to basically trade a 6spd for a ppg 5spd, or like 4 regular 5spds, i don't see the issue, but i also expect it to be workable in the future. it is a kit car people, you CAN build it however you want....

[fateo66]

besides the weight difference, the transmission mount points are also different for the STI (same as a subaru automatic). in any case, you should be able to basically trade a 6spd for a ppg 5spd, or like 4 regular 5spds, i don't see the issue, but i also expect it to be workable in the future. it is a kit car people, you CAN build it however you want....

That is not true, the 6MT is the same length as the 4EAT and therefore can use the same driveshaft. But the mounting point is in the exact same location as the 5MT, that is why you can use the same transmission crossmember when you do a 6mt swap.

[StatGSR]

That is not true, the 6MT is the same length as the 4EAT and therefore can use the same driveshaft. But the mounting point is in the exact same location as the 5MT, that is why you can use the same transmission crossmember when you do a 6mt swap.

I was under the impression that it used the mount locations of the auto (crossmember to chassis), i was pretty sure that my friend had to swap to the different tranny crossmember when he swapped one in his legacy, but maybe i was mistaken it was about 4 years ago now ( i checked, your right you can use a 5spd crossmember with a 6spd box). in any case, i know for a fact that the rubber mount that connects the tranny to the crossmember is different from the 5spd to the 6spd.

i agree 100% that the auto and sti use the same drive shaft, but that is really not relevant to this discussion.

[Etos]

I was under the impression that it used the mount locations of the auto (crossmember to chassis), i was pretty sure that my friend had to swap to the different tranny crossmember when he swapped one in his legacy, but maybe i was mistaken it was about 4 years ago now ( i checked, your right you can use a 5spd crossmember with a 6spd box). in any case, i know for a fact that the rubber mount that connects the tranny to the crossmember is different from the 5spd to the 6spd.

i agree 100% that the auto and sti use the same drive shaft, but that is really not relevant to this discussion.

The location is exactly the same, as is the crossmembers(it's got 3 parts). The trans mount for the 5 speed is much bigger since the crossmember being the same and the trans being smaller then the 6 speed, there's a bigger gap to fill.

Subarus are like legos, they make most of the parts between impreza to sti interchangeable. Obviously there are variations on specific parts(hubs being one) but for the most part they share alot.

[Funky]

Newbie here, but I've tried to do lots of reading on 818...

One important difference between WRX and the STI is the LSD (Limited Slip Diff) in the STI. What are the options when it comes to LSD with 818?

[Etos]

Newbie here, but I've tried to do lots of reading on 818...

One important difference between WRX and the STI is the LSD (Limited Slip Diff) in the STI. What are the options when it comes to LSD with 818?

That depends on what trans you're going with. The WRX 2.0 engine trans(02-05) is slightly different then the 2.5 engine(06+) and the non turbo trans are different too. So figure out what trans you are going with, what you want from the diff(there's many different kinds of diffs) and go from there.

[papajon1000]

Here you go


Center Diff Lock Tube
Can be used in FWD, RWD and 4WD.

Our Price: $ T.B.A



I have looked into this and it looks to me that may make them but not sell them.

Jon

[papajon1000]

Newbie here, but I've tried to do lots of reading on 818...

One important difference between WRX and the STI is the LSD (Limited Slip Diff) in the STI. What are the options when it comes to LSD with 818?

As Etos stated it depends on your transmission choice but if you search google it will give you many options. I know there are some JDM sti 5 speed transmissions that came stock with a front LSD as well.

Jon

[BrandonDrums]

Newbie here, but I've tried to do lots of reading on 818...

One important difference between WRX and the STI is the LSD (Limited Slip Diff) in the STI. What are the options when it comes to LSD with 818?

A popular and cheap front LSD that many 5mt Subaru guys use is just an OBX torsen-type LSD. Can be had for under 400 bucks and works great for track but might not be the best for fun/drifting since it's not a 50/50 locker.

More knowledgeable guys might have other specific recommendations based on your goals like drifting, autox, drag-racing, high power or whatever. Most of my understanding about LSD's is simply around the best types to reduce understeer in AWD configuration and the torsen-type LSD is about the best...and cheapest (from OBX at least) in that regard from what I understand.

[StatGSR]

the OBX is cheap, but i have never seen a recommendation that didn't start with "take it apart and re-torque it first, then its fine".... just something to chew on...

[shinn497]

I wonder if LSD would be wise since the drive train is reversed. Also, as I understand, MR cars tend to oversteer.

[jimgood]

I wonder if LSD would be wise since the drive train is reversed. Also, as I understand, MR cars tend to oversteer.
What drive train is reversed?

[shinn497]

Normally the subaru drives the front wheels with the transaxle. In the 818 it will be the rear. Much much more/less would rear wheels benefit from an LSD engineered for front wheels. In particular would it be necessary to retune the LSD to correct oversteer rather than understeer.

With that said, the GTM has dialed in understeer and it is also mid-engine.

[Niburu]

What drive train is reversed?

not reversed so much as moved back
what I think shinn is trying to get at is the lock-up point on the diff needs to tweaked
for it's new position you will want a different engagement point
I honestly don't know whats possibe on a torsen though

[PhyrraM]

I'm no expert, but in such a lightweight car it seems like a torsen-type might be better. A clutch type would likely need to run 'loose' anyways because a truely locked rear diff will make such a light car 'push' or understeer more than the same diff in a heavier car.

[wjfawb0]

I've owned two 05 STIs and I have driven a few WRX models. The crappiest thing about an STI 6 spd to me is the close ratio gears. It was kind of annoying when daily driving how short first gear was. I'd much rather have higher gearing in the first couple gears on a lightweight car like the 818.

[Etos]

My understand of diffs is a little sketchy but from what I understand, a LSD or even locked diff is highly preferable to an open diff on the track. If you run an open diff, at a turn the inside wheel is going to have the least traction thus that is where the an open diff wants to send power. When that happens it's taking power away from the outside wheel which is helping turn in and trying to give that power to the inside wheel, making the car want to turn out.

With an LSD, it keeps the power even between the 2 wheels thus more control over what the rear end does. I think the notion that an LSD is more likely to oversteer the car comes from people not knowing how the drive well. Oversteering an LSD just means you are overcoming the traction of both wheels and thus the back end kicks out.

[Samiam1017]

Thx for all the great info. Between this thread and wrx experts tread a lot of good info is being put out there. Ok know since most donors that are gonna be bought are gonna be very high mileage cars what's the best place for remanufactured long blocks or companies that rebuild and what price range would that land in. Thanks

[Nuul]

Depends on your pocketbook but there are a few options. It's usually the bottom end that gives out these engines - a spun bearing to be specific.


Stock WRX: Assembled EJ205 block, rods, crank and pistons (Subaru part #10103AB330). You bolt on the rest of the engine from your donor if it's in good shape. You shouldn't have to do much if any tuning for this option since you're basically stock.


STI Hybrid: Assembled EJ257 block, rods, crank and pistons (Subaru part #10103AC050 for nitrated crank #10103AC030 for normal). You bolt on the rest of your engine but you use the STI gasket set and you have your heads sent to Equilibrium Tuning (http://eqtuning.com/products.php) to be CNC'd to match the STI head chamber. You'll also need bigger injectors, fuel pump, a good tune and some odds n' ends to make this work right. The upside is that you get way more torque than you had with the 2.0L. But it's a lot of effort so this isn't for everyone.


Buy a re-manufactured engine for probably more than you spent on the donor to begin with.


Purchase the individual components that are showing wear. This is probably the most economical way if you are comfortable bonding the two halves of the engine together with Fujibond.

[Samiam1017]

I was accually thinking stock. Just bringing a 200000 mile motor back to life. Just for reference I would think most people would want to stick a fresh stock motor in the car and do upgrades later

[shinn497]

Also don't forget JDM engines! They are build to tight tolerances most versions feature forged internals. Japan has strict emissions laws that make owning a car past 50k very expensive so these engines have low miles and are plenty available.

The ej207 v7/v8 are the most popular. The v7 can make big power (totally forged) and the v8 has a twin scroll turbo (faster spooling). Both rev to 8k with their more advanced AVSC.

[Xusia]

If you run an open diff, at a turn the inside wheel is going to have the least traction thus that is where the an open diff wants to send power. When that happens it's taking power away from the outside wheel which is helping turn in and trying to give that power to the inside wheel, making the car want to turn out.

With an LSD, it keeps the power even between the 2 wheels thus more control over what the rear end does. I think the notion that an LSD is more likely to oversteer the car comes from people not knowing how the drive well. Oversteering an LSD just means you are overcoming the traction of both wheels and thus the back end kicks out.


I think you have that backwards. An open diff would allow the wheels to turn at wildly different rates, whereas a limited slip won't - it would limit the difference in wheel speed. In a very tight turn, the outer wheel needs to turn much faster than the inner wheel. An open differential will allow this. A limited slip differential is going limit the difference in wheel speed, sending more torque to the slower wheel (this is the INNER wheel in a turn) which is going to tend to cause the car to want to go straight (i.e. understeer).

http://auto.howstuffworks.com/differential2.htm

[Etos]

I think you have that backwards. An open diff would allow the wheels to turn at wildly different rates, whereas a limited slip won't - it would limit the difference in wheel speed. In a very tight turn, the outer wheel needs to turn much faster than the inner wheel. An open differential will allow this. A limited slip differential is going limit the difference in wheel speed, sending more torque to the slower wheel (this is the INNER wheel in a turn) which is going to tend to cause the car to want to go straight (i.e. understeer).

http://auto.howstuffworks.com/differential2.htm

But that is what I said :confused:

[Xusia]

The effect you described, where the inside wheel receives more power and therefore tends to make the car want to turn out - is true of an LSD, not of an open diff. Hence my statement I think you have it backwards. Sorry for not being more clear! :)

[StatGSR]

No, inside wheel spins with an open diff, because it is unloaded it will spin first if you are in the throttle. An open diff will send almost all power to the spinning wheel because it is the path of least resistance. The LSD will force power to still go to the outside wheel and will help reduce the wheel spin at the inside wheel...

I am talking about when the car looses traction, not just the differance of wheel speed under a no or low power situation.

[Xusia]

An open diff will send almost all power to the spinning wheel because it is the path of least resistance.

Which is the OUTSIDE wheel...

[Flashburn]

the OBX is cheap, but i have never seen a recommendation that didn't start with "take it apart and re-torque it first, then its fine".... just something to chew on...

This took me under an hour and it was the first time for me.

[Nuul]

Also don't forget JDM engines! They are build to tight tolerances most versions feature forged internals. Japan has strict emissions laws that make owning a car past 50k very expensive so these engines have low miles and are plenty available.

The ej207 v7/v8 are the most popular. The v7 can make big power (totally forged) and the v8 has a twin scroll turbo (faster spooling). Both rev to 8k with their more advanced AVSC.

I asked my father in law about that when I was in Japan last summer, I had heard that too. He looked at me funny when I said I had heard it was cheaper to buy a new car than keep an old one plated. Apparently there is a tiered taxation for cars but it seemed to have more to do with engine size. I remember the 500, 300 and (I think) 50...though I can't remember the exact designation for tiny car. I guess this was to persuade people into driving the small cars to help reduce how much oil they import.

As for the JDM engines I don't know that they are worth the premium you have to pay for them. I would LOVE to have an EJ207 v8 but for what it would cost me I just can't justify it.

[Flashburn]

I think you have that backwards. An open diff would allow the wheels to turn at wildly different rates, whereas a limited slip won't - it would limit the difference in wheel speed. In a very tight turn, the outer wheel needs to turn much faster than the inner wheel. An open differential will allow this. A limited slip differential is going limit the difference in wheel speed, sending more torque to the slower wheel (this is the INNER wheel in a turn) which is going to tend to cause the car to want to go straight (i.e. understeer).

http://auto.howstuffworks.com/differential2.htm

What type of LSD are you talking about?
Torsens?

[Funky]

Which is the OUTSIDE wheel...

Actually, it is the INSIDE wheel, because if you think about it, weight is transferred to the outside wheel in a turn right? An open diff sends power whichever way it is easier. Therefore if you have one wheel slightly off the ground, almost all of the power goes to that wheel because it requires less effort.

This problem is notorious with the low-end non-LSD versions Lotus Elise, which is why this subject concerns me with 818. There is a review video of the Elise demonstrating this that I'll try to find for you guys.

When this happens you will get a loss of power delivered to the , and therefore reduced directional control. StatGSR wrote a pretty good explanation (http://thefactoryfiveforum.com/showthread.php?5661-Why-not-Sti-What-s-the-differences&p=56624&viewfull=1#post56624).

Additionally, an LSD does allow difference in rotation speeds, but limits that to a certain amount. Therefore, an LSD is not truly 'Locked'. Another important point is that it ensures that both wheels get at least a certain MINIMUM amount of power at all times (for example 40% minimum on each side). I have personally noticed a huge difference driving cars with LSD in the canyons vs. non-LSDs, and that is also because of the rather symmetrical distribution of power. Not only to me it seem the threshold was higher, the car was more predictable as well, leading to better performance. If you can't have a GTR style intelligent asymmetric power distribution, the next best thing is symmetric before an open diff. After all, if LSD was a bad thing, you wouldn't see it in ALL super/high performance cars in production today.

[Funky]

Which is the OUTSIDE wheel...

Actually, it is the INSIDE wheel, because if you think about it, weight is transferred to the outside wheel in a turn right? An open diff sends power whichever way it is easier. Therefore if you have one wheel slightly off the ground, almost all of the power goes to that wheel because it requires less effort.

This problem is notorious with the low-end non-LSD versions Lotus Elise, which is why this subject concerns me with 818. There is a review video of the Elise demonstrating this that I'll try to find for you guys.

When this happens you will get a loss of power delivered to the , and therefore reduced directional control. StatGSR wrote a pretty good explanation (http://thefactoryfiveforum.com/showthread.php?5661-Why-not-Sti-What-s-the-differences&p=56624&viewfull=1#post56624).

Additionally, an LSD does allow difference in rotation speeds, but limits that to a certain amount. Therefore, an LSD is not truly 'Locked'. Another important point is that it ensures that both wheels get at least a certain MINIMUM amount of power at all times (for example 40% minimum on each side). I have personally noticed a huge difference driving cars with LSD in the canyons vs. non-LSDs, and that is also because of the rather symmetrical distribution of power. Not only to me it seem the threshold was higher, the car was more predictable as well, leading to better performance. If you can't have a GTR style intelligent asymmetric power distribution, the next best thing is symmetric before an open diff. After all, if LSD was a bad thing, you wouldn't see it in ALL super/high performance cars in production today.

[StatGSR]

Which is the OUTSIDE wheel...

Um, no. Because of body roll and camber, the outside wheel has plenty of traction, the inside is the one that will lift and lose traction and spin (with an open diff). Seriously, this isn't up for debate, thinking otherwise just makes you wrong.

Maybe there is confusion about which side is the outside, if your making a right hand turn, the left side of the car is the outside and visa versa

If both tires have traction, the outside will complete more rotations than the inside, that is because it has a longer travel path than the inside. the diff "slips" to allow this to be done, but an open diff or and LSD will allow this to happen, only a fully locked or welded will not allow this to happen smoothly ....

An LSD still allows the inner and outer wheels to spin at slightly differant speeds just like an open diff, so that both can make the correct amount of rotations to go around a corner, but unlike an open diff it will not allow large diffances in wheel spin, which basically means it wont leave you doing a one wheel burnout when one wheel loses traction.

[PhyrraM]

......, only a fully locked or welded will not allow this to happen smoothly ....

Or, possibly, a stiff LSD diff in a very light car. I'm just suggesting that it's possible the "better" diffs for a 3000 pound car are maybe not the same ones as for a 1800 pound car.

[StatGSR]

Thanks for being nit picky PhyrraM, we haven't even begun to talk about different types of LSDs here...

[Nelff]

OMG to the diff debate... If you want the inside wheel to spin and loose all power to motivate forward, by all means, keep an open diff. If, when the inside wheel unloads and starts to spin you want both wheels to lock up and spin at the same rate, get a 'normal' limited slip diff. (which is good for steering the back end at slo speeds) If, you want the power delivery to be split evenly at the limit where one or both wheels are trying to spin, get a torsen dif...

[el_jefe]

<sarcasm> I'm planning on using a Lincoln Locker on mine </sarcasm>

[Xusia]

Um, no. Because of body roll and camber, the outside wheel has plenty of traction, the inside is the one that will lift and lose traction and spin (with an open diff). Seriously, this isn't up for debate, thinking otherwise just makes you wrong.

Maybe there is confusion about which side is the outside, if your making a right hand turn, the left side of the car is the outside and visa versa

If both tires have traction, the outside will complete more rotations than the inside, that is because it has a longer travel path than the inside. the diff "slips" to allow this to be done, but an open diff or and LSD will allow this to happen, only a fully locked or welded will not allow this to happen smoothly ....

An LSD still allows the inner and outer wheels to spin at slightly differant speeds just like an open diff, so that both can make the correct amount of rotations to go around a corner, but unlike an open diff it will not allow large diffances in wheel spin, which basically means it wont leave you doing a one wheel burnout when one wheel loses traction.

Even if I have that wrong, which I may, my research indicates my original assertion is still correct: That an LSD has a greater tendency to understeer than an open differential. This is the opposite of what was originally said. I'd like to know the truth, so if this is wrong, please provide some sources I can learn from! :)

[el_jefe]

Even if I have that wrong, which I may, my research indicates my original assertion is still correct: That an LSD has a greater tendency to understeer than an open differential. This is the opposite of what was originally said. I'd like to know the truth, so if this is wrong, please provide some sources I can learn from! :)

It depends on the differential. Posi or a solid rear, sure, it'll make it push a bit. But with enough power, you can control the cars yaw with rear wheelspin. But most limited slips are just that, limited. They allow some slip so that the car doesn't understeer. I have driven both, and added both types to cars that didn't have them before, and if you have any kind of power going to the ground, having a limited slip is much more to your advantage than the possibility of some slight understeer.

[Etos]

The effect you described, where the inside wheel receives more power and therefore tends to make the car want to turn out - is true of an LSD, not of an open diff. Hence my statement I think you have it backwards. Sorry for not being more clear! :)

I think there needs to be more clarity on the situation. I am not talking about driving along normally on the highway/around town/off the track. For this- anything but a completely locked diff will work well.

I am talking about at the track/autox. You are confusing the rotations of the inside/outside wheel during a turn with an loaded/unloaded wheel at a turn. For instance, you are making a hard left turn. All the weight is going toward the right side(the outside). This wheel has not just the weight of the car but all the force of that weight at speed on it. The left side(the inside) is losing all this weight, thus it is becoming unloaded. With an open diff, it wants to send power to the path of least resistance. The outside wheel has the most resistance since it has all the force on the car on it while the inside wheel is losing resistance.

I am not saying ALL the power is being lost to the outside- only that power is being transferred and that amount changes on almost infinite variables of tires, diff, weight, speed, turn radias, suspension setup, front to back weight transfer, amount of power being sent etc etc. Nor am I talking about the difference in rotations an outside wheel makes over the inside.

[Evan78]

Even if I have that wrong, which I may, my research indicates my original assertion is still correct: That an LSD has a greater tendency to understeer than an open differential. This is the opposite of what was originally said. I'd like to know the truth, so if this is wrong, please provide some sources I can learn from! :)
An open diff can limit oversteer by preventing the outside wheel from receiving sufficient power to break traction. In an extreme example, if you try to do a donut in a car with an open differential, you may not be able to deliver enough power to the outside wheel (the one with all the traction) to overpower its available traction. You'll just end up spinning the inside tire without much donut action.

[Oppenheimer]

OK, so now we can't even agree on the definition of a 'donut'. To me, an open diff that results in laying down a simple, single strip circle of rubber is a classic donut.

What you are describing is a power-slide.

Is there a term to differentiate between over/understeer that occurs when tires are merely slipping vs that which occurs when tires have broken the threshold into full on sliding? It seems such a term could avoid a lot of confusion when discussing these things. A car that has strong understeer tendencies below threshold can easily become a fish-tailing monster with the simple application of a bit more throttle that pushes the back tires past threshold (for example high-powered RWD LSD equipped).

[Evan78]

Using your term, I'd define a donut as any power slide covering at least 360 degrees. I don't think it matters, I was just providing an example. Power slide, donut, whatever you want to call it, they are all oversteer conditions that are affected by how the rear diff behaves.

To answer you question, I'm not aware of any simple terms to differentiate the situation. I just say a car tends to over/understeer on corner entry or corner exit, etc.

[RossLH]

Even if I have that wrong, which I may, my research indicates my original assertion is still correct: That an LSD has a greater tendency to understeer than an open differential. This is the opposite of what was originally said. I'd like to know the truth, so if this is wrong, please provide some sources I can learn from! :)

A limited slip diff will oversteer on throttle during hard cornering in a RWD car. Imagine the scenario in which the front wheels are at their lateral grip limit, as is the outside rear wheel, and the inside rear wheel is spinning (such will happen with an open diff). When that rear differential locks up, the outside rear wheel will get a surge of power, and when that happens it will lose grip and the back end will come around. (Source: I'm a mechanical engineer with a background in autocrossing LSD'd S2000's.)

You can read up all you want, and ultimately what you'll find is that your current conception is misled. I recommend you actually get behind the wheel of a RWD car with an open diff, and subsequently a similar car with an LSD. The difference is very clear.

[BrandonDrums]

Even if I have that wrong, which I may, my research indicates my original assertion is still correct: That an LSD has a greater tendency to understeer than an open differential. This is the opposite of what was originally said. I'd like to know the truth, so if this is wrong, please provide some sources I can learn from! :)

To go along with the other replies, here's my shot at it.

It sounds like you're doing most of your LSD research on NASIOC where everyone has AWD as your statement is true but only primarily when the LSD is tight, splits torque 50/50 and placed on the front axle for a FWD or AWD car. For instance, on a FWD car with a tight LSD going into a turn, if there's resistance to allowing the wheels turn at different rates in the absence of forward power then there's a loss of grip on turn-in as one tire must slip to make a radius with the other front tire. Most of where anyone would want to put such a diff on the front axle is for rally where wheelspin is inevitable and the name of the game is throwing dirt and gravel to make your traction and you want the lock to happen sooner on throttle-in than on tarmac.

A regular, streetable LSD on a regular, streetable FWD car won't understeer on turn in but it might understeer more than an open-diff as the 50/50 lock might cause both the inside (unloaded) and outside (loaded) tires to spin from torque and thus loosing cornering grip as well. An Open-diff FWD car would still have one tire gripping through the corner. In a power-on scenario, an open diff will result in less understeer on a FWD than an traditional 50/50 split LSD but initial turn-in is unaffected.

To counter that scenario, a fancy electronic torsen style diff is what makes the famous euro-spec Ford Focus RS handle 300 HP in FWD format so well. The Wheels can turn at their separate rates and when slip begins, the diff applies torque proportionate to the grip of each wheel. That's where an LSD on the front axle reduces understeer, it's a different type of LSD that distributes torque in a very different way.

In RWD application, at lower speeds a tight 50/50 split LSD would cause some initial understeer but would mostly aid in oversteer. A good example is a drift car which has no rear diff at all, they drag one tire around when driving in the pits all the time making it hard to turn since the diff is always locked. Once slip happens, both rear wheels are spinning at the same rate making them both loose grip which results in oversteer but in a controllable way since both rear wheels are spinning at the exact same rate.

A regular clutch LSD allows the wheels to turn at different rates until there's a significant difference in wheelspeeds THEN it locks. So any LSD anyone will be putting in an 818 wouldn't cause understeer at low or high speeds, it would only aid in power-on traction and make oversteer situations more controllable.

Hope that helps.

[Xusia]

Thanks, guys. That's given me a lot to think about. I was thinking of "slip" as the difference in tire rotational speed only without regard to load or traction loss. I appreciate the honest efforts to educate me! :)

One thing I still don't understand, though, is the behavior of an LSD after lock up, but before traction is lost. How/where is the power distributed? Is the axle truly locked causing both wheels to rotate at the same speed? Or is the wheel speed of the inside wheel limited and power therefore transferred to the outside wheel? Or something else?!?

[SkiRideDrive]

To go along with the other replies, here's my shot at it.

It sounds like you're doing most of your LSD research on NASIOC where everyone has AWD as your statement is true but only primarily when the LSD is tight, splits torque 50/50 and placed on the front axle for a FWD or AWD car. For instance, on a FWD car with a tight LSD going into a turn, if there's resistance to allowing the wheels turn at different rates in the absence of forward power then there's a loss of grip on turn-in as one tire must slip to make a radius with the other front tire. Most of where anyone would want to put such a diff on the front axle is for rally where wheelspin is inevitable and the name of the game is throwing dirt and gravel to make your traction and you want the lock to happen sooner on throttle-in than on tarmac.

A regular, streetable LSD on a regular, streetable FWD car won't understeer on turn in but it might understeer more than an open-diff as the 50/50 lock might cause both the inside (unloaded) and outside (loaded) tires to spin from torque and thus loosing cornering grip as well. An Open-diff FWD car would still have one tire gripping through the corner. In a power-on scenario, an open diff will result in less understeer on a FWD than an traditional 50/50 split LSD but initial turn-in is unaffected.

To counter that scenario, a fancy electronic torsen style diff is what makes the famous euro-spec Ford Focus RS handle 300 HP in FWD format so well. The Wheels can turn at their separate rates and when slip begins, the diff applies torque proportionate to the grip of each wheel. That's where an LSD on the front axle reduces understeer, it's a different type of LSD that distributes torque in a very different way.

In RWD application, at lower speeds a tight 50/50 split LSD would cause some initial understeer but would mostly aid in oversteer. A good example is a drift car which has no rear diff at all, they drag one tire around when driving in the pits all the time making it hard to turn since the diff is always locked. Once slip happens, both rear wheels are spinning at the same rate making them both loose grip which results in oversteer but in a controllable way since both rear wheels are spinning at the exact same rate.

A regular clutch LSD allows the wheels to turn at different rates until there's a significant difference in wheelspeeds THEN it locks. So any LSD anyone will be putting in an 818 wouldn't cause understeer at low or high speeds, it would only aid in power-on traction and make oversteer situations more controllable.

Hope that helps.

I disagree slightly. I believe how tight your differential is tuned will definitely effect the understeer characteristics of the car not only on corner exit but on corner entry as well. Changing the preload if you will on the diff can effect not only the amount of engine torque going to each wheel, effectively changing the rotational characteristics of the car, but it also effects longitudinal acceleration as a whole which will effect load transfer and thus indirectly effecting the handling characteristics as well. In my SAE days we would normally search for the minimal amount of preload necessary to get the car off the corner quickly as to have minimal disruption on the balance of the car. Tuning for corner entry independently of corner exit through diff tuning however is a tricky business and I never quite got to that level.

Great discussion.

[BrandonDrums]

I disagree slightly. I believe how tight your differential is tuned will definitely effect the understeer characteristics of the car not only on corner exit but on corner entry as well. Changing the preload if you will on the diff can effect not only the amount of engine torque going to each wheel, effectively changing the rotational characteristics of the car, but it also effects longitudinal acceleration as a whole which will effect load transfer and thus indirectly effecting the handling characteristics as well. In my SAE days we would normally search for the minimal amount of preload necessary to get the car off the corner quickly as to have minimal disruption on the balance of the car. Tuning for corner entry independently of corner exit through diff tuning however is a tricky business and I never quite got to that level.

Great discussion.

You're absolutely right. I thought I had kinda covered that with the Rally Car example where an LSD on the front wheels with high pre-load (I just said "tight") causes understeer on turn in. Although I didn't also point out that scenario would cause understeer on corner exit and even mid-corner on power.

[Oppenheimer]

One thing I still don't understand, though, is the behavior of an LSD after lock up, but before traction is lost. How/where is the power distributed? Is the axle truly locked causing both wheels to rotate at the same speed? Or is the wheel speed of the inside wheel limited and power therefore transferred to the outside wheel? Or something else?!?

This is where the 'Limited' in LSD comes in. If it where a full locker, then when it locks, both axles (left and right) are locked into the same rotational speed. Can you say 'power-slide'? With an LSD, the whole idea is to push some of the torque to the axle with less traction, without fully locking, allowing them to still rotate at different speeds. But send gobs more power than the tires can handle on the surface you are on, and even an LSD will effectively fully lock.

There are all sorts of LSD's. Some crude (simple slipping clutch), some sophisticated (electronically controlled), some clever (TorSen - Torque Sensing, a clever gearset that mechanically senses and distributes torque).

[Xusia]

With an LSD, the whole idea is to push some of the torque to the axle with less traction...

That sounds like exactly the opposite of what everyone has been saying, and what I've always understood. Doesn't an LSD send power to the wheel WITH traction (I know you said axle, but since we are talking RWD only, I assume that's not what you meant. Or was it?)? Because wouldn't an open diff follow the path of least resistance, essentially sending power to the wheel with the least traction?

[Etos]

That sounds like exactly the opposite of what everyone has been saying, and what I've always understood. Doesn't an LSD send power to the wheel WITH traction (I know you said axle, but since we are talking RWD only, I assume that's not what you meant. Or was it?)? Because wouldn't an open diff follow the path of least resistance, essentially sending power to the wheel with the least traction?

The LSD is simply a compromise for a full lock diff that won't fully lock 100% of the time. Basically under little power it acts much like an open diff, allowing the wheels a great amount of difference in rotation. Once power is applied it simply attempts to lock the power difference. Where an open diff tries to send power to the path of least resistance, an LSD attempts to overcome that. Depending on the type and settings these have huge variations right down to more sophisticated OEM methods by using the brakes to force an artificial resistance to send power the other direction.

Diffs are a tricky, complicated subject no doubt. Essentially it comes down to what kind of driving you'll be doing with the car. Strictly street driving anything but a locked diff will do. When it comes to the track, alot of personal preference, street to track driving ratio and budget take over.

I personally am looking at a torsen style with a taller gear ratio for wider powerband.

[Oppenheimer]

Yes, an open diff will send power to the least resistance. LSD attempts to overcome that while maintaining drivability. Full locker throws driveability out the window in extreme effort to get launching traction (or for overroad adventure).

An LSD is an otherwise open diff with some way to push torque to the wheel that otherwise would get all the torque. But it can do it in a way that slips to allow both wheels to still spin at different speeds to go around a corner. Like Etos said, it can be a simple slipping clutch between the two axles, when they try to spin at vastly different speeds, the clutch drag tends to pull the wheel with traction along with the one that is spinning out of control, or something more sophisticated that is electronically controlled, or even just ABS where the out of control wheel gets some braking applied.

A full locker is similar, its just that when it engages, it does so by completley locking to the two axles so they are forced to spin at the same speed. Great for a drag start, not so much for driving around a corner in the rain.

Yes, axle. Whether IRS or solid axle, the two rear wheels both have a drive axle.

[Etos]

To those who are interested in higher power but fears of breaking the 5 speed, here's a GB for a trans brace:

http://forums.nasioc.com/forums/showthread.php?t=2341424

These are nothing new in the subaru world, just rare and far between. For clarity, what breaks the 5 speeds is case flex allowing the gears to not have optimal contact and shearing the ends of the gears. This causes a snowball effect as there's less and less contact to the point of losing entire teeth and allowing the next engagement of teeth to bang into eachother thus causing massive chipping.

I personally spoke with one of subarus engineers on this topic a few years ago and he agrees that is what is happening.

[BipDBo]

To those who are interested in higher power but fears of breaking the 5 speed, here's a GB for a trans brace:

http://forums.nasioc.com/forums/showthread.php?t=2341424

These are nothing new in the subaru world, just rare and far between. For clarity, what breaks the 5 speeds is case flex allowing the gears to not have optimal contact and shearing the ends of the gears. This causes a snowball effect as there's less and less contact to the point of losing entire teeth and allowing the next engagement of teeth to bang into eachother thus causing massive chipping.

I personally spoke with one of subarus engineers on this topic a few years ago and he agrees that is what is happening.

That's very interesting. It looks like a very inexpensive solution. If the problem is case flex, I wonder, how would the case flex differently, when converted to 2WD? Would it make a difference? If so, would it make the problem worse or better? My guess would be that it would decrease the flex, considering that the front axles come out of the case at the large bell, but I'm not sure what part of the case is flexing, or which way it's flexing, so my guess doesn't mean much.

[Nuul]

To those who are interested in higher power but fears of breaking the 5 speed, here's a GB for a trans brace:http://forums.nasioc.com/forums/showthread.php?t=2341424

Thanks for the link. I don't think we're going to put the stress on the 5MT that the WRX does but it's still an area of concern for me to say the least. I'll probably end up buying one just in case when the time comes. $315 is pretty cheap insurance for a trans.

[SkiRideDrive]

To those who are interested in higher power but fears of breaking the 5 speed, here's a GB for a trans brace:

http://forums.nasioc.com/forums/showthread.php?t=2341424

These are nothing new in the subaru world, just rare and far between. For clarity, what breaks the 5 speeds is case flex allowing the gears to not have optimal contact and shearing the ends of the gears. This causes a snowball effect as there's less and less contact to the point of losing entire teeth and allowing the next engagement of teeth to bang into eachother thus causing massive chipping.

I personally spoke with one of subarus engineers on this topic a few years ago and he agrees that is what is happening.

Cool idea. Looks like it would be easy enough to make at home. It will be interesting to see what kind of power levels they can achieve on that stock gearset. It would also be nice to see the effect of putting all the power through just the front driveshafts vs. splitting the load through the awd.

[Etos]

That's very interesting. It looks like a very inexpensive solution. If the problem is case flex, I wonder, how would the case flex differently, when converted to 2WD? Would it make a difference? If so, would it make the problem worse or better? My guess would be that it would decrease the flex, considering that the front axles come out of the case at the large bell, but I'm not sure what part of the case is flexing, or which way it's flexing, so my guess doesn't mean much.

I think it depends alot on peoples setups. Engine/trans mounts are going to affect it in different ways. Too soft and you get a pogo stick affect of the trans banging back and forth. Too stiff and you force more of that power on the gearsets. I don't think the fact that it doesn't connect the rear to a driveshaft changes anything as it's the same setup from the factory.

Like a debate that was ongoing in another thread, TQ is TQ. So long as the wheels aren't slipping, the trans is taking the same abuse be it AWD or RWD. Granted a big reason for broken trans are people launching the cars from a dead stop and the AWD not allowing that power to go anywhere else but the trans from double the grip. Regardless, past that it's still the same amount of stress being put on the trans.


Thanks for the link. I don't think we're going to put the stress on the 5MT that the WRX does but it's still an area of concern for me to say the least. I'll probably end up buying one just in case when the time comes. $315 is pretty cheap insurance for a trans.

If the GB gets to 10+ people, that becomes $265. For all those people wanting to get super expensive PPG gears out of fears of a broken trans, this is a MUCH MUCH cheaper solution even at the $400 retail. Only reason anyone should get a PPG trans is if they want different gearing and willing to pay big time premium for that. I personally would rather go with this brace and a taller diff. Taller diffs IMO is a must for this car as the gear ratios are extremely short making the car too snappy especially with the turbo kicking in.

[StatGSR]

Cool idea. Looks like it would be easy enough to make at home. It will be interesting to see what kind of power levels they can achieve on that stock gearset.

the 500 hp dyno with that forester wasn't done on a stock gear set, at least that's what i read in the thread.

[Funky]

So lets say going with a 2007 WRX donor, can you think of any decent LSDs to go with it?

[StatGSR]

Cusco, Modena, Kaaz..... do some research, its the best way to learn about things.