|
jamal posted:notice how in my example the diff is at full lock? Here's a question for you jamal: If for example the front of the car were on ice, and the rear were on dry pavement, with the differential completely lock in "50:50" mode, and you gave it some gas would the front tires spin? Would the rear tires spin?
|
#
?
Feb 4, 2009 08:21
|
|
|
|
| # ? May 12, 2024 20:04 |
|
|
they would turn the same speed. except there is no resistance between the ground and the front wheels, so there is 0 torque going to the front wheels. If the engine produces, for example, 50 lb-ft of torque, and the differential is locked, and the front tires have 0 traction, that 50 lb-ft of torque must all go to the rear wheels to move the car forward.
|
|
#
?
Feb 4, 2009 08:24
|
|
|
jamal posted:they would turn the same speed. except there is no resistance between the ground and the front wheels, so there is 0 torque going to the front wheels. If the engine produces, for example, 50 lb-ft of torque, and the differential is locked, and the front tires have 0 traction, that 50 lb-ft of torque must all go to the rear wheels to move the car forward. You just proved yourself wrong. If the front wheels are turning at all, then there is torque is being applied to it. If you don't apply torque to something, its not going to magically rotate on its own. the torque going to the front of the car isn't helping to move it, but the fact that the wheels aren't helping to move the car doesn't mean no torque is being sent to it.
|
|
#
?
Feb 4, 2009 08:29
|
|
|
ok, sure. I was ignoring drivetrain loss. But only the torque required to turn the actual mechanical components in front of the differential is applied. so I guess the torque bias is actually only something like 0.5:99.5 jamal fucked around with this message at 08:36 on Feb 4, 2009 |
|
#
?
Feb 4, 2009 08:33
|
|
|
With 50/50 split, if it is truly locked, you can just pretend the diff isn't there. You've gone the transmission output directly driving two sets of wheels. The inputs to the front and rear diffs will be at the same speed and that speed will be determined by the wheels with grip. The torque applied will cause the wheels with grip to start accelerating and the gripless wheels will match. 50:50 and locked aren't quite the same thing. 50:50 implies that the torque is the same between the front and rear, while locked implied that the rotation between the front and rear are the same. You can't have 50:50 distribution on varied grip and be locked.
|
|
#
?
Feb 4, 2009 08:37
|
|
|
If it were actually 50:50 in this scenario, the front wheels would accelerate to a much higher speed than the rears. Because they are locked, however, the wheels accelerate in unison, and this rate will be determined by the wheels with grip, and as a result, almost all of your torque will be sent to the wheels with grip.
|
|
#
?
Feb 4, 2009 08:39
|
|
|
A quick picture after some late night snow fun:
|
|
#
?
Feb 4, 2009 09:28
|
|
|
atomicfire posted:You just proved yourself wrong. If the front wheels are turning at all, then there is torque is being applied to it. If you don't apply torque to something, its not going to magically rotate on its own. the torque going to the front of the car isn't helping to move it, but the fact that the wheels aren't helping to move the car doesn't mean no torque is being sent to it. Wheel rotation != useful torque. Torque on clutch-type LSDs is directly proportional to traction. You are being dumb and running your argument into the ground.
|
|
#
?
Feb 4, 2009 09:51
|
|
|
jamal posted:notice how in my example the diff is at full lock? Yes exactly it is at full lock. In a completely locked diff - the torque still goes equally to the front and to the rear. Disregard what the effective work it can do, that is not the issue when you are discussiong torque split. Torque goes to both front and the rear equally when there is no chance front of rear axle can move independant of each other. How much effective work can be done is very different. And where that work is done is different, this is true but it's not axle torque split. What you have in a locked diff is effectively a four wheel axle. In fact, lets go to the logical conclusion and lets say you have a four wheel truck axle. Two wheels a side. The outside wheel is tarmac, the inside wet ice. There is still 100% of torque availible of that axle going via both wheels - if one has traction and the other doesnt, they still have the same torque turning them. So what does this front torque actually do if there is no traction on the front but lots on the back? Well.... nothing. It's wasted. BUT because the diff is locked, there is torque availible to drive the wheels with traction. And thence the car will still move. IF the diff was open, the torque will go path of least resistance and you will get a flurry of front wheelpin, no matter what the base torque spilt is. Front and rear will be trying to do the same work. Wether that work is achieved..... different story.
|
|
#
?
Feb 4, 2009 09:56
|
|
|
Cat Terrist posted:
sorry, that is not correct. You can't have "wasted" torque. The motor can only generate as much torque as the driveline and wheels can transfer to the ground. That's physics: for every action there is an equal and opposite reaction. So say we have a 50:50 open diff. If the rear wheels have 50 lb-ft of traction, and the fronts have 20 lb-ft of traction, the motor can only produce 20 lb-ft of torque at the front wheels to move the car. Because the diff is open and biased 50:50, it means that only 20 lb-ft can go to the rear for a total of 40 lb-ft to the wheels. As soon as the motor produces 41 lb-ft, the front wheels spin. After that the motor can not produce any more torque because there is no additional load. The horsepower generated will depend on rpm. If, given the same 20 lb-ft of front traction and 50 lb-ft of rear traction, with a totally locked diff the motor would be able to produce 70 lb-ft of useful torque, and the differential torque split would be 20:50 (or 28:72). jamal fucked around with this message at 12:15 on Feb 4, 2009 |
|
#
?
Feb 4, 2009 10:04
|
|
|
Turbo Car posted:Wheel rotation != useful torque. I was going to quit arguing since none of us can really agree on an answer, until I saw this. 04-05 STi's center diff's don't use LSD's, they are open diffs. The 06-08's use clutch type LSD's but its still a moot point because we are discussing a situation where the clutch plates are locked at 50/50 with the DCCD set to LOCK. In this situation it doesn't matter if its a helical LSD, a clutch LSD, or a viscus LSD, its locked at 50/50 as if it were welded. to jamal: I finally see where you are coming from. Though I can't say I agree, I understand your argument, so i'll stop arguing with you.
|
|
#
?
Feb 4, 2009 11:22
|
|
|
Nerds. I press pedal and car goes!
|
|
#
?
Feb 4, 2009 15:51
|
|
|
jamal posted:the entire internet seems not to understand how dccd works. You think it's bad here, try going to nasioc and starting a discussion about any diff related subject.
|
|
#
?
Feb 4, 2009 16:59
|
|
|
ab0z posted:Fixed. To quote the owner of Modena Gears: "DCCD is as active as my post-marriage sexlife" atomicfire posted:04-05 STi's center diff's don't use LSD's, they are open diffs. Nein. '04 ran helical, '05 went to full electromagik clutches. Helical is also a clutch type, albeit different in operation. Turbo Car fucked around with this message at 17:25 on Feb 4, 2009 |
|
#
?
Feb 4, 2009 17:20
|
|
|
Turbo Car posted:Nein. '04 ran helical, '05 went to full electromagik clutches. Helical is also a clutch type, albeit different in operation. Nein. 04-05 were open planetary type with the torque split determined by the ring gears. The switch to a limited-slip center differential didn't occur until the 06+. All years had electromagnetic clutch packs to control lockup either from the default torque split of 35/65 in the 04-05's to 41/59 in the 06+ all the way up to a 50/50 when fully locked. http://forums.nasioc.com/forums/showthread.php?t=1074497 quote:the 2006 STi Tranny uses a LSD type center differential instead of an open differential like in the 04-05 models. http://forums.nasioc.com/forums/showpost.php?p=16219822&postcount=11 quote:-The obvious differences in outward appearance http://en.wikipedia.org/wiki/Subaru_Impreza_WRX_STI#Rev._D_.282004.29 quote:Rev. F (2006) In all fairness I think you're thinking of the front differential. The 04 STi's had a SureTrac front differential, which is just fancy for Crappy Viscous Limited Slip. The 05+ got the true limited slip front differential, I forget if it was clutch or helical type. Then in the 06+ the limited slips started to appear in the center differential, making all three diffs limited slips. atomicfire fucked around with this message at 18:14 on Feb 4, 2009 |
|
#
?
Feb 4, 2009 17:57
|
|
|
Car is due to go into the shop tomorrow. Of course, tonight something changes on the way home in relation to the noises I mentioned before. The something that changed is this, I have a gear oil leak coming from the very very back of the transmission. I can't see the exact origin, but it's dripping from the crossmember that goes directly underneath where the rear axle comes out of the tranny. So, guesses? Friction noise, constantly varies, clicking on deceleration load, and now gear oil leak from the rear of the transmission. This is going to be expensive, isn't it  ?
|
|
#
?
Feb 5, 2009 01:44
|
|
|
And after all of that it turns out that the restrictor pill WAS missing from the wastegate line, the mechanic is working to make the correctly sized one now, aiming to have the car at between 15 and 17psi, should be a decent improvement from 9! edit Literally the smallest part on the car... Who'd have thought. Chunderbuss fucked around with this message at 02:13 on Feb 5, 2009 |
|
#
?
Feb 5, 2009 01:52
|
|
|
Let's look at 50:50 torque split forgetting about locking for now. Angular acceleration is determined by input torque and moment of inertia (which in a system like a car I'm sure isn't as simple as moment of inertia, but since it is a number that represents resistance to spinning, we'll say that 100% traction is some value x, and that ice with almost no traction is some smaller value y). alpha = torque / moment of inertia. Looking at a 50:50 diff that always distributes the torque evenly, what happens when you have wheels on ice is that they will accelerate much faster, and will spin faster. In an open differential this can happen. You'll move at the rate half of the engine's output torque can accelerate the wheels with grip and the other wheels will spin like crazy. Now, looking at a 100% locked diff, not 50:50, nothing, just a solid welded differential, you know that the front and rears will be spinning and accelerating in unison. So, going back to that formula, you can back solve. The wheels are spinning at the same rate, but one set has a smaller moment of inertia, so you end up with torque = alpha * moment of inertia, and given alpha is constant between the two and moment of inertia varies, the torque distribution will vary. It will be biased to the wheels with more resistance to spinning. Now, you've got a 50:50 locked diff. Your torque is constant between the front and rear, as is your alpha. So, with either of those previous formulae, you end up with a fallacy. T / x = T / y if and only if x = y. So, in summary, I just wrote three paragraphs that would make a physicist vomit, but it demonstrates that 50:50 and locked are not the same and cannot be the same. Hopefully this puts an end to this discussion.
|
|
#
?
Feb 5, 2009 02:43
|
|
|
Oh, hey, wow. Look at this. http://forums.nasioc.com/forums/showthread.php?t=1249894&highlight=gear+oil+leak+rear Describes my problem to a T, including the leak. I think my center diff has failed. Estimate on cost to fix? Hmm, looks like people were quoted $1200 or so to fix. Not horribly expensive, annoying none the less though. bull3964 fucked around with this message at 04:37 on Feb 5, 2009 |
|
#
?
Feb 5, 2009 04:07
|
|
|
A locked diff means that equal torque is being sent to all wheels - the center diff is "locked", meaning that - by definition - it can't reallocate torque. Words are all well and good, but I'll prove myself correct with an experiment you can try at home (if you own an STI). In a safe, dry paved place and at a dead stop, roll the diff control all the way forward ("lock"). Crank the wheel as far to one side as it will go, hold it there and accelerate. The car will bog down because it's sending equal torque to all the wheels, even though the inner wheels want to move slower. But you always ask Subaru if you don't believe me: http://www.driveperformance.subaru.com/version1_2/blueprint.asp quote:The percentage of lock-up from the top “LOCK” stage to the bottom is: Looks like I was wrong about the rear wheel torque split by 15%.
|
|
#
?
Feb 5, 2009 07:41
|
|
|
Artemis J Brassnuts posted:A locked diff means that equal torque is being sent to all wheels Contrariwise, if an Audi with a locked rear diff is going through some mud on its left side, the left rear wheel will spin but give 0 torque, whereas the right rear wheel will spin and apply some torque to the dirt. Locking is what enables 0:100 ratio when one wheel loses all grip.
|
|
#
?
Feb 5, 2009 07:58
|
|
|
Artemis J Brassnuts posted:A locked diff means that equal torque is being sent to all wheels and there's what everyone keeps getting completely wrong. An OPEN diff means equal torque is getting sent to all wheels. Just because one wheel is spinning does not mean it's getting more torque. Torque is simply a twisting force that has nothing to do with rpm. One spinning wheel will be getting slightly more power, but no more torque than the wheel sitting there on pavement that can't move the car forward. jamal fucked around with this message at 09:07 on Feb 5, 2009 |
|
#
?
Feb 5, 2009 08:07
|
|
|
nm posted:Nerds. Incase you guys missed it. I just had a new downpipe put on yesterday and flashed to stage 2 with the AP..Very Nice  I'm debating on getting a used sti turbo of sorts, and getting a perrin email map...or just getting a perrin map now. I dont expect to get a turbo for a few months The DP is catless, though I do have a cat in the rest of the exhaust..it sounds pretty hot  Much louder then before. Can hear turbine spin and wastegate a bit more too.I think i've gotten to stage 2 for well under $1000, to include the AP. not bad I have to say.
|
|
#
?
Feb 5, 2009 12:48
|
|
|
Diff is being torn apart as we speak. So far my running tab is $130ish for an hour and a half's amount of diagnostic work for tearing it apart.
|
|
#
?
Feb 5, 2009 16:55
|
|
|
Those fuckers claim that I need to replace the entire transmission. They said that to even get the center diff off involves dropping the transmission and it's 8 hours book labor. I'm pretty much livid at this point. I need to find a local shop that will do the work that the car needs, not just shotgun replace. The best part is, they want to use a salvage transmission with 52k miles on it which only has a 100 day warranty. gently caress that noise, I have no idea what may have been done to that transmission before it became "salvage."
|
|
#
?
Feb 5, 2009 21:42
|
|
|
bull3964 posted:Those fuckers claim that I need to replace the entire transmission. They said that to even get the center diff off involves dropping the transmission and it's 8 hours book labor. Scour NASIOC for someones tranny that was swapped.
|
|
#
?
Feb 5, 2009 23:07
|
|
|
Lando posted:Scour NASIOC for someones tranny that was swapped. The thing is, there's no reason to replace the tranny. Even if the center diff imploded (which I'm pretty sure it hasn't since I can still drive the car) the chances of it taking out anything else inside the transmission are minimal at best. They have not provided a good case why I should replace the entire transmission rather than just removing the rear extension housing and rebuild the center diff. Even if the labor was the same for both jobs, all the parts that make up the center differential total to about a 3rd of that $2800, new. As far as I know, he didn't even drain the gear oil to see if there were chunks of metal in the general transmission case. The diagnosis seems to be a "transmission make noise, replace transmission" even though the car clearly exhibits signs of the center diff binding on full lock turns and the gear oil is leaking from the extension housing. I mean, I can make a diagnosis that if a car has a flat tire, and you replace the car, it will be fixed. That doesn't mean it's the right course of action though.
|
|
#
?
Feb 5, 2009 23:16
|
|
|
Really, what it boils down to, is time. They can charge me 8 hours labor to swap a tranny because that's book time. We all know that a skilled person can do it in 3. That difference is pure profit. Meanwhile, removing and rebuild the center diff may also be 8 book hours, but it may take them 6 hours to do in reality. My labor costs are the same (but the parts costs are way less), but they lose out on those 3 hours that they could be charging someone else for. Then, lets factor in the fact that if they swap the transmission, they have a perfectly good transmission that they can fix and sell or put in someone else's car, also for a profit. The whole situation is so transparent. Goddamn am I angry.
|
|
#
?
Feb 6, 2009 00:47
|
|
|
bull3964 posted:Really, what it boils down to, is time. Is this a subaru dealer or independent shop?
|
|
#
?
Feb 6, 2009 01:12
|
|
|
ab0z posted:Is this a subaru dealer or independent shop? Dealership. I haven't been able to find any independent shops that specialize in subarus.
|
|
#
?
Feb 6, 2009 01:48
|
|
|
So that stupid little plastic idler pulley that the AC compressor belt runs over decided that -10F mornings weren't its style and committed bearing suicide today. Funny thing is that it must have been bad for a while and the whine I had attributed to a crappy alternator is gone. The car sounds wonderful now. I guess I had gotten used to the whine.
|
|
#
?
Feb 6, 2009 02:16
|
|
|
bull3964 posted:Dealership. I haven't been able to find any independent shops that specialize in subarus. Sorry if I missed it, but what year is your car?
|
|
#
?
Feb 6, 2009 02:54
|
|
|
seriously? no Subaru specialty shops in Pittsburgh? that's kind of shocking, we have at least one in Portland as well as several Subaru tuning shops. I guess the difference is that there are a ton of Subarus in the PNW.
|
|
#
?
Feb 6, 2009 03:30
|
|
|
Dude get on nasioc into the MAIC area and ask for a reputable local shop. Theyre prices are usually a lot cheaper too.
|
|
#
?
Feb 6, 2009 03:35
|
|
|
Lando what EM are you using? AP or OS?
|
|
#
?
Feb 6, 2009 03:38
|
|
|
MMD3 posted:seriously? no Subaru specialty shops in Pittsburgh? that's kind of shocking, we have at least one in Portland as well as several Subaru tuning shops. I guess the difference is that there are a ton of Subarus in the PNW. There should be an absolute poo poo ton of Subarus in Pennsylvania as well, they're all over the North East.
|
|
#
?
Feb 6, 2009 03:51
|
|
|
jamal posted:and there's what everyone keeps getting completely wrong. And now I feel shame for participating in derailment.
|
|
#
?
Feb 6, 2009 04:12
|
|
|
ab0z posted:Sorry if I missed it, but what year is your car? '02 with 90k miles on it. Lando posted:Dude get on nasioc into the MAIC area and ask for a reputable local shop. Theyre prices are usually a lot cheaper too. I've been a member of Nasioc forever and I do know most of the people that run really modified cars in my area. They pretty much all have to do their own work. We have good parts vendors, but no one to really install them. It just kills me. http://bull3964.googlepages.com/transfer_case_detail01.jpg This is where the transfer case interfaces with the transmission. All of my problems (99.999% sure) are on the other side of that wall. How exactly are fragged components there going to hurt the rest of my transmission? There isn't enough space for any major pieces to get to the rest of the gears. If there was any more damage in the rest of the transmission, it would have been the result of the center diff seizing hard and locking up, hurting the actual gears themselves. That really hasn't happened. I have had some instances where there has been noticeable drag, but nothing violent enough to break gears.
|
|
#
?
Feb 6, 2009 04:34
|
|
|
for 8 hours of dealer labor you can buy an entire used 5-speed and have a competent shop swap it in. Call Andretech in Gaithersburg, MD. They're possibly the best subaru transmission shop in the country.
jamal fucked around with this message at 07:31 on Feb 6, 2009 |
|
#
?
Feb 6, 2009 07:27
|
|
|
|
| # ? May 12, 2024 20:04 |
|
|
jamal posted:for 8 hours of dealer labor you can buy an entire used 5-speed and have a competent shop swap it in. I don't know how you can just ignore it. It is taking all my strength not to start arguing again.
|
|
#
?
Feb 6, 2009 07:29
|
|