Losing traction when turning

[disillusions]

So my issue is a little different than the standard traction loss associated with launches, or hard cornering. When I'm making a turn during city driving, especially with the steering wheel at it's lock position (on either side, but especially left), it feels as if the wheels are slipping out from underneath the car. The loss of traction can be so bad that the entire car shakes.

The car this is happening on is a 95 Eclipse GSX with an automatic transmission. As far as any modifications that I could see effecting it would be a urethane master mount kit, and wheels from a 3G (though it was happening with it's stock wheels as well).

I know that this issue has happened to one other person, but they did not find a solution. Has anybody else on here experienced this? Or perhaps know what may be causing it? Even better would be a solution to fix it, but I know it might be asking a bit much.

Oh, and as far as power adders go, it only has hard pipes, intake, and a 1G bov; since I'm sure someone will ask.

 

[dboyle23]

Are they really slipping or does it only feel like they're slipping?

 

[rEclipserGST]

"Especially with the steering wheel in the locked position..." what do you mean? Like the wheel is turned all the way while turning? Sounds like you are trying to turn to0o0o hard possibly. Please elaborate.

 

[PieEyedPiper]

When was the last time you inspected your suspension?
Had your wheels balanced and checked their tire pressure? Don't forget that our cars understeer like crazy.

 

[leecastillo]

does your car feel like its struggleing to make power during a turn

best option is to have a friend drive the car while you stand out side to be sure if your tires are spinning or not

 

[disillusions]

Are they really slipping or does it only feel like they're slipping?

Not sure, I haven't had the chance to have another person drive it while I look at the wheels from outside the car. I can only go by what I feel/hear; but the feels is accompanied by the sound of almost skidding tires.

"Especially with the steering wheel in the locked position..." what do you mean? Like the wheel is turned all the way while turning? Sounds like you are trying to turn to0o0o hard possibly. Please elaborate.

Yes, I mean the wheel is turned all the way while turning. I would normally agree that I'm trying to turn too hard, but it is only turned all the way when doing a U-turn from a stand still. There isn't much throttle being applied at all; however, it also happens just making a simple left turn on the street.

When was the last time you inspected your suspension?
Had your wheels balanced and checked their tire pressure? Don't forget that our cars understeer like crazy.

Had them checked before I put them on the car, everything was A-OK. I know they understeer, but would it be a problem from 0-5mph just making a simple left turn? Or a U-turn?

does your car feel like its struggleing to make power during a turn

best option is to have a friend drive the car while you stand out side to be sure if your tires are spinning or not

YES! That is exactly the feeling; and once it straightens out, the thing takes off like a da*n bullet. As I told the others, I'll try and get someone to drive it as soon as possible. Do you know what could be causing it though? This has been irking me for a while now.

 

[TunaTalon]

With the steering wheel at lock the center LSD directs more torque to the slower turning front tires. While turning, the inside tire is going slower than the outside tire so the front open diff directs all of the torque to the outside tire.

All of the engine power is going to the outside front tire. With a one wheel drive it's pretty easy to make that tire spin, or hop.

 

[kmetiuk]

Have you checked your idle? I know years ago when I still had an automatic that my idle was set to high. It was always applying power to the wheels, even when I was not pressing on the accelerator. What actually happened to make me realize this was a problem was I was turing around in a cul de sac in winter and my front tires hit a slippery patch and iot pushed me right into a parked car. And when I say pushed, I mean it pushed me for about 100 feet with me frantically hammering on the brakes. Since the idle is too high it engages the torque converter and applies torque to the wheels which means pressing on the brakes is only going to lock the front wheels and prevent you from truning. Maybe this is your issue? The only other thing I can think of is that your center diff is locked or jammed somehow and is no longer open and is causing your wheels to hop around the cornere. Kind of like when you lock the diff in a big truck or jeep and try and make a turn...they hop around like crazy. Hope this helps.

Oh and that car years ago was a RWD, hence why they locked the fronts with brakes.

 

[jtmcinder]

With the steering wheel at lock the center LSD directs more torque to the slower turning front tires. While turning, the inside tire is going slower than the outside tire so the front open diff directs all of the torque to the outside tire.

Oh, my, no. First off, I doubt that his center VC is locking at all at that speed, but, if it is, a locked diff doesn't "direct" torque anywhere; it merely makes all torque available to both outputs. Second, the fronts drive a larger arc than the rear, so they turn faster, not slower. Third, an open diff allows all torque to go to the output with less grip, which, in the case of the front diff, would be the inside front, not the outside.

In short, every single thing you said was wrong.

To the OP: have you checked your front diff? Lift the entire nose and, with the car in gear and handbrake on, spin either wheel to verify that (a) it can be done and (b) the opposite wheel rotates the opposite direction. Of course, I'm assuming that you don't have something like a welded center that you forget to mention.

 

[disillusions]

With the steering wheel at lock the center LSD directs more torque to the slower turning front tires. While turning, the inside tire is going slower than the outside tire so the front open diff directs all of the torque to the outside tire.

All of the engine power is going to the outside front tire. With a one wheel drive it’s pretty easy to make that tire spin, or hop.

Even at such low speeds? o_0

Have you checked your idle? I know years ago when I still had an automatic that my idle was set to high. It was always applying power to the wheels, even when I was not pressing on the accelerator. What actually happened to make me realize this was a problem was I was turing around in a cul de sac in winter and my front tires hit a slippery patch and iot pushed me right into a parked car. And when I say pushed, I mean it pushed me for about 100 feet with me frantically hammering on the brakes. Since the idle is too high it engages the torque converter and applies torque to the wheels which means pressing on the brakes is only going to lock the front wheels and prevent you from truning. Maybe this is your issue? The only other thing I can think of is that your center diff is locked or jammed somehow and is no longer open and is causing your wheels to hop around the cornere. Kind of like when you lock the diff in a big truck or jeep and try and make a turn...they hop around like crazy. Hope this helps.

Oh and that car years ago was a RWD, hence why they locked the fronts with brakes.

Actually my idle seems to be a bit low on that car; to the point where if it is placed in drive, it won't want to move forward from the idle speed alone. In reverse it does move a bit though. Not too sure if that has any significance to my problem though.

Oh, my, no. First off, I doubt that his center VC is locking at all at that speed, but, if it is, a locked diff doesn't "direct" torque anywhere; it merely makes all torque available to both outputs. Second, the fronts drive a larger arc than the rear, so they turn faster, not slower. Third, an open diff allows all torque to go to the output with less grip, which, in the case of the front diff, would be the inside front, not the outside.

In short, every single thing you said was wrong.

To the OP: have you checked your front diff? Lift the entire nose and, with the car in gear and handbrake on, spin either wheel to verify that (a) it can be done and (b) the opposite wheel rotates the opposite direction. Of course, I'm assuming that you don't have something like a welded center that you forget to mention.

I haven't checked my front diff recently, but this car has gone through the ringer when it comes to drive train issues. To the best of my, admittedly limited, knowledge, the front diff checked out OK when it was last at my shop, and no welded center; but do solid mounts in the rear count?

You know what, let me lay out the problems it has had to the best of my ability, since I don't know if it will help.

When I first bought the car, it had a failing transmission; which was then rebuilt at AAMCO (I was young! Don't judge me! =]). Later down the line, the center diff went on the car and the warranty company I had said that they would only replace it and hold it under warranty of it was put in as one piece (still attached to a tranny they sent). So that went in. Shortly after I was getting a horrible noise from the back; and a little while later my rear diff fried. Once that happened I brought it to my current shop that specialized in DSMs/Evos. They got a replacement from the warranty company, and that ate itself up. Come to find out (much later down the line), that the gear ratio in my tranny was different from the rear diff, which caused it to fail. Both AAMCO and the warranty company gave me the run around, so they found a used rear diff with a matching gear ratio and put it in. No more smoking/shaking rear diff.

Now I don't know how legit anything I was told was, seeing as I thought there was only one A/T offered with the AWD 2G eclipses; but the issues that were there were certainly not fabricated. In any case, the car has spent more time in the shop than on the road with me, so I can't say for sure if it was doing that turning stuff before the rear diff went. I can be 10000% sure it happened after though.

I'm hoping some of the details in the story helped, otherwise I just wasted your time in reading it.

 

[TunaTalon]

Oh, my, no. First off, I doubt that his center VC is locking at all at that speed, but, if it is, a locked diff doesn't "direct" torque anywhere; it merely makes all torque available to both outputs. Second, the fronts drive a larger arc than the rear, so they turn faster, not slower. Third, an open diff allows all torque to go to the output with less grip, which, in the case of the front diff, would be the inside front, not the outside.

A DSM center LSD is not a locking differential; it is a viscous coupler that directs torque evenly front and back when front and back are turning at the same speed. When the front and rear speeds are different torque is directed preferentially to the end that is turning slower. The front and rear are never "locked" in a DSM, that's for the muscle cars of my youth.

An open diff sends more torque to the wheel that's turning faster. In the normal case of a car going straight that is indeed the tire with less grip. If the speed differential is because the car is turning the diff still sends more torque to the faster turning wheel. When in a tight turn the outside tire is going faster than the inside tire.

My two AWD auto Talons both tended to make the outside front tire hop with the OE tires but better tires cured that.

Edit:
After reading the story in Post 10 the problem probably had nothing to do with the original DSM design.

 

[jtmcinder]

A DSM center LSD is not a locking differential; it is a viscous coupler that directs torque evenly front and back when front and back are turning at the same speed. When the front and rear speeds are different torque is directed preferentially to the end that is turning slower. The front and rear are never "locked" in a DSM, that's for the muscle cars of my youth.

Please. I beg you. Just stop. You have no idea what you are talking about and you're embarrassing yourself.

As I explained in my tech article (which you really, really need to read), a VC is activated by a difference in output speeds. So, when the fronts and rears are turning (on average) at the same speed, the VC does absolutely nothing. (The rest of what you wrote about the center is incomprehensible, but I doubt that what you were trying to say was any more accurate.) As to the front and rear ever being locked ... you do realize that all LSDs short of an Evo's AYC works by locking, right? The front on a stock DSM is open, but when you put a LSD in there, it works by locking. Not all the way, but that's what they do. The rear is the same, only some stock DSMs have a LSD device back there.

An open diff sends more torque to the wheel that's turning faster. In the normal case of a car going straight that is indeed the tire with less grip. If the speed differential is because the car is turning the diff still sends more torque to the faster turning wheel. When in a tight turn the outside tire is going faster than the inside tire.

Actually, it sends more power to the output that spins faster, not torque (but there's just about no chance you'll understand that). In the case of turning with no tire slip, the rear output turns slower, so the rear gets less power. The front and rear outputs are always getting half the torque, because the center is a fricken spider-type diff.

I haven't checked my front diff recently, but this car has gone through the ringer when it comes to drive train issues.

OK, I'm going to stop you right there. If you've been having drivetrain issues and now have turning issues, then it's time to check all the diffs.

I've posted this before, so I'll just do a summary here. It's best if you devise your own tests based on an understanding of the drivetrain, but this will get you going.

Lift one wheel at a time (with the car in gear) and try to turn it. It should wiggle a bit (because there's a lot of slop), but none should turn freely. If any wheel turns freely, check the axle. Fix any broken axles before continuing.

Assuming the car passes that set of tests, lift the nose with the car in gear (plus the handbrake ) and spin a wheel. Now it should turn freely and the opposite should turn the other way.

Check the rear diff in the same way (note to Tuna: without the handbrake this time ). If your car has a rear VC, this will be a little harder than the front, but will still work.

You can only check the center if the front and rear are both OK. Lift one side of the car and turn a front. The rear should turn the other way. It can take a bit strength to do this test, as you could be fighting two VCs at the same time.

 

[TunaTalon]

Nope, not embarrassed. Just trying to explain things. Mitsubishi might be embarrassed to learn that their viscous coupling is a locking differential though.

From Mitsubishi-Motors-PR

Mitsubishi VCU (Viscous Coupling Unit) AWD System – Endeavor LTD

The full-time AWD system used on Endeavor is the result of engineering principles refined by years of competition in the World Rally Championship and the Paris to Dakar Rally. Reliable, efficient and effective this full time AWD system uses a viscous coupling type centre differential to regulate torque split between the front and rear wheels. Under normal driving conditions torque is split evenly (50:50). When the system detects slip in any of the wheels, the VCU (Viscous Coupling Unit) automatically regulates torque split in proportion to the rotational speed differentials, sending power to the wheels with the most grip. This system provides optimum traction in all driving conditions, and is especially comforting in the rain and on slippery or snow covered roads.

 

[jtmcinder]

Nope, not embarrassed. Just trying to explain things. Mitsubishi might be embarrassed to learn that their viscous coupling is a locking differential though.

From Mitsubishi-Motors-PR

You really are going to regret this some day.

How do you think a VC "reroutes" torque from one output to another? Is it done with little people inside, picking up handfuls of torque and running from one output to the other? Is it done by waving the Mitsu magic wand and poof! the torque shifts? Is it done by instantaneously altering the gearing on the center diff?

Or is done by locking the two outputs together? Not hard ... the VCs in Mitsus are not very strong ... but still by locking the outputs together?

In your defense - since I really like you - I must say that both Mitsu and Scooby have said some really silly things in their ads. Why they insist on treating us like tunafish instead of intelligent adults is beyond me. But, of course, even Quaife has played a bit loose in their ads: "automatic torque biasing?" GMFB. It locks. Just like all other non-Evo LSDs. It just locks.

 

[josh1095]

sounds like a differntial problem to me. you ever take an old 4wd truck and turn in a parking lot? it hops and shakes like hell, ### all wheels are turning at a 1:1ratio. equal turns. at slow speed this is far more pronounced than high speed because the wheel is turned much further making the difference in rotations more apparent.

 

[phunny]

I like French Fries

 

[disillusions]

sounds like a differntial problem to me. you ever take an old 4wd truck and turn in a parking lot? it hops and shakes like hell, ### all wheels are turning at a 1:1ratio. equal turns. at slow speed this is far more pronounced than high speed because the wheel is turned much further making the difference in rotations more apparent.

I've actually never been in a 4wd truck; I'm a city boy . But I'll get more information once I get my car back; it currently suffers from an exhaust leak.

I like French Fries

You know just what to say Phunny.




Now, Tuna and Cinder, as much as I love a good heated debate; throwing out conflicting information truly isn't going to help me discern the cause of my wheel-hopping woes, but rather confuse me to no end. I would respectfully ask if you two could just hold off for a bit until I get this all sorted out; then after that, we can do it like Thunderdome. Two men enter, one man leaves!

Cinder, thank you for the step-by-step guide; as soon as my car is back I will be sure to test it all out. The one thing I will say is that all of my diffs were taken apart, checked, and rechecked. I also do have one of those rare rear-LSD DSMs. I didn't think of it until just now, so forgive my lack of given information. So unless my shop is completely inept, which I doubt considering the amount of high end cars that frequent it, then I'm at a loss.

In any case, thank you to everyone who is contributing helpful information along with steps I can take to further diagnose the problem. Oh and of course, thank you once again from the comic relief, Phunny.

 

[jtmcinder]

A VC in the rear isn't all that rare. I believe all "premium package" 2Gs had it.

One other thing to consider: alignment. A massive amount of front toe-in could cause your symptoms.

 

[disillusions]

A VC in the rear isn't all that rare. I believe all "premium package" 2Gs had it.

One other thing to consider: alignment. A massive amount of front toe-in could cause your symptoms.

Well the amount of work it took to track one down when I needed to replace it made me thing it was a rare commodity; alas, I stand corrected.

I'll give that a shot first, then your steps.

 

[Only2G4Me]

I mean no offense jtmcinder, but no actual "locking" occurs in a viscous coupling. There are two plates with fins on them, and a very thick viscous fluid in the diff. the plates are attached to the output shafts. When one plate is spinning faster than the other it drags the fluid and the other plate up to speed. Effectively splitting up the power sent to the wheels. Viscous diffs do not have any parts that lock together.

 

[jtmcinder]

That's what I mean by "locking." To mean, "locking" is anything that acts to force the two outputs to turn at the same speed. That's why I say that all LSDs (below that of an Evo) act by locking.

If you wish to restrict the word "locking" to situations where cams/bolts/whatever make it impossible for the two outputs to turn at different speeds without something breaking, then you're free to do so. But that's not how I use the word.

If any of the grief above was caused by my using the word "locking" in a way that differs from yours, then I'm sorry about that, because I'd like for people to know what I mean when I say things. It would be a shame for me to not be able to explain things because I use words in an idiosyncratic manner.

Clearly, no offense taken.

edit: with all that said, when you reach the "hump phenomenon," there is direct contact in a 2G's center VC. See my tech article.

 

[kmetiuk]

When I say "locking" I don't mean it actually locks from design. There could be a particle in there causing the fins not to pass by each othere, essential locking the diff. ITf its hopping like crazy and the idles are fine, my money is on a diff issue.

 

[TunaTalon]

You really are going to regret this some day.

How do you think a VC "reroutes" torque from one output to another? Is it done with little people inside, picking up handfuls of torque and running from one output to the other? Is it done by waving the Mitsu magic wand and poof! the torque shifts? Is it done by instantaneously altering the gearing on the center diff?

Or is done by locking the two outputs together? Not hard ... the VCs in Mitsus are not very strong ... but still by locking the outputs together?

In your defense - since I really like you - I must say that both Mitsu and Scooby have said some really silly things in their ads. Why they insist on treating us like tunafish instead of intelligent adults is beyond me. But, of course, even Quaife has played a bit loose in their ads: "automatic torque biasing?" GMFB. It locks. Just like all other non-Evo LSDs. It just locks.

I agree sales pitches aren't reliable. (Except for me of course, when I was selling engineering services I was pure as the driven snow.) So let's go back to the original source of the technology untainted by management, sales or interpretation by word smiths. A torque control viscous coupling was patented by Ferguson Limited in 1977 in patent number US 4040271. From the patent abstract:

Viscous shear couplings for controlling the action of an interaxle or interwheel differential gear in a vehicle transmission are described. The couplings comprise a pack of interleaved annular plates which are free to float axially in an annular enclosure containing a viscous liquid. The enclosure incorporates a spring-loaded piston shiftable to vary the enclosure volume. The quantity of viscous liquid, in the range 85% to 100% of the minimum volume at ambient temperature, and the spring strength are chosen to give a predetermined pressure build-up within the enclosure with rise in temperature during shearing of the viscous liquid to tailor the torque-transmitting characteristics of the coupling to particular needs.

Just so we're not completely disagreeing the VC may reach the hump point and reach a near lock but heat is required for the control of torque and heat is generated by slippage.



Thank you for ending the flaming.

To the OP:
After reading your post number ten it is more likely that your problem was related to the defective "repairs" to the car, not to design defect or wear and tear.

 

[kmetiuk]

You really are going to regret this some day.

How do you think a VC "reroutes" torque from one output to another? Is it done with little people inside, picking up handfuls of torque and running from one output to the other? Is it done by waving the Mitsu magic wand and poof! the torque shifts? Is it done by instantaneously altering the gearing on the center diff?

Or is done by locking the two outputs together? Not hard ... the VCs in Mitsus are not very strong ... but still by locking the outputs together?

In your defense - since I really like you - I must say that both Mitsu and Scooby have said some really silly things in their ads. Why they insist on treating us like tunafish instead of intelligent adults is beyond me. But, of course, even Quaife has played a bit loose in their ads: "automatic torque biasing?" GMFB. It locks. Just like all other non-Evo LSDs. It just locks.

No, YOU really are going to regret this some day and GMFB....that is not true. They can automatically transfer torque, either mechanically or electronically. Do you think being properly informed is some kind of black ass magic that miraculously comes out of your ass and populates your brain? That Google somehow knows what you're about to comment on in a forum an magically waves a wand and poof! Your no longer a misinformed condescending tool?

Go look up a Torsen. They do it automatically via mech. They do it via the thrust washer inside of the diff housing, by using the gear layout that they did, the forces acting against the worm gears inside apply an axial force to the housing. This in turn applies more torque from one wheel to the other. A torsen is actually brillant. How that guy figured those worms out is beyond me. Super super smart. I'm not saying the Torsen is the greatest diff, just the design behind it is smart.

Now go look up Honda's SH-AWD. They do it automatically via electronics over mech. They have electro magnets inside that apply a varying amount of torque to each wheel. The layout is similar to a viscous diff, but instead of fins, they have little clutch plates inside. 12 plates per wheel, 6 on the wheel side and 6 on the diff side all sandwiched together. Very similar to the clutch in a sport bike. The electro magnets apply a force and squeeze the sandwich together. They can go from fully opened to fully locked. That is how they can apply 70% of the engine torque to only one wheel based off the various sensors around the car

Now go lookup a disc type differential, they are similar to a CVT. They have two discs that transfer torque and are in contact with two cones. By varying the angle at which they make contact, they vary the gear ratio, thereby transfer torque again.

These are all non-Evo LSD's and they are all visocus realted, so I dunno where you get your info....but ya.

Just so we're not completely disagreeing the VC may reach the hump point and reach a near lock but heat is required for the control of torque and heat is generated by slippage.

This is not true in regards to heat. As a fluid heats up it become less viscous. Hence why automatics have trans fluid coolers and also why when a torque converter in an auto is bagged on and heated up it slips more.

 

[PieEyedPiper]

Black ass magic?

Strong introduction, mate.