NOT the Torsen... [Archive]

jdstuff

02-20-2006, 04:55 PM

So, for all of the teams out there that build their own drivetrains, instead of buying pre-fab Taylor or ATV units....

1. How many of you actually take a look at how the differential affects the vehicle's handling characteristics?
2. If you are, in fact, looking at vehicle handling, how many of you still think a Torsen is the best choice?
3. If you're not running a Torsen, what types of diffs are you considering?

It is my opinion, that a torsen diff is not ideal as a REAR differential in a racing application. It seems to me that in order to be fast around a corner, a diff should: be nearly 'open' during turn-in, not adversly affect the balance of the car upon power application, and not promote understeer at WOT.

The torsen does not seem to fit any of these criterion. Being a proactive diff, it is already torque biasing upon turn in. It seems that this can sometime make the car a little unstable on turn-in. Also, torsens bias torque to the outer, heavily loaded wheel, which could be seen to contribute to power-on oversteer upon corner exit.

Being a torsen team for many years, this year we will be running a small salisbury. With a salisbury, you have the capability to adjust virtually every parameter that you can think of. Preload adjustments allow you to specify the breakaway torque needed for the diff to come 'on', thus affecting how the vehicle turns-in. Varying the friction surfaces allow you to tune how much the diff locks. Varying the leading and trailing ramp angles allow you to independently specify the RATE at which the diff locks and unlocks. With a torsen, none of this is possible!

It just seems surprising to me how many people neglect the differential as a valid contributor to a vehicle's handling.
Anyway, just curious on anyone else's thoughts & ideas....I hope to start a decent discussion!

Regards,

jdstuff

02-20-2006, 04:55 PM

Marshall Grice

02-20-2006, 05:38 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">a diff should: be nearly 'open' during turn-in, not adversly affect the balance of the car upon power application, and not promote understeer at WOT. </div></BLOCKQUOTE>

For clarification are you running dual rear disks or a single(diff mounted)?

from your criterion it sounds like you want a purely open diff. I think you'd quickly determine that excessive inside wheel spin would be a bigger problem then meeting your ideal diff requirements. I think the main reason people choose a torsen unit (other then just following the herd) is because it does act as an open diff during turn in, assuming you're not applying any input torque. Having a limited slip diff of most any type is going to improve rear end stability, not reduce it as you seem to think. Also the diff isn't biasing torque to the outside wheel, rather the opposite, which is where the understeer during accleration comes from.

I don't see how having an adjustable diff solves any of the problems you've brought up.

also, is this a student built salisbury unit or did you source it from somewhere/something?

jdstuff

02-20-2006, 06:21 PM

Marshall,

When we were running a torsen, we had a single inboard, diff-mounted rotor. So there was a torque input into the diff, and therefore some sort of biasing effect...so I will have to disagree with you in saying that the diff is open during turn in (at least in this configuration). This year, we are running dual outboad rotors.

"Having a limited slip diff of most any type is going to improve rear end stability, not reduce it as you seem to think"

Sorry if I was unclear before...but I meant to say that a LSD will improve stability, I totally agree with you there! We just weren't seeing that with the torsen, especially under trail-braking. As for the torque biasing to the inside wheel, I guess I was mistaken...for some reason, I was always under the impression that it was the opposite.

As far as adjustability goes, I think there is something to be gained in investigating the progressive locking/unlocking rates resultant of the different ramp angles. Then again, I could be wrong....that's why I started this post!

Most of the "guts" of the salisbury is from a older BMW 3-series small case diff. Some others will be student built, as will the housing.

I suppose some of the handling characteristics I've described could be fixed by simply moving to dual-outboard rotors. Then again, I've always thought the torsen to be a bit heavy, and I think that I can build a lighter unit that will still get the job dpne by moving to a salisbury.

Thanks for your input....anyone else?

Anthony V

02-20-2006, 07:45 PM

Marshall, i have to respectfully disagree with you too. i think what jason was trying to get at is that the salisbury allows you to tune the diff characteristics differently on turn entry vs. turn exit. the different ramps allow you to adjust the entry and exit characteristics independently. a 'brake' torque will utilize your 'turn entry' ramp whereas a 'driving' torque will utilize the 'turn exit' ramp. this is what allows you to isolate the 2 types of torques.

you can set it up so it acts very open on turn entry but then locks on exit. you determine how aggressively each occurs.

also, a torsen drives the outsite (loaded) tire in corner exit. the meaning of the torque biasing ratio: when the inside (unloaded) wheel starts to slip, the difference in torque causes the gears to 'bind' together. this is how/where the bias ratio multiplies the torque to the wheel with traction. so torsen diffs tend to promote throttle on oversteer. however, if you lost complete traction on one wheel it will go full open.

detroit locker style diffs drive the inside wheel on turn exit.

Underthefloor

02-20-2006, 08:56 PM

If the wheels are spinning at different speeds, a torque sensing diff will bias to the slower moving wheel when accelerating. If both wheels are spinning at the same speed, the TS diff will bias to the tire with better grip. When cornering, the inside tire can break loose and spin up until it reaches the speed of the outside tire. At that point, the torque would be transferred to the outside wheel.

There are benefits and downsides to running a LSD but they will be different for every car. Chassis and suspension geometry and setup will affect the way the tbr effects handling so a diff that works well for one car may not work well for another.

We are planning to run a non preload clutch pack differential. The clutches are energized from the gear pressure angle only and not from ramps. We can adjust the tbr by changing the order of the plates. Our goal is to find the lowest tbr without having excess wheel spin. Our current car has a huge under steer problem when accelerating out of corners so a lower tbr should help reduce this and increase lap times.

Boston

02-20-2006, 09:14 PM

Under, can you explain how torque is biased to a wheel with better grip when both wheels are spinning at the same speed? I'm picturing the torque to be equal under this condition.

absolutepressure

02-20-2006, 10:55 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Also, torsens bias torque to the outer, heavily loaded wheel, which could be seen to contribute to power-on oversteer upon corner exit. </div></BLOCKQUOTE>

Well, these guys seem to like the idea. A little oversteer is sometimes a good thing. (Not like they are smokin the tires though)

http://www.acura.com/index.aspx?initPath=RL_Learn_Featu...e_SHAWD!landi ng:true (http://www.acura.com/index.aspx?initPath=RL_Learn_FeaturesOptions_Techn ologyInnovation_Performance_SHAWD!landing:true)

click watch demo after everything pops up

I'm tryin to get my Dad to buy one of these

Ppada

02-21-2006, 12:29 AM

I'm not a drivetrain guy but take a look at this!

http://www.tsiriggakis.gr/

We will propably use the torsen.

KU_Racing

02-21-2006, 08:15 AM

Boston, the Torsen doest work the way a typical 'Salisbury' type diff does- it doesnt sense the motor torque and then lock up if a wheel spins. The Torsen uses a difference of torque applied at the contact patch of the tire to bias torque. Because of the way the gears work, the diff senses how much torque is on the least-loaded tire, then applies the same amount of torque, multiplied through the bias ratio of the diff, to the other wheel.

on another note, our team uses a honda rubicon cam and pawl diff. open diff on turn-in, then locks them together after one wheel slips. nice and light too.

Andrew Nabb

02-21-2006, 09:01 AM

Something to think about when choosing a diff, how does it operate when a wheel is unloaded?

For the torsen and sure trac style (what kettering runs) if a wheel is off the ground there will be no tractive effort applied to the other wheel. You will obviously have to keep this in mind in your suspension design and setup.

Marshall Grice

02-21-2006, 09:24 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">i think what jason was trying to get at is that the salisbury allows you to tune the diff characteristics differently on turn entry vs. turn exit. </div></BLOCKQUOTE>

well sure an adjustable diff allows you to change various parameters but does changing them "fix" anything related to Jason's criterion?

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">When we were running a torsen, we had a single inboard, diff-mounted rotor. So there was a torque input into the diff, and therefore some sort of biasing effect...so I will have to disagree with you in saying that the diff is open during turn in (at least in this configuration). </div></BLOCKQUOTE>

exactly why i asked what brake configuration you had. it makes a big difference in how the car behaves. Never the less, if you are not braking or engine braking the torsen is "open". A preloaded salisbury does not open up. Obviously the preload is adjustable and you could remove all preload to get it to act like a torsen.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> As for the torque biasing to the inside wheel, I guess I was mistaken...for some reason, I was always under the impression that it was the opposite </div></BLOCKQUOTE>
The torsen is trying to keep both tires spining at the same speed. The TBR is the limit of force that it can apply to try to equalize the speeds. there are other threads that discuss the torsen operation at length, i would refer to those.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> I suppose some of the handling characteristics I've described could be fixed by simply moving to dual-outboard rotors. Then again, I've always thought the torsen to be a bit heavy, and I think that I can build a lighter unit that will still get the job dpne by moving to a salisbury.
</div></BLOCKQUOTE>

I agree with that completely. Sounds like good justification to me. My impression from your first post was that you were going to a salisbury to address the criterion you had mentioned, which i dont' think a salisbury will do. It should act "just" like an adjustable torsen, only with components that wear out.

Underthefloor

02-21-2006, 11:53 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Boston:
Under, can you explain how torque is biased to a wheel with better grip when both wheels are spinning at the same speed? I'm picturing the torque to be equal under this condition. </div></BLOCKQUOTE>

Friction based diffs such as the torsen or salisbury act like an inefficient open diff. Since the friction is a function of normal forces in the diff, the tbr is relatively constant if there is no preload. As long as both wheel are spinning at the same speed, the friction LSDs will act just like a solid axle. It can vary torque from one tire to the other depending on the available traction. A LSD works the same way up to the point where the tbr is met and then one tire will begin spinning faster.

Preload on Salisbury diffs mainly effect the tbr at lower torque outputs. On passenger cars, preload is used to provide some torque to a wheel with traction when one tire is on ice or some other slippery surface. Preload will also effect the turn in and apex when cornering if there is low engine torque going throug the diff. Read "Development of a Limited Slip Differential" for more info on tbr and clutch pack diffs in general.

drivetrainUW-Platt

02-21-2006, 03:55 PM

atv parts are so small and pretty....

drivetrainUW-Platt

02-21-2006, 05:42 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by absolutepressure:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Also, torsens bias torque to the outer, heavily loaded wheel, which could be seen to contribute to power-on oversteer upon corner exit. </div></BLOCKQUOTE>

Well, these guys seem to like the idea. A little oversteer is sometimes a good thing. (Not like they are smokin the tires though)

http://www.acura.com/index.aspx?initPath=RL_Learn_Featu...e_SHAWD!landi ng:true (http://www.acura.com/index.aspx?initPath=RL_Learn_FeaturesOptions_Techn ologyInnovation_Performance_SHAWD!landing:true)

click watch demo after everything pops up

I'm tryin to get my Dad to buy one of these </div></BLOCKQUOTE>

What size oars would you guys recomend we put on our car give previous track layouts, also should we go with wood or some sort of plastic/polymer, hope its raining hard this year!

absolutepressure

02-21-2006, 11:15 PM

He's jealous he isn't smart like asians.

How else would you describe oversteer to someone who doesn't even know what horsepower is (yes, I know people like that, and yes, they can afford that car)

jdstuff

02-22-2006, 02:49 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">

exactly why i asked what brake configuration you had. it makes a big difference in how the car behaves. Never the less, if you are not braking or engine braking the torsen is "open". A preloaded salisbury does not open up. Obviously the preload is adjustable and you could remove all preload to get it to act like a torsen.

</div></BLOCKQUOTE>

Ture.....which is why for our application, I will be running the salisbury with little to no preload.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">

on another note, our team uses a honda rubicon cam and pawl diff. open diff on turn-in, then locks them together after one wheel slips. nice and light too.

</div></BLOCKQUOTE>

How do you guys like the preformance of the Rubicon diff? I've heard of one or two team running them in recent years. Any problems in upsetting the balance of the car when the diff engages mid-corner....or is it pretty smooth?

Jon

02-22-2006, 07:41 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Ppada:
I'm not a drivetrain guy but take a look at this!

http://www.tsiriggakis.gr/

We will propably use the torsen. </div></BLOCKQUOTE>

Isn't that just a cam and pawl???

drivetrainUW-Platt

02-22-2006, 08:16 PM

looks like one to me, nothing impressive, just bearings instand of pawls....woopie, we will see how that Rubicon beast puts out here in a month hopefully....

rjwoods77

02-22-2006, 09:29 PM

What are the rough dimensions on that rubicon diff? Also I have seen alot of different honda models thrown around. Are the all the same diff?

romkasponka

02-23-2006, 07:59 AM

Does the clutch type LSD generates more heat?

kozak

02-23-2006, 08:18 AM

NOT TO HIJACK THE THREAD BUT WHO IS RUNNING DIFF-LESS (I.E. NO DIFF) AND WHAT DO YOU GUYS THINK ABOUT THIS APPROACH.

drivetrainUW-Platt

02-24-2006, 09:53 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by kozak:
NOT TO HIJACK THE THREAD BUT WHO IS RUNNING DIFF-LESS (I.E. NO DIFF) AND WHAT DO YOU GUYS THINK ABOUT THIS APPROACH. </div></BLOCKQUOTE>

called a spool....with proper suspension geometry and driveing that lifts a wheel in a corner it can be done, simplifies the heck out of the drivetrain

drivetrainUW-Platt

02-24-2006, 10:01 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by drivetrainUW-Platt:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rob Woods:
What are the rough dimensions on that rubicon diff? Also I have seen alot of different honda models thrown around. Are the all the same diff? </div></BLOCKQUOTE>

http://i22.photobucket.com/albums/b309/duwem/diff2.jpg

http://i22.photobucket.com/albums/b309/duwem/diff1.jpg

mouse as frame of referance, thats out of a 2004 TRX-350 (same as the 450) its in a lot of the newer models. the older housings werent as beefy and had a removable ring gear, we will machine it off and replace with chain drive, the housing minus the flange for the ring gear is 3.11" dia and from the bearing flanges end to end its 4.125" ...besides a few oil pasages its sealed up as it sits (its not bolted tight together in the middle there)...death to the torsen! </div></BLOCKQUOTE>

kozak

02-24-2006, 10:45 AM

HOW HEAVY IS IT W/O THE RING GEAR. IT LOOKS LIKE A REALLY GOOD OPTION TO THE TORSEN. IS THAT A KEY WAY I SEE. IF YOU DON;T MIND ME ASKING HOW MUCH WAS IT AND WHERE DID YOU PROCURE IT, DEALER, EBAY?

rjwoods77

02-24-2006, 11:27 AM

I looked at the torsen t-1 cutaway diagram and it is 4" diameter and 4" to bearing flanges. So that means thast this honda diff is the same size?
I thought they were supposed to be smaller?

T1 torsen:

http://www.torsen.com/files/University%20Special%20012000.pdf

Scotty

02-24-2006, 02:48 PM

Very cool...What are the internals to this diff..Is it an LSD.

drivetrainUW-Platt

02-25-2006, 07:42 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Scotty:
Very cool...What are the internals to this diff..Is it an LSD. </div></BLOCKQUOTE>

cam and pawl style...I told you guys too much already...look it up like we had to

kozak

02-25-2006, 01:43 PM

alright be that way.

Andrew Nabb

02-26-2006, 06:38 PM

Just as an FYI, if you carefully order the "older" style rubicon diff you can get the two piece unit with the removable ring gear. We got one last year and have 3 of the things floating around in the shop.

Andrew Nabb

02-26-2006, 07:19 PM

Oh and the rubicon diff is a "suretrack" style diff. Look up suretrack (or suretrak) on this site and there should be some info on it. If not i can email people what we have (basic instructions)

Greg 08

02-26-2006, 09:13 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> Oh and the rubicon diff is a "suretrack" style diff. Look up suretrack (or suretrak) on this site and there should be some info on it. If not i can email people what we have (basic instructions) </div></BLOCKQUOTE> I searched suretrack and suretrak and came up empty. Am I missing something? email gjehlert@mtu.edu

Alan

02-27-2006, 04:27 AM

Try Suretrac

Scotty

02-27-2006, 09:52 AM

Thanks guys...I will do a search on SureTrac ,I appreciate the info....I was just curious to see what the internals where......

Scotty
Taylor Race Engineering

kozak

02-27-2006, 11:04 AM

Alan how much did you get your diff for. i found them for like $150.

Alan

02-27-2006, 11:14 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by kozak:
Alan how much did you get your diff for. i found them for like $150. </div></BLOCKQUOTE>

We got our first one in the summer of '01 when the 500cc version of the Rubicon just came out. Back then I paid 330 (that was with the Honda employee discount). 150 sounds like a good deal.

kozak

02-27-2006, 11:18 AM

I ACTUALLY FOUND IT ON BIKEBANDIT. IT THE SITE WHERE YOU CAN LOOK AT THE EXPLODED VIEW OF THE DIFFERENT ASSEMBLIES. DOES ANYONE HAVE ECPERIENCE WITH BIKEBANDIT.COM OR PLACES LIKE THIS

NetKev92

03-07-2006, 07:45 PM

One good link that popped up when serching.

http://www.rmahc.com/differentials.html

Boston

03-07-2006, 10:28 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by NetKev92:
One good link that popped up when serching.

http://www.rmahc.com/differentials.html </div></BLOCKQUOTE>

The internet is full of incorrect information on a variety of subjects, but the dissinformation on differentials is astounding. This website claims the Torsen bias torque to the outside wheel in a turn:

"they send more torque to the more heavily laden outside wheel"

Not to knock you personaly Kev...

Boston

03-14-2006, 09:15 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Anthony V:
also, a torsen drives the outsite (loaded) tire in corner exit. the meaning of the torque biasing ratio: when the inside (unloaded) wheel starts to slip, the difference in torque causes the gears to 'bind' together. this is how/where the bias ratio multiplies the torque to the wheel with traction. so torsen diffs tend to promote throttle on oversteer.
</div></BLOCKQUOTE>

It seems like you are saying that the torsen will bias torque to the outside wheel in a turn, causing oversteer. This is wrong; a bias of torque at the TBR towards the slower moving (inside) tire is a precondition for differentation. This means the torsen will always promote understeer on acceleration.