Welcome to the North American Subaru Impreza Owners Club Thursday March 28, 2024
Home Forums Images WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC Archives > NASIOC Archives > STi Forum Archive

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.







* As an Amazon Associate I earn from qualifying purchases. 
* Registered users of the site do not see these ads. 
 
 
Thread Tools Display Modes
Old 05-12-2003, 10:53 PM   #51
afpdl
Scooby Guru
 
Member#: 26361
Join Date: Oct 2002
Chapter/Region: TXIC
Location: Sugar Land (Houston), Tx
Vehicle:
My shenanigans are
cheeky and fun

Default

Quote:
Originally posted by strangerq


Good links, thanx. Will note though it does mention indeed that locking the diff effectively alters torque distribution so it's not consistent with what you are saying...also I think the 36:64 is a slightly different ratio than the USDM:

"DIFF FREE: No front and rear LSD (Front to rear power distribution=36:64)
DIFF LOCK: The front and rear have an almost direct connection. (Front to rear power distribution=front/rear load distribution)"
Can you show me where exactly in any of those threads it says that torque dist. changes from 65f 35r while tires have traction?
I know torque can change when a wheel looses traction but I did not find what you are saying in those articles.
* Registered users of the site do not see these ads.
afpdl is offline  
Sponsored Links
* Registered users of the site do not see these ads.
Old 05-12-2003, 10:59 PM   #52
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

Quote:
Originally posted by afpdl


Can you show me where exactly in any of those threads it says that torque dist. changes from 65f 35r while tires have traction?
I know torque can change when a wheel looses traction but I did not find what you are saying in those articles.
LOCK: front to rear power distrubution = load distribution (roughly 50-50).
FREE: Front to rear power distribution = 36:64.

I'm not sure what you are asking unless you are distinguishing power distribution from torque distribution, in which case you've lost me again.
strangerq is offline  
Old 05-12-2003, 11:01 PM   #53
afpdl
Scooby Guru
 
Member#: 26361
Join Date: Oct 2002
Chapter/Region: TXIC
Location: Sugar Land (Houston), Tx
Vehicle:
My shenanigans are
cheeky and fun

Default

I was asking in what thread you found ^^^this and who posted it.
afpdl is offline  
Old 05-12-2003, 11:04 PM   #54
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

Quote:
Originally posted by afpdl
I was asking in what thread you found ^^^this and who posted it.
It's from your scoobynet link showing John Felsteds post of the english DCCD manual.....it's a direct quote (cut and paste) from the manual.
strangerq is offline  
Old 05-13-2003, 12:29 AM   #55
RafalW
Scooby Specialist
 
Member#: 8193
Join Date: Jul 2001
Chapter/Region: MWSOC
Location: Illinois, USA
Vehicle:
2008 WRX STI
Blue

Default

Quote:
Originally posted by afpdl
here is the subaru manual for the dccd it mentions nothing about adjusting torque distribution. http://www.scoobynet.co.uk/bbs/threa...hreadID=143750
Quote:
DIFF FREE: No front and rear LSD (Front to rear power distribution=36:64)
DIFF LOCK: The front and rear have an almost direct connection. (Front to rear power distribution=front/rear load distribution)
Now I would like to know what is: "front/rear load distribution".
RafalW is offline  
Old 05-13-2003, 04:24 PM   #56
brainrally
Scooby Specialist
 
Member#: 3683
Join Date: Jan 2001
Chapter/Region: MWSOC
Location: Collinsville, IL 62234
Vehicle:
2007 Corvette
Black

Default

Perhaps what it's saying is that as the load distribution (or weight distribution? are these synonymous?) changes, the torque distribution changes. More *load* to the front, more power to the front. More *load* to the rear, more power to the rear.

At least that's how I interpret it. So this doesn't equate to 50/50 torque (power) distribution.

Edit: Having given it more thought, I think Load Distribution may refer to the torque load at the wheels, that is, the amount of traction.

Still trying to figure this out. Not that it matters, since I won't likely be getting an STi (not for a few years anyway).

Last edited by brainrally; 05-13-2003 at 05:33 PM.
brainrally is offline  
Old 05-13-2003, 04:31 PM   #57
johnfelstead
Scooby Specialist
 
Member#: 856
Join Date: Feb 2000
Chapter/Region: International
Location: Cambridgeshire, England, UK
Vehicle:
2123 JDM STi
Black

Default

Quote:
Originally posted by strangerq
`

Now add to that that DCCD-A supposidly responds to throttle position, g-force, etc.


John: We understand that you have a similar car, but that does not help us understand your explanation. After all you are adamently contradicting even Subaru's own explanation of how DCCD works. If you wish to dispute statements to the effect that DCCD alters the effective torque distribution, then it would seem that you must start by getting Subaru to stop saying that it does.
OK, i'll send SOA a mail.

The DCCD-A doesnt affect the way the centre diff works, all it is doing is using sensor inputs to alter the % lock of the centre diff, it's just replacing your index finger on the dial with something more inteligent. So the fact it measures throttle position etc has no relevence whatsoever to how the diff unit works.

Add to that the bumf from SOA says there are 6 positions, again that is wrong, the DCCD-A is infinately variable from 0-100% locked, there are 6 LED's on the dash to give you an indication of roughly where the settings are, but that doesnt mean it has 6 distinct positions, the dial you adjust is a potentiometer, not a 6 position switch. Having spent some time with the first comercially available active centre diff controller, way before the DCCD-A came out, i know how the DCCD electromagnet control ECU works too.

What manufacturers release to the public isnt always technically correct, they are just trying to explain things in a manner that is easily understandable.
johnfelstead is offline  
Old 05-13-2003, 06:21 PM   #58
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

Quote:
Originally posted by johnfelstead


OK, i'll send SOA a mail.

The DCCD-A doesnt affect the way the centre diff works, all it is doing is using sensor inputs to alter the % lock of the centre diff, it's just replacing your index finger on the dial with something more inteligent. So the fact it measures throttle position etc has no relevence whatsoever to how the diff unit works.

Add to that the bumf from SOA says there are 6 positions, again that is wrong, the DCCD-A is infinately variable from 0-100% locked, there are 6 LED's on the dash to give you an indication of roughly where the settings are, but that doesnt mean it has 6 distinct positions, the dial you adjust is a potentiometer, not a 6 position switch. Having spent some time with the first comercially available active centre diff controller, way before the DCCD-A came out, i know how the DCCD electromagnet control ECU works too.

What manufacturers release to the public isnt always technically correct, they are just trying to explain things in a manner that is easily understandable.
Excellent. I really do think you should write Subaru. I'm not kidding. Because your explanation of DCCD functionality is not merely subtley different from Subaru's....it completely contradicts them. Virtually every sentence of their explanation of their own technology is therefore completely wrong and possibly intentionally misleading.

Just a taste and I quote.......
"Taking A Curve With DCCD
As the driver releases the throttle and applies the brakes approaching a turn, the center differential locking force is minimal. As the driver turns the car into the apex of the curve, the lateral g force increases. As a result, DCCD begins to lock the differential, transferring power to the front wheels. As the car exits the corner and the driver applies the throttle, DCCD continues to progressively lock the differential to help prevent oversteer and maintain stability."

Now according to you...the act of locking the dccd does not transfer power (ie - redistribute torque), and therefore locking the DCCD based on G force is irrelevant to to the issue of torque distribution.

I still don't know how DCCD works, but either you or Subaru are completely wrong about its functionality.
strangerq is offline  
Old 05-13-2003, 06:29 PM   #59
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

Quote:
Originally posted by brainrally
Perhaps what it's saying is that as the load distribution (or weight distribution? are these synonymous?)
My interpretation would be that load distribution equal weight distribution taking acceleration/deceleration into consideration.

ie - the car is front heavy, approx 59/41 but under acceleration it is less front heavy, under braking it is more so.

Don't lose sight of the issue though......we are discussing whether the effective torque distribution changes or remains fixed......., the manual's explanation clearly implies the torque distribution is not the same when diff locked as it is when diff free.....the whole paragraph is dedicated to that premise.
strangerq is offline  
Old 05-14-2003, 05:42 PM   #60
rhung
Scooby Newbie
 
Member#: 3073
Join Date: Nov 2000
Location: Winnipeg, Manitoba, Canada
Default

Man... I have been banging away at this for a while trying to get my head around it but... I think I have something. After doing a bunch of reading on SAE articles, TBR definitions, differentiating LSC and LSD (man the Audi Haldex people are uptight), it took one look at the trusty AutoZine article on differentials to make it click. Basically, I forgot that Subaru uses a Viscous Coupling Differential Lock system for their AWD centre differential. This is basically an open diff with a viscous coupler at one of the output driveshafts.

Now, having never seen a DCCD equipped STi, I'm taking a stab here but I'm guessing that it essentially replaces the Viscous Coupler with a Electromagnetic controlled wet multiplate clutch system. With the DCCD-A, the ECU gathers data from the various sensors and outputs a current (of various amperage) to stimulate a magnetic field thus causing the clutch plates to be "magnetized" and thus lock. On manual mode, a current at a constant level (but manually adjustable using the dial) is outputted causing various levels/percentage of lock.

If this is what it is, John's description makes sense. The torque split, as he refers to it, will be 35/65 since the centre diff will be geared to as such. No amount of adjustment on the DCCD will change the mechanical torque split of the centre diff. The DCCD just adjusts the amount of lock of the multiplate clutch system.

So, if the diff is open, then we essentially have an open centre diff that has a torque split of 35/65 with no anti-slip properties. Get both front or rear wheels on glare ice at a standstill, and you're likely not going anywhere. With any other DCCD setting, the torque split is still 35/65 as the clutch system tries to prevent any slip between the two output shafts.

Does that make any sense?

Ray
rhung is offline  
Old 05-14-2003, 06:04 PM   #61
RafalW
Scooby Specialist
 
Member#: 8193
Join Date: Jul 2001
Chapter/Region: MWSOC
Location: Illinois, USA
Vehicle:
2008 WRX STI
Blue

Default

Quote:
Originally posted by rhung
Does that make any sense?
No.
Why everybody forgets about treating locked diff as a solid connection? No slip means exactly this. And at this point there is no way to delive different torque to each end of the car so it's 50:50. And this is exactly what Subaru says but some people here think they know better.
RafalW is offline  
Old 05-14-2003, 06:30 PM   #62
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

Quote:
Originally posted by RafalW

No.
Why everybody forgets about treating locked diff as a solid connection? No slip means exactly this. And at this point there is no way to delive different torque to each end of the car so it's 50:50.
Well, let's operate on that assumption and move on to the next question.

What is happening at the various points between full-lock and full-slip?

Subaru implies that under cornering situation in DCCD-A the diff moves toward lock.

Does this imply that the diff treats the conering scenero as a slip situation because of the differences in wheel speed between front and rear wheels.....and so redistubutes torque?

Does it actually require wheel "spin" per se in the rear wheels or
is the natural difference in wheel speed under cornering suffice to trigger?

What happens when the front wheels are the ones spinning under DCCD-A? Does this cause the diff to close or open?

If the diffs functionality can be expressed entirely in terms of wheel slip (odd that Subaru often avoids the term wheel slip in it's DCCD explanations but uses it liberally in it's VCD explanations), then can we express it in terms of a certain pertentage of diff lock allowing a certain amount of wheel slip...... or not?

This is why I have a problem with circular explanations that state that DCCD simply progressively locks or unlocks the differential.....
and therefore....what? Different amounts of slip between front and rear are allowed? Torque is redistributed to a different degree? It still has not been explained.
strangerq is offline  
Old 05-14-2003, 06:47 PM   #63
RafalW
Scooby Specialist
 
Member#: 8193
Join Date: Jul 2001
Chapter/Region: MWSOC
Location: Illinois, USA
Vehicle:
2008 WRX STI
Blue

Default

Quote:
Originally posted by strangerq
Subaru implies that under cornering situation in DCCD-A the diff moves toward lock.

Does this imply that the diff treats the conering scenero as a slip situation because of the differences in wheel speed between front and rear wheels.....and so redistubutes torque?

Does it actually require wheel "spin" per se in the rear wheels or
is the natural difference in wheel speed under cornering suffice to trigger?

What happens when the front wheels are the ones spinning under DCCD-A? Does this cause the diff to close or open?

If the diffs functionality can be expressed entirely in terms of wheel slip (odd that Subaru often avoids the term wheel slip in it's DCCD explanations but uses it liberally in it's VCD explanations), then can we express it in terms of a certain pertentage of diff lock allowing a certain amount of wheel slip...... or not?

This is why I have a problem with circular explanations that state that DCCD simply progressively locks or unlocks the differential.....
and therefore....what? Different amounts of slip between front and rear are allowed? Torque is redistributed to a different degree? It still has not been explained.
The most important is to understand DCCD is proactive not reactive. It means the computer collects all the data from sensonrs and makes decision what to do to create desired (preprogrammed) solution. So in turn it can just read Gs and steering wheel angle to know it's in turn and act accordingly. Slip is detected as well but it's not necessary for changes to happen.
Viscous are reacting to a different speed on both ends and this creates a lock but DCCD might even leave it opened or lock more then just reaction to this difference.
RafalW is offline  
Old 05-15-2003, 11:59 AM   #64
Russ in CT
Scooby Newbie
 
Member#: 36869
Join Date: May 2003
Default SOA description notes a PLANETARY gearset

Did you note that the SOA description of DCCD states the center diff uses a PLANETARY gearset? Yes, the same sort of gear system used in automatic transmssions, which allows for the output gearing to be altered simply by applying clutch pressure to various points within the planetary gearset.

An automatic transmission can switch output gearing without meshing and unmeshing any gears, and so could the DCCD. But is this how it actually works? I don't know, haven't done enough research yet.

But it would appear that the DCCD could indeed vary torque split by altering the effective gearing, assuming the clutch is acting on the planetary gearset the same way that the clutches in an auto tranny act on its planetary gearset.

So is this how the (US version STi) DCCD works? Like a 6 speed automatic tranny, with a thumb wheel 'Tiptronic' mode, and a fully auto mode? switching in 6 possible steps between a gear ratio of 65:35 and 50:50?

Or does it (the US STi DCCD) have fixed 65:35 gearing, and all that the clutch does is vary the amount of 'lock-up', or slip, front to rear (as has been posted?)

If all the clutch does is vary the amount of slip, why would the DCCD need a planetary gearset?

Obviously one doesn't have to have all these answers to drive it, even to drive it well. But that doesn't stop us from wanting, needing, to know. I for one am still wanting for a more thorough understanding.
Russ in CT is offline  
Old 05-15-2003, 02:04 PM   #65
reks
Scooby Specialist
 
Member#: 9567
Join Date: Aug 2001
Location: Marlboro, MA
Vehicle:
2005 Legacy GT Wagon
Silver

Default

I thought torque distribution and slip were two different things.

Follow my example:

Let's say I put 100lbft of torque through a differential that is 35:65 split. Let's say both sides have traction and can transfer the full amount of torque. On one side I will measure 35lbft of torque and on the other I will measure 65lbft of torque. This is what the 35:65 split means.

Now let's say I take that same differential, leave the differential completely open and put the 35% side on a slippery surface. The amount of torque being applied to the slippery surface is 0lbft. It's slipping: no torque is applied. Now we need to calculate how much torque is being applied to the other end. Since this is an open differential, the same amount of torque is applied to the other end (by percentage). This is effectively 65% x 0 or 0 torque.

Now if we lock the differential fully, 65% of the full 100lbft of input torque or 65 will be transferred instead of the fully open case (which is 0).

I would propose that incrementally moving through the DCCD locks would provide for increasing/decreasing amounts of slippage. The effect you would see is less power being provided to the rear if the front is slipping and vice versa. The amount of fall off would vary in proportion to the openness of the DCCD setting.

For example, let's say DCCD is set to 50% slippage. I would expect to see 50% x 100 x 65% or about 32.5lbft of torque on the 65% side if the 35% is fully slipping.

In this way controlling the slippage of the DCCD will control the amount of effective torque being put down by the wheels which are NOT slipping. You have no way of varying the torque of wheels which are slipping. They are slipping after all and the amount of torque they put down is dependent on the tire and the surface. Nor can you vary the base percentage of the input torque that either side will receive. However, you are varying the *effective* amount of torque that makes it to the ground.

So, if you wanted to get really layman you could say that the DCCD differential is a 35:65 torque split with the ability to put more torque forward or aft as needed (read: when the wheels are slipping).

Now the auto setting of DCCD checks various sensors and adjusts the amount of lock for you. That's pretty tricky and you would need to know exactly what sensors they are looking at and exactly what they are trying to do to figure out how it works. But it must work pretty well because most reviewers seems to think that Auto is a good setting if you're not 100% sure what you're doing.

Note that this is much different from most clutching systems. These systems usually employ a disengaged clutch giving 100% of the power to the front wheels (FWD). When there is slippage the clutch can fully engage giving a 50:50 locked torque distribution between front and rear. The advantage of this system is fuel economy under 'normal conditions' at the expense of response time to slippage.

Incidentally, most 4x4 setups use fully locking differentials. Since there is no slip in these systems, they recommend against using them on dry pavement. I would imagine that subaru would recommend against using full DCCD lock if no slip is antcipated. Then again, 'full' DCCD lock might allow for some slippage (not 100% locked) which would make this less stressful.

Anyway, there's my conjecture and explanation given the stuff that I've seen and my rather limited understanding of the mechanics involved.

Last edited by reks; 05-15-2003 at 02:09 PM.
reks is offline  
Old 05-15-2003, 03:21 PM   #66
rhung
Scooby Newbie
 
Member#: 3073
Join Date: Nov 2000
Location: Winnipeg, Manitoba, Canada
Default

Quote:
Originally posted by RafalW

No.
Why everybody forgets about treating locked diff as a solid connection? No slip means exactly this. And at this point there is no way to delive different torque to each end of the car so it's 50:50. And this is exactly what Subaru says but some people here think they know better.
Again, this is only my theory and is open for discussion. I think what we have to remember here is that the centre differential of Subaru's actually comprises of two devices. Using the WRX, it has a centre differential (open) and at one of the output driveshafts, a viscous coupler. Look at the Viscous Coupling Differential Lock section in this url.

Autozine's article on diffs

See the open centre diff and the separate viscous coupler attached to only one output driveshaft? The centre diff provides the torque split via the side gears or, the gears that are connected to the driveshafts within the differential. The VC is a 1 input, 1 output device that provides the anti-slip function. The VC does not alter torque split. It can't as it has no gears to apply a different torque ratio. At the same time, the centre diff cannot change torque splits because there's only one set of gears.

This applies to the STi's centre differential as I believe (based on John's comments about removing the electromaget and the centre diff is still there) that the STi's DCCD comprises an centre open diff like in the diagram, but replaces the VC with a electromagnetic wet multiplate clutch system. The electromagnetic setup still has no gears to alter torque split as this is derived from the centre diff's gears.

Quote:
Did you note that the SOA description of DCCD states the center diff uses a PLANETARY gearset? Yes, the same sort of gear system used in automatic transmssions, which allows for the output gearing to be altered simply by applying clutch pressure to various points within the planetary gearset.


This is the way I see it. STi's DCCD centre diff uses a planetary gearset to achieve a 35:65 rear bias torque split. However, I still think its an open centre diff that has a planetary gearset at one end of the side gears. It would use the standard side gear as the input, the planet gears as the stationary medium and the ring gear for output. However, that's all it will do. There is no changing of gears at the planetary gearset. To do so smoothly would require building the smoothest, non shuddering, in an extremerly small package mini auto transmission within a differential that is the size of a hand. The planetary gearset just provides a gearing increase for one end thus giving it a higher torque bias.

Ray
rhung is offline  
Old 05-15-2003, 03:26 PM   #67
rossscarbro
Scooby Specialist
 
Member#: 34627
Join Date: Mar 2003
Location: Riverside county, Socal!!
Vehicle:
2006 Evo IX GSR
Graphite Grey

Default

Nice try guys.... I'm totally lost. Anyone else, or am I such a noobie that I'm just too dense?
rossscarbro is offline  
Old 05-15-2003, 04:10 PM   #68
RafalW
Scooby Specialist
 
Member#: 8193
Join Date: Jul 2001
Chapter/Region: MWSOC
Location: Illinois, USA
Vehicle:
2008 WRX STI
Blue

Default

Quote:
Originally posted by rhung
Using the WRX, it has a centre differential (open) and at one of the output driveshafts, a viscous coupler.

WRX has voscous in center and rear. Only front is open.
Quote:
The VC does not alter torque split. It can't as it has no gears to apply a different torque ratio. At the same time, the centre diff cannot change torque splits because there's only one set of gears.

Did you ever drive a MT car? Do you know what is a clutch? Can it vary torque without any gears?
Quote:
This is the way I see it. STi's DCCD centre diff uses a planetary gearset to achieve a 35:65 rear bias torque split. However, I still think its an open centre diff that has a planetary gearset at one end of the side gears. It would use the standard side gear as the input, the planet gears as the stationary medium and the ring gear for output. However, that's all it will do. There is no changing of gears at the planetary gearset. To do so smoothly would require building the smoothest, non shuddering, in an extremerly small package mini auto transmission within a differential that is the size of a hand. The planetary gearset just provides a gearing increase for one end thus giving it a higher torque bias.
OMG!
Why do you need AT in a center diff?
Answer my question I already asked:
How it's possible to have a different torque delivered to front then rear when diff is fully locked?
RafalW is offline  
Old 05-15-2003, 06:22 PM   #69
Hummer
Scooby Specialist
 
Member#: 31892
Join Date: Jan 2003
Location: KC
Vehicle:
2004 STi
Silver

Default

Quote:
Originally posted by rossscarbro
Nice try guys.... I'm totally lost. Anyone else, or am I such a noobie that I'm just too dense?
I am with you bro. I am used to a one wheel drive GTP. This thread makes my brain ache. It is a very civilized discussion though....that might be my problem.
Hummer is offline  
Old 05-16-2003, 03:54 AM   #70
boostaddict
Scooby Newbie
 
Member#: 35871
Join Date: Apr 2003
Chapter/Region: SCIC
Location: Yorba Linda, CA
Vehicle:
2003 WRX Wagon MBP
Spun Bearing @ 176k

Default

Pertaining to differentials in general, I don't see how a locked diff can have a 35:65 torque distribution. A locked diff has a 50:50 distribution. If "gearing" were to give it anything but 50:50 you'd have wheels going different speeds, and a lot of tension on the diff while cruising. At 65mph, a 35:65 split in a "locked" diff would give you 34.45 mph on one end of the car, and 65 mph on the other. I don't find that believable.

The basic function of a differential is to allow the shafts at opposite ends to turn at different speeds. The limited slip, variable, torsen, locker, and other features only enhance the basic function. This basic function is achieved through slippage (of some sort) inside the diff. Controlling this slippage, ie. DCCD, controls the amouont of torque/power applied the each shaft. With some slippage occuring, both shafts can turn the same speed, but the torque applied to each can be different. Without slippage, the shafts must turn the same speed, hence a locked situation would yield a 50:50 distribution.

I don't think I need to go into specifics regaridng the STi or anything else similarly equipped, but if there's questions or contradictions, of course post them.
boostaddict is offline  
Old 05-16-2003, 08:54 AM   #71
strangerq
Scooby Specialist
 
Member#: 13095
Join Date: Dec 2001
Default

I think the argument that locked diff = approximately 50/50 torque, and is distinct from open diff's 35/65 is fairly clear and persuasive. It is consistent with the manuals on DCCD and no one has presented any argument directly to the contrary. (ie - explaining how diff can be locked and still transmit variable torque front to rear; notwithstanding issues of traction loss, etc..)

Now, let's deal with settings other than open and locked.

If the dccd clutch is in an intermediate postion allowing some slip, does that imply that torque distribution is generally somewhere between 50/50 and 35/65?

Is the analogy that has been offered of an manual tranny clutch, which of course does vary torque distribution between the engine and the transmission by varying slip, appropriate, or is it flawed?

If the tranny clutch analogy is appropriate then i'm satisfied that we've answered the most pressing question about DCCD's funtionality. If it isn't, uh.....then we're still lost.
strangerq is offline  
Old 05-16-2003, 01:00 PM   #72
boostaddict
Scooby Newbie
 
Member#: 35871
Join Date: Apr 2003
Chapter/Region: SCIC
Location: Yorba Linda, CA
Vehicle:
2003 WRX Wagon MBP
Spun Bearing @ 176k

Default

Quote:
(ie - explaining how diff can be locked and still transmit variable torque front to rear; notwithstanding issues of traction loss, etc..)
I'm not sure if were implying this, but a "locked" diff does not deal with loss of traction. Both ends will receive 50% of the torque even if one end has 100% slippage. The torque from the slipping side will not be transfered to the traction side.


Quote:
If the dccd clutch is in an intermediate postion allowing some slip, does that imply that torque distribution is generally somewhere between 50/50 and 35/65?
I would assume that, yes, but it would depend on the design.

A fully open diff (ie. FWD) still has some natural resistance to slippage. The specific amount, or 35:65 possibly, would be determined by the design of the diff. If the diff were designed to have a different ratio, say 0:100 or 15:85, then the DCCD would have to have the electromagnet energized all the time to achieve the default 35:65 ratio, but that's entirely possible.

I think it'd be a much simpler design to have the 35:65 built into the diff and use the elctromagnet to adjust it tighter from there.


Quote:
Is the analogy that has been offered of an manual tranny clutch, which of course does vary torque distribution between the engine and the transmission by varying slip, appropriate, or is it flawed?
No, that analogy is close, but not 100% accurate. In the manual tranny case the clutch is varying the amount of torque applied from the source. A diff must distribute 100% of the torque applied to the input shaft (from the source) to either of the two output shafts, there's nowhere else for it go. Of the various types, torsen, clutch, helical, etc., the STi DCCD must be a clutch type since I think that's the only type that can be electronically variable.

Last edited by boostaddict; 05-16-2003 at 01:06 PM.
boostaddict is offline  
Old 05-16-2003, 02:49 PM   #73
rhung
Scooby Newbie
 
Member#: 3073
Join Date: Nov 2000
Location: Winnipeg, Manitoba, Canada
Default

Quote:
Originally posted by RafalW
[b]
WRX has voscous in center and rear. Only front is open.
Ok. Let's look at it from your end. The WRX AWD system only uses a viscous coupler. Now, a viscous coupler has a single input and a single output unlike a differential that has a single input and two outputs. So if the VC only has a single input and a single output, how is AWD achieved? With only a single in and single out, the engine can't drive its torque to both ends of the car at the same time.

Well, there is a way but it results in a part time AWD system like that of VW's old Syncho system. Let's take this a bit further and describe it. The syncro system works like this:

The front diff gets power all the time. The driveshaft to the VC is driven in step with the front diff. Then, the power goes into the viscous coupling, where the viscous fluid is stable unless there is a great difference in speed of the input and output. Since its stable, it means no torque is transferred to the rear end. In the case of front diff wheelspin, the input and output shaft speeds of the VC are very different. This causes the viscous fluid to quickly heats up in the VC which stiffens up the coupling between the input and the output. Now, torque is being transferred.

So, in this situation, there is no way for full time AWD system and certainly not a normally operating 50:50 split of the WRX. So how does Subaru do it? It adds a open differential to work alongside the VC. So, the WRX has a viscous coupler alongside an open differential for its AWD. Those two devices work in conjuntion to create the LSD. Incidentally, the open front diff of the WRX is located in the transmission. The centre open diff and VC is located in the transfer case. Two separate devices.

So how does it all work? The engine spins the ring gear of the open differential and the two side gears of the diff are part of the front and rear driveshaft. One input, two outputs. The centre diff allows for rotational differences but it always outputs equal amounts of torque since the side gears are geared as such. The VC then receives as an input, the front or rear driveshaft to create the anti-slip function. I think, the VC is connected to the front driveshaft. If the front wheels begin to slip, the VC will try to slow the wheels down masking the fact that the front wheels are slipping to the open diff. Since the open diff doesn't sense wheel slippage, it happily continues distributing torque to the rear. But what happens if both the rear wheels slip (on ice)? Well, in this configuration, the rear end just spins like mad since there is no VC unit to mask the rear slippage to the open diff.

Quote:
Did you ever drive a MT car? Do you know what is a clutch? Can it vary torque without any gears?
Yes to all questions and I can see where my statement is off. But let's be clear about this... a clutch is a transfer medium. It's not a torque multiplier like gears.

Quote:
OMG!
Why do you need AT in a center diff?
Answer my question I already asked:
How it's possible to have a different torque delivered to front then rear when diff is fully locked?
The AT comment was just a remark on a prior post that said that because of the planetary gearset the DCCD uses, one could vary the gear ratios and hence, the torque split of the centre diff. Now while it is possible I guess, the use of all the bands and clutches makes it sound like an AT. Now fitting a mini AT is just an extreme comment but I still think the packaging, smootheness requirements, and the fact that using planetary gearsets in this manner results in very definitive gear ratios (no variation in between) doesn't really match up to what Subaru's DCCD does. At least, not to my mind.

But onto your last question, if the diff is fully locked, how is it possible to deliver different torque splits front and rear? Well, if you believe my explanation that the AWD of Subaru's WRX comprises both a VC and an open centre differential, its quite easy. Since the VC is external to the open centre diff, we can see that regardless of whether the VC fluid is shearing or stable, the ultimate torque split is determined by the gearing of the open diff. Let's take an example. Assume the standard WRX with its 50:50 distribution. On normal traction filled conditions driving in a straight line, the VC fluid is stable and the open diff distributes 50:50 torque split. In a turning situation, the rear end wants to travel slower since it is travelling a shorter distance than the fronts. The centre diff compensates for the difference in rotation speeds by letting the pinion gears turn but it still applies torque 50:50. Now right here, you might say this doesn't make sense. If the speeds are different, then torque must be delivered somewhere else. First, remember what torque is: torque is rotational force. You can generate the same amount of torque at different rotational speeds. Look at car engines. With a two different car engines, you can have engine A develop 150 lb-ft of torque at 5000 RPM while engine B develops 150 lb-ft of torque at 3500 RPM. Different speeds but same values of torque. So ultimately, how is torque defined within a differential? By the gearing of the differential. The pinion gears within the diff allow for different rotational speeds but it still applies a 50:50 torque split. As for the VC, well it could be stable or shearing but it doesn't change the gearing of the diff.

So now, change the gearing within the diff to accept a 35:65 split and my statement still holds. It's the gearing of the diff that determines the split. Not the VC or electromagnet.

Ray (who after all the research, lights are coming on)
rhung is offline  
Old 05-16-2003, 05:08 PM   #74
RafalW
Scooby Specialist
 
Member#: 8193
Join Date: Jul 2001
Chapter/Region: MWSOC
Location: Illinois, USA
Vehicle:
2008 WRX STI
Blue

Default

Quote:
Originally posted by rhung
So how does it all work? The engine spins the ring gear of the open differential and the two side gears of the diff are part of the front and rear driveshaft. One input, two outputs. The centre diff allows for rotational differences but it always outputs equal amounts of torque since the side gears are geared as such. The VC then receives as an input, the front or rear driveshaft to create the anti-slip function. I think, the VC is connected to the front driveshaft. If the front wheels begin to slip, the VC will try to slow the wheels down masking the fact that the front wheels are slipping to the open diff. Since the open diff doesn't sense wheel slippage, it happily continues distributing torque to the rear. But what happens if both the rear wheels slip (on ice)? Well, in this configuration, the rear end just spins like mad since there is no VC unit to mask the rear slippage to the open diff.
So you know it for a fact or you're just guessing that WRX with rears on ice would go nowhere? Why VC needs to mask anything? When the difference is speed heats up it progressively locks both end so it transfers effectively more torque to non slipping wheels (because these are rotating slower). So VC is going from open to closed in a variable manner.
Quote:
Yes to all questions and I can see where my statement is off. But let's be clear about this... a clutch is a transfer medium. It's not a torque multiplier like gears.
But it can progressively transfer the torque from one to the other end. Just like VC in WRX. So even if you apply 10 to fronts and 90 to rears VC in slippage conditions would transfer part of the torque from one to the other end up to 50:50 situation at a full lock.
Quote:
But onto your last question, if the diff is fully locked, how is it possible to deliver different torque splits front and rear? Well, if you believe my explanation that the AWD of Subaru's WRX comprises both a VC and an open centre differential, its quite easy.
Not really. Locked diff makes front and rear solid. So whatever you apply to each end it will add up to a full torque applied to both of them making it in fact 50:50. This is a reason this thread started with a warning about strange sounds at full lock. You can still deliver 35 to front and 65 to rear but it will still in effect give you 50:50 with locked a diff.
RafalW is offline  
Old 05-16-2003, 06:27 PM   #75
johnfelstead
Scooby Specialist
 
Member#: 856
Join Date: Feb 2000
Chapter/Region: International
Location: Cambridgeshire, England, UK
Vehicle:
2123 JDM STi
Black

Talking

with no power to the electromagnet the diff is open, it is in that state when the diff is set to open or if the power to the centre diff ECU fails.

A little picie to keep you going, this is me with the car set to an open centre diff, the rears are spinning hence the tyre smoke, yet still providing a large forward/sideways force, the fronts are freewheeling.

If it were raining i could show you how having the diff fully locked i could still get the car in that angle, but the front wheels would be pointing straighter with all 4 wheels spinning, still with more power to the rears than the front in a 65:35 split.



johnfelstead is offline  
 

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Come on i-clubbers, what's one thing that you are ashamed of actually liking. Eric SS Off-Topic 72 01-02-2009 07:42 PM
The ONLY thing that will be on the news for the next 4 weeks tacomaprime Off-Topic 11 11-14-2007 12:56 PM
New, 2005 2.5RS or used, 02-03 WRX for a first-time Subie driver? live- Newbies & FAQs 19 02-10-2005 02:29 AM
haha that new Neon...first time ive seen one yet yzfrone Mid Atlantic Impreza Club -- MAIC 33 12-19-2003 07:24 PM
One thing that is wrong with the world (music related) JGard Off-Topic 18 03-05-2003 07:19 PM

All times are GMT -4. The time now is 08:25 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2024 Axivo Inc.
Copyright ©1999 - 2019, North American Subaru Impreza Owners Club, Inc.

As an Amazon Associate I earn from qualifying purchases.

When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission
Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.