D or 3 ?

[svxistentialist]

Originally Posted by Trevor

I do not have US manuals. From what official source do the US figures originate?

Exactly how is there a torque split during normal driving when there is no wheel slip, and as I understand it, only a clutch involved in respect of torque distribution?

As a matter of interest the JDM split is actually, 36.4% F and 63.6% R and this is as a result of a gear ratio inherent within the centre diff.

I'm glad you cleared that up, Trevor. The seat of my pants was telling me that a 35/65 split wasn't exactly right.

I don't know the answer to your first question, but the US type gearbox was in use in Legacys and other vehicles, and I think the torque split has been explained in Subaru technical journals such as the End Wrench.

At present I don't have a scan of them on my computer, but perhaps somebody in the US could chime in and quote the source?? John or Earl maybe?

Joe

[ridered777]

Originally Posted by Trevor

I do not have US manuals. From what official source do the US figures originate?

Exactly how is there a torque split during normal driving when there is no wheel slip, and as I understand it, only a clutch involved in respect of torque distribution?

As a matter of interest the JDM split is actually, 36.4% F and 63.6% R and this is as a result of a gear ratio inherent within the centre diff.

Trevor,

I will give you what I have heard, but take it as neither fact, nor truth.

I THOUGHT the US SVX's (like all the US Subaru's) got a viscous type CENTER differential. They are commonly used as center differentials, but not usually as side to side diffs.

This is due to the inherent characteristics, which also line up with what I've experianced when driving under low-traction situations.

I am unsure if you are farmilar with viscous type differentials, I'd guess you are, but I take nothing for granted. So a quick take on them (again, I am mearly a 19 year old with a thing for cars, and have no authority on the subject- take nothing as fact or truth):

There a several plates inside of the diff, plate 1 would be connected to the front axle, plate 2 to the rear, 3 to the front, etc...

Then it is filled with a viscous fluid (where the name comes from), and the properties of that fluid causes it to "bind" when heated. The friction on the fluid from the plates spinning at different speeds (which occurs when the axles are moving at different speeds, obviously) causes the heating of the fluid, making the diff "seize", and the plates are forced to spin at the same speed, creating 50/50 split.

The property I was referring to that I experience, is that when I have loss of traction to front wheels, it takes a good 2 seconds or more for the power to go to the rear. This would account for the time of the fluid to become more solid, and cause the corroboration of the two axles. This is also why it is un-suited for front and rear diffs- by the time it kicks in, it's behind the curve.

Now, to elaborate, I thought the split was controlled via electronics to be 90/10 (front) for the US models, and when wheel slippage occurred, the mechanical aspects of the viscous diff (see above) over-rode the sub-par split, and cause the car to split the power 50/50. This would work, because the front and rear could be individually controlled via the TCU, but at the same time, the loss of traction and binding of the diff could merely cause the signal to be over-ridden, easily controllable by a passive sending unit.

It's very long winded, and probably completely wrong, but that's my $.02

Craig

[Nomake Wan]

Okay, I guess I have to jump in, here...

I translated a Japanese website a while back which stated that different countries got different transmissions. The UK, Austria, New Zealand, and some other countries got the same VTD-4WD transmission that Japan had.

The other countries, like Canada and the USA, got the ACT-4 transmission, the split of which is 90/10 normally. Same transmission used on the original Legacy all around the world.

The site is here: http://isuzupiazza.fc2web.com/whatsvx.htm

I translated it here: http://www.subaru-svx.net/forum/show...67&postcount=4

Enjoy.

[Trevor]

Craig,

Unless what has been continually posted here is incorrect, the US SVX incorporates a centre clutch, rather than a differential of any sort.

For all to think on, I repeat my questions ---

From what official source do the US figures originate?

Exactly how is there a torque split (front/rear) during normal driving, when there is no wheel slip, and as I understand it, only a clutch involved (in the US system) in respect of torque distribution?

i.e. When a US SVX is travelling in a straight line with all wheels rotating at the same speed, drive to the front, and a clutch transmitting power directly to the rear with no clutch slip.

[svxistentialist]

Originally Posted by Trevor

Craig,

Unless what has been continually posted here is incorrect, the US SVX incorporates a centre clutch, rather than a differential of any sort.

For all to think on, I repeat my questions ---

From what official source do the US figures originate?

Exactly how is there a torque split (front/rear) during normal driving, when there is no wheel slip, and as I understand it, only a clutch involved (in the US system) in respect of torque distribution?

i.e. When a US SVX is travelling in a straight line with all wheels rotating at the same speed, drive to the front, and a clutch transmitting power directly to the rear with no clutch slip.

Trevor, the following is verbatim from the Technical Service Information description I have in the ATSG manual. It does not quote numbers, but offers an idea of the control mechanism.

<<<
This transmission is SUBARU's unique, elctronically conrolled Multi-Plate-Transfer (MP-T) system, full-time 4WD automatic transmission. Its design is based on the above-mentioned electronically controlled, 4-speed fully-automatic transmission for 2WD. This transmission is equipped with a transfer hydraulic controlled system, including a duty solenoid and hydraulic multiplate clutch in the transfer section on the rear of the transmission.

The transmission control unit stores optimum transfer clutch torque [duty ratio] data corresponding to various vehicle operating conditions. As signals are sent from various sensors indicating actual operating conditions [car speed, throttle opening angle, range position, wheel slip, etc] the control unit selects the optimum duty ratio for current conditions. It, then, controls the torque of the transfer clutch [hydraulic multi-plate clutch] by activating the duty solenoid to control hydraulic pressure. This system allows the transfer clutch torque to be finely controlled corresponding to the various vehicle operating conditions, in contrast to the conventional 4WD automatic transmission where transfer clutch torque is controlled only by vehicle speed and throttle angle.

<FEATURES>

In addition to the features of the 2WD automatic transmission, the 4WD has the following features:
1) It permits transfer clutch capacity to be finely controlled by an electronic system. This eliminates the tight corner braking phenomenon which occurs when the steering is turned fully at low speeds.
2) It permits optimum distribution of rear wheel driving power corresponding to engine output and gear position.
3) It facilitates non-slip control. This improves the ability to get off bad roads or slippery surfaces at low speeds.
4) The '1st hold' range improves driving performance.
>>>

As you can see here, the torque output of the multiplate clutch is controlled by the duty ratio of the solenoid, Solenoid C I think. The rate applied for given sensor inputs is measured against a map in the control unit and then applied accordingly.

The last feature mentioned is another name for "Manual" in the SVX gearbox. I think 1st hold is a more correct descriptive of what it does, the word Manual is a misconception waiting to happen.

Hope that helps.

Joe

[Trevor]

Thank you Joe.

However I am fully conversant with the operation of the system. What is more I am able to differentiate between sales dept generated hype and fact.

My questions remain unanswered. In particular regarding the origin of the figures quoted covering the torque split in respect of the US clutch controlled system.

Cheers, Trevor.

[Hondasucks]

Would have been a lot cooler if "manual" mode made it act as a manumatic, with 1 being 1st gear and D being 4th gear, within reason of course (automatically upshifting at redline, not allowing a downshift that would over-rev the engine, and not allowing an upshift that would lug the engine too much (i.e., 4th gear at 20MPH) I think some of the automatic Supras had a "Manual override" function that basically did this... Would have been a lot cooler than a "Manual" button that was really a "2nd gear start", which is what it should have been labeled. (Some Toyotas have a "2nd gear start" button, not sure exactly but I think it's on the new Land Cruisers, and it does exactly what the Manual button does in the Subarus)

[Trevor]

Originally Posted by Trevor

Thank you Joe.

However I am fully conversant with the operation of the system. What is more I am able to differentiate between sales dept generated hype and fact.

My questions remain unanswered. In particular regarding the origin of the figures quoted covering the torque split in respect of the US clutch controlled system.

Cheers, Trevor.

Joe.

[svxistentialist]

Originally Posted by Trevor

Thank you Joe.

However I am fully conversant with the operation of the system. What is more I am able to differentiate between sales dept generated hype and fact.

My questions remain unanswered. In particular regarding the origin of the figures quoted covering the torque split in respect of the US clutch controlled system.

Cheers, Trevor.

Originally Posted by Trevor

Joe?

Trevor, apologies for the non-reply, I've been busy at work and did not have time to "dig" further.

I don't have the USA manual any more, and the detail I supplied above came from a Technical Auto Transmission repair manual for the Subaru 4EAT, and not from a sales or marketing leaflet. The information in this manual is itself culled or cloned from the Subaru WS manual.

Unfortunately it does not offer a percentage on the torque split, I did mention that in my post.

I have seen the split reported as 90/10 and as 85/15. I don't know which is correct, possibly the 90/10? I don't know for sure, but I would hope some of the USA members could come up with a reference, seeing as how it is their gearbox variant.

Joe

[Trevor]

Thanks Joe,

The extreme front rear/bias reported in respect of the US cars, is rather strange and would account for the under steer criticism. If this could be brought more towards 50/50, possibly these cars could be transformed and I have an idea in this regard.

My second still unanswered and open question:-

Exactly how is there a torque split (front/rear) during normal driving, when there is no wheel slip, and as I understand it, only a clutch involved (in the US system) in respect of torque distribution?

i.e. When a US SVX is travelling in a straight line with all wheels rotating at the same speed, drive to the front, and a clutch transmitting power directly to the rear with no clutch slip.

[svxistentialist]

Well, that's a fair question, and I don't actually have the answer, other than what I surmise from the ATSG Repair manual description as above.

If we take it that in normal dry road operation the split for argument's sake is 90 front, 10 rear, and that when driving on sheet ice rear massive wheel slip is detected, and torque to the front axle is reduced and torque to the rear axle is increased, thus giving an optimal balance of 50/50 in the worst conditions, then other percentages must lie between these extremes.

That would seem to imply that the fully engaged rear clutch multiplate is capable of delivering 50% of the engine drive torque to the back, but that in normal dry road operation it actually only delivers 20% of this amount, and this is set in the software based on a duty ratio setting for the C solenoid, and hence by the hydraulic pressure on the plates.

Does this make any sense?

Joe

[crazyhorse]

Originally Posted by Trevor

Thanks Joe,

The extreme front rear/bias reported in respect of the US cars, is rather strange and would account for the under steer criticism. If this could be brought more towards 50/50, possibly these cars could be transformed and I have an idea in this regard.

My second still unanswered and open question:-

Exactly how is there a torque split (front/rear) during normal driving, when there is no wheel slip, and as I understand it, only a clutch involved (in the US system) in respect of torque distribution?

i.e. When a US SVX is travelling in a straight line with all wheels rotating at the same speed, drive to the front, and a clutch transmitting power directly to the rear with no clutch slip.

My pet theory, and it's just that, is that the "torque split" is the Hype you mention. It's more like "duty cycle release" as the Duty C increases, or decreases its pulse width. Theoretically this would create a torque split, but only in the broadest sense. Rather than being true AWD it's an electronically coupled 4wd. Any time the transfer clutch is engaged it's 50/50, I'm guessing that in a straight line, with no slip, the Duty C is operating at 10% cycle closed, 90% cycle open. That's why some of the XT6'ers can put a switch on the Duty C solenoid to "lock" the transfer clutch. NOT something I'd want to monkey with personally.

Just my opinions, for what they're worth.

[NiftySVX]

The us transmission does have the clutch type system (I've layed eyes on it myself). The figures for the "nominal" torque split are really kind of hard to interpret. The best way to explain it is that the computer, at any given time, is between 90/10 and 50/50. It varies so much and so often that it is really hard to pin down an exact value, I think that is why it is so illusive in the technical manuals. It is determined by the programming in the computer, but of course it can never exceed 50% rear because it would have to somehow disconnect the front wheels which are driven directly via the reduction gear. In first gear, with pretty heavy throttle, it is way more than 90/10. On the highway, it is probably pretty close to that. The system operates in what i call "predictive modeling". It works to anticipate conditions, for example the TCU may see the following inputs: Light engine load, low throttle, 4th gear. It would then operate the transfer clutch in a conservative manner, and not get real aggressive with applying the transfer clutch, because conditions for wheel spin are unfavorable (not likely) so it may be almost completely FWD. On the other hand, it could see: high engine load, large throttle opening. These are favorable conditions for wheel spin, therefore, the clutch is applied more aggressively to prevent wheel spin from ever occurring in the first place. And of course, if there is a large difference between the speed sensor on the front diff and the one on the rear output shaft,(more than just normal difference due to turning) it defaults to a virtual "lock" condition. It seems to fluctuate somewhere inbetween the two extremes just depending on what inputs it receives. The whole idea of it is to prevent wheel spin from ever occurring, which at the time was very advanced at the time. Some others (like toyota) used a wet clutch like the Subaru 4EAT, but it did not attempt to prevent wheel spin, it only handled wheel spin once it had already occured. So, the answer to what is the nominal torque split is it depends. Is it 90/10? yes, sometimes. Is it closer to 35/65? When it needs to be, yes

[Nomake Wan]

These will have to be confirmed by Phil, since he can make heads or tails of the TCU data.

From a Subaru tech guy:

"The TCU will split the torque however is needed according to a program unless the speed sensors indicate a slip. You can observe this by looking at the data stream from the TCU for the MPT clutch solonoid (solonoid C). As you accelerate the split will be 50/50 no matter what gear range, when you are cruising it goes to FWD (95%) to save fuel.

If the A/T stayed at 90/10 like is often stated the MPT clutch would wear out in a month. This is clutch pack and it will not last if it is slipping. It will be at partial engagement only to lesson shock, otherwise 0% duty is 50/50, 95% duty is FWD or 100/0. There is no 100%duty.

Under hard braking (ABS) the A/T shifts into 3rd with no engine braking (coasting) and the split goes to 100/0 (FWD)."

This brings up the age-old question of "Am I damaging something in FWD?" If it's true that 95% duty cycle means the clutch pack is 0% engaged, then the only thing you could be damaging would be your front half-shafts (if you drive aggressively). On my old Legacy FWD, I did snap a half-shaft or two while doing silly things.

Anyway, it does make sense. Makes me wonder what's up with the Japanese transmissions. All the translations I've read make perfect sense... but if I add in the duty cycle split information I just learned, then stuff starts to fall apart again. I'm going to need a little time to think about it.

[Trevor]

My question:-

Exactly how is there a torque split (front/rear) during normal driving, when there is no wheel slip, and as I understand it, only a clutch involved (in the US system) in respect of torque distribution?

After having slept on it, the answer:-

When the car is travelling on an even surface and in a straight line, the road and forward motion in effect lock the front and rear wheels as if they were connected by a chain or a toothed belt.

Torque can be applied to the front wheels with less to the rear because the rear can be partly coasting and be pulled along i.e. driven by the front. In this situation the clutch need not be applied at full pressure and a state of tension can exist between the plates but there will be no real slip.

Say 90% torque front, 10% rear results in the missing 80% being provided by the coupling of forward motion. When the car enters a corner in these circumstances, and the situation changes, there is surely the foregone conclusion of understeer.

It will be seen that much of the possible issue of clutch wear which has been suggested is negated.