Poll: Would you spend $1000 for complete 5EAT TCU programmability?

[fishbone]

For a geared tranny control is important because you can dictate shifting behavior, shift points, stress on clutches, etc.

With a CVT, there are no gears. I am somewhat confused as to what you would like control over?

[ClimberDHexMods]

Center diff open

 

The way our 5eats work. This would mean the car will stay in 45/55 mode and will not go into 50/50

*There is only 2 operating modes for the center diff

 

This can be a plus

 

The 5eat planetary center diff doesn't work that way. As I understand it, 5eat functionality is similar to early STi DCCD that operated around an open center diff

 

What happens is we have 46/54 front/rear TORQUE split when center diff is unlocked. But when wheel spin or any loss of traction is detected, and you have the rear drive shaft suddenly spinning 1.2x faster than the front pinion (for example), you still have that 46/54 split, but the actual speed difference between front and rear outputs could be anything. To make the front and rear output speeds be the same, it is necessary to activate the center diff clutch pack, which effectively locks the two outputs together. The 50/50 isn't so much a power distribution selection, as it is a anti-slip selection. So if you want to have an open diff all the time, that is fine, but you may find that you suddenly end up with rwd from time to time.

[underground000]

^Not sure what you mean. Torque is actively sent front/back based on conditions. If you are braking, it goes forward. If you are accelerating it goes to the rear. You do not have to have slip for this to occur. It sounds like the center diff is being made dumber? We have an active awd. Why make it passive?

 

I thought whenever enough slip occurs on a wheel (speed sensor) the center diff activates becoming 50/50 power transfer. And I thought the center diff only has 2 operating modes. 45/55 and 50/50

[underground000]

The 5eat planetary center diff doesn't work that way. As I understand it, 5eat functionality is similar to early STi DCCD that operated around an open center diff

 

What happens is we have 46/54 front/rear TORQUE split when center diff is unlocked. But when wheel spin or any loss of traction is detected, and you have the rear drive shaft suddenly spinning 1.2x faster than the front pinion (for example), you still have that 46/54 split, but the actual speed difference between front and rear outputs could be anything. To make the front and rear output speeds be the same, it is necessary to activate the center diff clutch pack, which effectively locks the two outputs together. The 50/50 isn't so much a power distribution selection, as it is a anti-slip selection. So if you want to have an open diff all the time, that is fine, but you may find that you suddenly end up with rwd from time to time.

 

I said it because 1st post said you may not have control of the center diff lockup control. And thought lockup would make it 50/50

[cipher_nemo]

For a geared tranny control is important because you can dictate shifting behavior, shift points, stress on clutches, etc.

With a CVT, there are no gears. I am somewhat confused as to what you would like control over?

 

Ratios algorithm, delays in adjusting ratios, when the ratios change, when the CVT switches back to auto when in temporary manual mode, timing and speed of manual ratio changes, etc., etc. Use your imagination here. Just as you might want control over your EAT, I want control over my CVT. The CVT is much more complicated in terms of algorithms since it is always adjusting ratios throughout the entire power curve. I could understand someone asking your question for EAT, but for CVT things are even more processor intensive.

[fishbone]

I thought whenever enough slip occurs on a wheel (speed sensor) the center diff activates becoming 50/50 power transfer. And I thought the center diff only has 2 operating modes. 45/55 and 50/50

I think ClimberD pwned this explanation up above. Except he is incorrect on a teeny-tiny bit. It's 44.3% front, 55.7% rear

If we only had 45/55 and 50/50, why would 5% make driving behavior so radically different? I don't know what the maximum transfer bias can occur front vs rear but under slip I imagine it can be 10/90% front/rear. In my own experience, the front wheels were barely spinning while the back wheels were melting a hole in the ice I was on. It was only after the center diff locked up that they were spinning equally and thanks to the LSD locking up as well, that ONE wheel that had traction, got me out. LGTs are so awesome

[fishbone]

Ratios algorithm, delays in adjusting ratios, when the ratios change, when the CVT switches back to auto when in temporary manual mode, timing and speed of manual ratio changes, etc., etc. Use your imagination here. Just as you might want control over your EAT, I want control over my CVT. The CVT is much more complicated in terms of algorithms since it is always adjusting ratios throughout the entire power curve. I could understand someone asking your question for EAT, but for CVT things are even more processor intensive.

Right, but I was approaching it from the perspective of "WOT vs NOT".

At WOT, there's still a bunch of stuff the geared tranny does. A CVT, it holds the revs at peak power. I may not be much of a dreamer here and I am certainly not knocking the CVT, but for us we have shift points which deals with when, how, how long, line pressure, solenoids, etc. If you are operating in "manual mode" which is only a simulate geared change with pre-set ratio points then it is truly unfortunate CVT is the only offering on the 2.5i these days. You don't run the risk of glazing or burning clutches due to shitty shifting. The single most significant point of failure in a CVT is the belt.

[ClimberDHexMods]

Except he is incorrect on a teeny-tiny bit. It's 44.3% front, 55.7% rear

 

You must not have heard, the cool kids are into rounding

 

Must have been your back wheel because front and you'd be going nowhere.

[fishbone]

You rounded the wrong direction though

Yes, it was a rear wheel that got me out of trouble. It took like 10 seconds of slipping for stuff to lock up though, it was kinda disheartening to see the one major drawback of a viscous-type awd in a real-life scenario. 10 seconds seems like an eternity.

[cipher_nemo]

The single most significant point of failure in a CVT is the belt.

 

That would be outdated information in regards to CVTs. The 5th gens do not use a rubber belt, they use a grouping of chains and are estimated to be one of the most reliable parts of the transmission.

[ClimberDHexMods]

You rounded the wrong direction though

Yes, it was a rear wheel that got me out of trouble. It took like 10 seconds of slipping for stuff to lock up though, it was kinda disheartening to see the one major drawback of a viscous-type awd in a real-life scenario. 10 seconds seems like an eternity.

 

Oh hell I'm going to take myself out back now

 

Viscous differential lockup is heat-dependent. If I understand the process correctly, it will work a lot better in a pre-heated differential. Since you were using it in frozen conditions, it took about 10 seconds to heat up enough to start transferring torque to the still wheel. Torsen is nice because you don't have to wait for fluid to chemically react to the slip:)

[fishbone]

That would be outdated information in regards to CVTs. The 5th gens do not use a rubber belt, they use a grouping of chains and are estimated to be one of the most reliable parts of the transmission.

That is what I meant, I just used the wrong word for it. I did not say it was not reliable. On the contrary. I simply stated where the stress in a geared tranny goes on clutches, in a cvt it goes on the belt/chain/whatever drives the pulleys. When a clutch fails, the equivalent in a CVT is the chain failing and slipping on the pulleys, or worse, snapping.

Another aspect is the fact that this discussion is motivated by the GT crowd modding their engine and desiring control over the 5EAT to protect it and improve performance. On a stock car, this is almost a non-issue.

If we address specific functionality you do not like or would like control over, who knows, there might be something for you out there.

[cipher_nemo]

That is what I meant, I just used the wrong word for it. I did not say it was not reliable. On the contrary. I simply stated where the stress in a geared tranny goes on clutches, in a cvt it goes on the belt/chain/whatever drives the pulleys.

Another aspect is the fact that this discussion is motivated by the GT crowd modding their engine and desiring control over the 5EAT to protect it and improve performance. On a stock car, this is almost a non-issue.

 

Ah, I understand what you mean now. And yes, seems all of the modding/tuning goes towards the GT crowd since pretty much nothing for me with a plain 2.5i Legacy except for the old bolt-on intake/exhaust stuff.

[fishbone]

The GT "gets the love" sort of speak because there is legit power to be had out of it.

[BAC5.2]

Oh hell I'm going to take myself out back now

 

Viscous differential lockup is heat-dependent. If I understand the process correctly, it will work a lot better in a pre-heated differential. Since you were using it in frozen conditions, it took about 10 seconds to heat up enough to start transferring torque to the still wheel. Torsen is nice because you don't have to wait for fluid to chemically react to the slip:)

 

No it isn't.

 

Viscous couplings are shear driven. They aren't chemical reactions either, shear is a mechanical property. Fluid viscosity increases as shear rate increases, and the differential locks up. No shear, no slip, no lockup.

 

Viscous diffs don't rely on expansion due to fluid temperature increase.

[ClimberDHexMods]

Then we said almost the same thing through semantics. I'm not an engineer and I don't know the lingo, so bear with me. The difference in speed of the viscous coupling plates produces heat in the fluid between the plates, which in turn causes the fluid to heat up and thin out and result in more friction between the plates via the fluid. Before the fluid gets a chance to heat up, it will still perform its function. But after it heats up from prolonged exposure to friction, it will work better than when cold. Is this correct or am I off?

 

I never said expansion. Yeah heat from friction is not chemical, my b.

[BAC5.2]

Then we said almost the same thing through semantics. I'm not an engineer and I don't know the lingo, so bear with me. The difference in speed of the viscous coupling plates produces heat in the fluid between the plates, which in turn causes the fluid to heat up and thin out and result in more friction between the plates via the fluid. Before the fluid gets a chance to heat up, it will still perform its function. But after it heats up from prolonged exposure to friction, it will work better than when cold. Is this correct or am I off?

 

I never said expansion. Yeah heat from friction is not chemical, my b.

 

It is not a heat thing. Its a shear thing. As two plates submerged in fluid spin at different speeds, the fluid thickens. As that differential increases, the fluid behaves like a fluid with higher viscosity.

 

Fluid properties are temperature dependent, which is why viscous differentials are less effective at low operating temps.

[ClimberDHexMods]

Ah thank you for the explanation!

[seabass07]

I'm a bit late to the party...

 

For me, $1000 is a bit too much for just a software solution. If it was $1000 for full installation etc, then it would be worth it. But $1000 for me to do the install and tuning is a bit steep. I also just spent $5000 to move across country...so I'm feeling a bit cheap right now. Other than that...

 

I could lose cruise control if it meant full control of my trans.

 

I have no problem disabling a CEL with romraider.

 

I can live with dash errors.

 

I would prefer to have control over the torque converter lockup. After towing a 4000lbs trailer, I REALLY wish I had the ability to lock up the TCC at will while towing through the mountains.

 

I would personally prefer an on/off switch for the center diff. It would mean that I could lock the diff when going into rough conditions. If I'm in deep snow, I'd rather just leave it on. I also wouldn't have to wait for wheel spin for it to lock up. But I wonder how the clutch pack would hold up if I left it on by accident in dry conditions...or more likely, if the wife left it on.

 

BTW, it's easy to get lost in the 45/55 to 50/50 talk. The center diff does have 2 modes, but they are simply open (45/55) and locked by the clutch...50/50 is misleading. As Frank_ster put it in a PM...it's more like 100/100. It's fully locked and the front and rear diffs have access to 100% of power until the clutch pack slips. It doesn't just change the distribution. And it is Subaru PR that leads to all of the power transfer talk. It doesn't actively transfer any power to any axle other than simply locking the center diff. They like to make diagrams of power flowing to the wheel with the most grip...

 

I understand the limitations of the megasquirt and the pricepoint reflects those limitations. But $1000 is definitely steep. If the price was lower on your option, then it would be the best option for me. I have no issue with taking the time to tune it exactly as I want it. So for me, all of the negatives are livable, but the price is not. Having said that, when compared to the cost of just a valve body mod, you get a LOT more for your money with this option. Maybe I'd feel different about it if I had a transmission fund with $1000 in it. I'd prefer this over a transmission upgrade or VB mod.

[frank_ster]

i actually upgraded to megasquirt 3 with the mega shift . and i have figureed out a way for the mega shift and mega squirt to do absolutely everything the stock tcu will do .. as for the cruse light and and cruise not working i think we can some how simulate the car becoming manual tranny.

[ClimberDHexMods]

i actually upgraded to megasquirt 3 with the mega shift . and i have figureed out a way for the mega shift and mega squirt to do absolutely everything the stock tcu will do .. as for the cruse light and and cruise not working i think we can some how simulate the car becoming manual tranny.

 

I would love to hear what you have come up with! We will be in touch.

[frank_ster]

hey yall ..

so mega squirt 3 is new from the squirt guys..

 

ms3 has another expantion module that allows for 8 sequential injection and 8 sequential ignition modules. what does this mean ? well if your not using 8 cylinders like i am i have 2 more pwm outputs. ready to use..

 

what does that mean ? well the amazing guys at ms provided tables for these extra out puts to be able to contol how everyou like. these tables can be simply rpm vs map or map vs mph . or or or .. they can be used for variable cam timming .. fancy boost contoroler.. or center diff and or torque lockup

currently the center diff on the subaru works with map more throttle you give the tighter is gets. basically you don't need much center diff action when your putzing around at idle. but you do as you give her a shot of throttle, so this will act the same as stock.

the torque converter on the ms shift was originally on and off depending on speed. but if you were to combine this with the table where you could have it engage as its pulling through the gear as stock !

 

and the ms3 can be configured to be a tranny controler on its own ! so having up wards 0f 10 pwm outputs. standalone sd card data logging usb and serial conectivity.

 

with this stuff i feel confidant i can make my car stand up and tango !

[seabass07]

Thanks for that!

[Lookatmyrust]

Another thing i think people would like is of you didnt have to cut wires. can you just have plugs included that will do this?

[seabass07]

I don't think there is a plug and play solution for this type of thing. I think you have the option of keeping cost down and doing a LOT of work yourself or paying a lot of money for professional installation and tuning. At least that is how it appears from my limited research.