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

[ClimberDHexMods]

Frank_ster,

 

Is your plan to simply piggyback the megashift controller over the stock TCU to try to fool it into being happy? That's my hope/intention. I'm sure you know, there is a fluid pressure on/off switch on each of most solenoids, so that will need to be tricked, but can perhaps be done with the stock TCU's solenoid outputs themselves. Perhaps separately, resistors may be used to fool the stock TCU into thinking it still controls all solenoids. Then there are purposeful solenoid disconnects that can lock the TCU into a set gear so it doesn't pull timing on what it thinks are still shifts... but there may be a failsafe through CAN... so a lot of guess and check is my plan until one of us gets a better idea of how boobytrapped the TCU is against contradictory input signals.

 

 

Separately, I don't know what your center diff frictions looked like when you pulled them apart each time, but mine are worn after only 30k miles.

 

With this as a concern, I'm now thinking center diff control would be best accomplished with simple ON/OFF functionality (aka 100 or 0 for those who like nice ratios) to prevent slipping. While mine are not burned, they do appear to be missing a lot of clutch material. The PCS controller or megashift can perhaps use 2 separate inputs (lateral G sensor and MAP, or lateral G sensor and TPS; second one would react more intuitively). With these to determine center diff on/off, either controller can then output simple on/off command to the center diff.

 

Though I will first try to get the stock TCU to keep command of the center diff lockup solenoid since it will still receive all the inputs it needs...

 

 

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?

 

There are connectors out there somewhere in the world, but I have had no bites trying to track down Subaru connectors. My thought is to cut the entire harness and add in molex or fancy deutsch connectors. So if you part out you just plug in your two aftermarket connectors to complete the TCU harness. Homemade harnesses are nice, but they can be a lot of work.

[frank_ster]

david .. what makes you think the clutches are worn out ? did you measure thier thickness ?

 

i don't see how any kind of abuse could ever wear these clutches out .. other than runing diffrent size tires front and back.

these clutches are the same type used for shifting they are very large and probably only see maybe 30 revelutions on a normal day.

[seabass07]

Could differences in tire wear front to back have caused it? I'm curious about how sensitive it is.

[ClimberDHexMods]

david .. what makes you think the clutches are worn out ? did you measure thier thickness ?

 

i don't see how any kind of abuse could ever wear these clutches out .. other than runing diffrent size tires front and back.

these clutches are the same type used for shifting they are very large and probably only see maybe 30 revelutions on a normal day.

 

I will measure to confirm. I use to be a ski racer, so lots of snow might explain it. They lacked the slotting and texture of other clutches.

[ClimberDHexMods]

More tech on center diff control with the PCS controller:

 

There are 9 PWM outputs and 3 digital on/off outputs. The 3 digital outputs are mostly useless...

 

I'm dedicating 4 PWM outputs to the 4 main shift solenoids

1 PWM output to Line Pressure Solenoid

1 PWM output to TCC solenoid.

So that's 6 definitely spoken for.

 

Things that can be either/or (PWM or digital, but would be slightly better as PWM):

A- The 5th shift solenoid, for Low Coast Clutch, which produces the forceful engine braking in 1st and 2nd gears. On our LGTs it is activated only in Sport Mode and even then only at certain unknown speeds. On the 350Z it is on 100% of the time in Sport Mode in 1st and 2nd gears.

B- An aftermarket boost dump solenoid, which I currently have as "PWM vs Shift cut", where I can actuate the stock BPV by intercepting the reference line by closing a solenoid for some tbd fraction of a second (since the car will shift very quick with this controller). This was Frank_ster's idea, and a great idea at that.

 

So that leaves a couple options for controlling the center diff with this specific controller.

I can do PWM vs Any Input, so its up to a creative imagination and comprehension of ideal center diff control (I lack both).

 

I have a mind to use two PWM outputs for the center diff: one based on TPS or MAP, and one based on the stock lateral G sensor.

I can use a relay to make one of these variables on/off, and the other variable based on the corresponding input.

For instance, one could do: PWM of varying % relative to the lateral G sensor. So if you are going perfectly straight, you would get 100% PWM on the center diff. Turning gradually would be 75%, turning sharply would be 0% PWM. Fill in the blanks in a sloping fashion. To add to that, there would be on/off functionality added to this. Below 5psi or 30% throttle, a relay would cut off the lateral G sensor-based PWM signal, and the diff would be open. Above that threshold value, the PWM signal that is always going in the background, would be allowed through. So I can set it up that way, and tune it by feel based on chatter and overall feeling of lateral grip.

I have read the DCCD threads, and they are all about using the full range of 0-100% duty cycle depending on conditions, which indicates that DCCD clutches are fine with some slipping, and that it can produce good driveability to do so.

Anyone have any thoughts on designing this with the rather similar 5eat center diff in mind?

[seabass07]

The way I understand what you are saying is that you want to use it only based on the lateral G sensor and available power (TPS or MAP). Does that mean you would activate it only during hard turns? It seems that if you lost traction during a hard turn that the resulting lower Gs from the skid would reduce your center diff engagement. It also leaves out the wheel speed sensors which would normally lock up the diff during slippage while starting out on a slippery surface.

 

If I had to choose between using the lateral G sensor or using the wheel speed sensors, I'd use the wheel speed sensors. Although I don't know if that is feasible with the limited outputs. Would you need all 4 sensors monitored separately?

I might be misunderstanding your intentions though.

[ClimberDHexMods]

I'm not quite sure I understand your post, but lets see if this answers it:

 

The lateral G sensor would be referenced to determine how much center diff lockup pressure to apply.

No lateral Gs (accelerating in a straight line) would be full center diff lock, which is ideal since you want the front and rear drive shafts to be spinning at the same speed.

A little lateral G (turning slightly) would be partial center diff lock.

A hard turn (like at a street corner or tight hairpin turn) would be no NO center diff lockup, which would be ideal since the front and rear drive shafts want to spin at different speeds when cornering.

 

A potentiometer could be installed to lower or raise the acceleration threshold on the fly to control at what level all of this center diff clutch logic would be activated (for snow/ice/rain/dry, etc).

 

I agree that using wheel speed sensors would be good, but I'm not sure they would add much value, nor do I have any idea how to interface with them. The DCCD Pro controller uses a 2-axis G sensor, and a hookup to TPS.

 

Stock looks like the very pretty illustration on page 3: http://dccdpro.com/main/wp-content/downloads/DCCDPro%20Auto.pdf

 

My version would require me to get a little more creative with my outputs, because as I have it written above, my acceleration function is on/off, not a pretty slope. But I don't know enough about electronics to know how to take two PWM power or ground outputs and tie them into one solenoid... or anything along those lines...

[seabass07]

OK. I misunderstood you. I didn't realize you would be keeping the diff locked under no lateral G. I was thinking the opposite for some reason. Is there reason to be concerned about keeping it fully locked most of the time. Would this have any impact on wear on the clutches since they would absorb any difference in tire size? The wheel speed sensor would be beneficial in reducing the amount of time the diff is locked when going in a straight line, but I can't think of an easy way to do that with so few outputs.

[ClimberDHexMods]

The diff would be locked under no lateral acceleration, but only above x acceleration threshold. Having it locked during light acceleration or cruising on the highway would be pointless since there would be no chance of slipping (except on ice) under those conditions. So 98% of the time, the diff would be unlocked. The only exception would be on a race track, where it would be locked partially or fully a lot of the time. But even then the rules don't change. I'm not so worried about tire wear... I have snow tires I need to burn up before I can switch the summers back on... and if I break a rear vlsd, oh shame

[seabass07]

I meant clutch wear due to different wear on tires, not tire wear.

[frank_ster]

the clutches for center diffrential do not have groves or anything in them ..

 

when i purchased my car it had new japanese tires in the front and the stock re92' in the back. and clutches were fine ..

as for spining ? i live in winter land .. normal drive to city is 70 km .. and during winter my clock says it takes over 200km ! thats how much spining i'm doing !!!

 

i'm sure the clutches are almost brand new..

 

as for pwm vs digital out puts for shifting .. well i don't see that making any real diffrence.. i'm quite sure you will have them at 100% anyways. so they will effectivly digital anyways.

[ClimberDHexMods]

Good to know

 

For PWM vs digital on the shift solenoids, there is a 2d table in the PCS software where you can scale your shift solenoid ramp-up/ramp-down from 100% (normal soft cushy around town shift) to 0% (for WOT, meaning yes simple On/Off) depending on your MAP level.

 

I'm was more referring to the logic behind AWD lockup solenoid in mentioning PWM vs ON/OFF, wondering if there is really any extra value to be had in using PWM for varying amounts of lockup, vs digital's all/nothing selections, for AWD's 3D lockup table. You probably know a lot more about these things than I do.

 

Screenshots of the aforementioned ramp up/down of a pwm shift solenoid, though you can set it to ANYTHING you want, and I haven't tweeked these details yet, as I'm yet to video record FreeSSM to see what the stock TCU is doing, for a base map.

 

http://legacygt.com/forums/attachment.php?attachmentid=97998&stc=1&d=1301077037

 

http://legacygt.com/forums/attachment.php?attachmentid=97999&stc=1&d=1301077037

[seabass07]

Out of curiosity, what are your plans for downshifts with 0 throttle for engine braking?

[frank_ster]

ok I see what your getting at. but your making it more complicated than it needs to be.

 

typically speed of shift .. or shift firmness is controlled my simple orifice with dictates time to fill the shift clutch. that is combined with line pressure to dictate over all shift time.

 

i see you have the ablity to control the valve slowly open then begin shift and then slowly close to finish off the shift as you would drive a manual tranny.

 

I feel that your reading to much into this, the important factor with auto tranny is the the torque converter. ie big marshmellow this will absorb any harsh shifting. at lower throttle positions, and this is also combined with lower line pressure.

 

i'm fairly certain giving full power to the shift sol when its time for a shift it will still be lacking with the big hp numbers. modification to the valve body ie ipt or just drilling some holes bigger will be necessary.

 

for control i am only using megashift over 5 microswitches on a "punch card " or "cams"

and full line pressure all the time for shifting.

Is because i want to try the electronic control.

 

the tcu you are using is amazing how everything it can do and control but its way more control than any one could ever use.

[ClimberDHexMods]

I see what you are saying.

 

Shame to hear you don't think the stock valve body will hold the high HP even at 100% of its mechanical ability. I wouldn't know since I've had a modified one for a while, which I am hoping to one day sell. But in comparing the RE5R05A and 5eat valve bodies on paper, I bet you or I could adapt a TransGo kit with little trouble. It comes with springs...

[frank_ster]

have you taken apart a ipt valve body to see what was changed ?

[ClimberDHexMods]

No, but seeing as I have both an unmodified and a IPT modified valve body sitting here uninstalled, now would be a really good time to find out...

[ean611]

No, but seeing as I have both an unmodified and a IPT modified valve body sitting here uninstalled, now would be a really good time to find out...

 

I'm very interested in this. I'd love to know what they really do.

[frank_ster]

they probably drill like 3 holes and change a spring

[ClimberDHexMods]

LieutenantColumbo's TransGo VB vs IPT VB would indicate about half the work going into the IPT VB as the TransGo retrofit. But he also had his trans rebuilt at time of switch, and a new trans does feel different than a worn one.

[fishbone]

ClimberD a comparo of the IPT vs stock would be priceless. I'm wondering if you're about to get a cease and desist

Like many of us here I am just curious what gets changed vs the price point

[ean611]

LieutenantColumbo's TransGo VB vs IPT VB would indicate about half the work going into the IPT VB as the TransGo retrofit. But he also had his trans rebuilt at time of switch, and a new trans does feel different than a worn one.

 

what do you mean "half"?

 

The transgo kit is still "under development" last I checked....

[ClimberDHexMods]

what do you mean "half"?

 

It's a largely uneducated guess on my part. How about we skip my guessing and wait until we know something more concrete.

 

The transgo kit is still "under development" last I checked....

 

There is a TransGo shift kit for the RE5R05A, of which our 5eat is a close variant. We have one extra solenoid for the center diff transfer clutch. Aside from that, the valve bodies are functionally the same, and are probably very similar in fashion.