My Touring Wagon EJ25 build, Part 2

User avatar
Alex
Elder Member
Posts: 5405
Joined: Wed Nov 23, 2005 10:00 am
Location: Perth
Contact:

Post by Alex » Thu May 31, 2012 9:08 am

So your low range is that low that it's useless??

Lol!

Sent from my HTC Velocity 4G using Tapatalk 2
my07 Outback
my13 Hyundai i45(shhhh)
my02 Gen3 Liberty limited ed.

previously
L-series wagon, LSD, EJ20turbo, 29in tyres, 'wanky wagon'
2000 gen3 outback, lifted, otherwise stock.

User avatar
thunder039
Junior Member
Posts: 998
Joined: Mon Jun 20, 2011 6:41 pm
Location: victoria AUS

Post by thunder039 » Thu May 31, 2012 9:33 am

glad to hear your happy with the car!

sorry i must of missed it what did you do for the rear LSD?
2004 subaru forester -gone
1999 subaru forester- no more :(
1989 subaru brumby- sold!
2008 zook jimny -sold!
2003 mitsubishi pajero - missus car
2013 nissan d22- set up for long distant touring

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Thu May 31, 2012 12:01 pm

I'm doing roughly 3,000rpm with 4.11's @ 100kph going by the GPS:)
Alex wrote:So your low range is that low that it's useless??

Lol!

Sent from my HTC Velocity 4G using Tapatalk 2

Haha, Na its not as low as that, Its just that when it was in low range, the revs were really high and the car didn't get much speed climbing the dune in 1st.
So thats why i stuck to normal for the day...

thunder039 wrote:glad to hear your happy with the car!

sorry i must of missed it what did you do for the rear LSD?
It though i still had the pics of the rear diff. Must have been in my old thread.
Nothing special, just used a CLSD diff off an RX and changed the crown and pinion to 4.11 to match the gearbox

Image

Image
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
El_Freddo
Master Member
Posts: 12175
Joined: Tue Oct 04, 2005 10:00 am
Location: Benders (aka Bendigo) Vic

Post by El_Freddo » Thu May 31, 2012 6:31 pm

d_generate wrote:I'm pretty sure seagull only shaved a corner off my 1.44 crown wheel but I guess the 1.19 gears are a much different diameter.
Wow Tom! That's a super tall diff - 1.44!! :p
guyph_01 wrote:The new box is great, I love the 4.11, it give the car this great acceleration feel in all the gears over the 3.7:p

Low range was useless up the sand dune, It was too Low
Both of these statements make me excited about getting my gearbox in :twisted:

I'm guessing you're happy with the changes this box has made for your subi?

Cheers

Bennie
"The lounge room is not a workshop..."
Image
El Freddo's Pics - El_Freddo's youtube

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Thu May 31, 2012 7:46 pm

El_Freddo wrote: I'm guessing you're happy with the changes this box has made for your subi?

Cheers

Bennie
Yeah more than happy. Only thing now is to get rid of that bearing noise. Hopefully only the rear transfer case will have to come off to fix it.

Don't think i'd wand a front LSD. Main reason is for the sand i don't really need one. On your side of the country though maybe id use one as you have a lot more dirt/mud tracks. But i'd hate the locking feeling on the road..
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Wed Sep 26, 2012 1:44 pm

hi guys,

Its been a while since i updated:)
Well the car has been driving as a daily since the gearbox rebuilt. Occasionally driving the temp project cars around but mainly the L.
Normal wear and tear has caused a steering rack boot and a rear cv boot to fail. Temp fixed with duct tape for the moment but will have to fix them before i head out in the sand again.
Rear arms bushes and bearings are making a hell of a noise. Those have never been changed since i owned the car. New bushes and bearings are already installed on another set of arms but waiting to get the rear 5 stud hubs to fit them unless one side fails soon.

On a positive side, Dave was able to sort out my check engine light and hunting revs issue. The translated jap diagrams i found turned out not to be the right ones as no colour wires matched but allowed us (Well mostly Dave) to understand a bit more on my wiring. ****, everything is controlled by the ecu, even the rear glass dimister/defogger!!!
First thing first we found out that the alt is supposed to be controlled by the ecu. When Dave cut the loom blind, it wasn't hooked up and cause high input to the maf causing an engine code. The alt is charging at 14.4-6.
Unfortunately the wires on the ecu don't match so we have no clue where to hook the wire from the alt to.. So i've got two options, find the right wiring diagram (unlikely) Or replace that ecu with a std one from an early wrx, foz or outback to get the lower charge. So no change there yet!!

Then Dave tackled the starter circuit. He hooked the test light to the starter wire from the ecu and started the car. the test light went on when the key was turn till the engine starter and Dave didn't understand why the code was coming on as this show it was connected. He did that a few times till we realised that when the car was started with the test light in the circuit, the check engine light was off and the car didn't rev up and down by itself. When the test light was off, the engine light back on and it reved. weird!!!
So Dave worked out that there must have had a relay in there that provided that resistance and the ecu recognised it. So we added a relay just to provide the resistance and TADAA it was fixed:D lol.

So yeah, that simple.
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Wed Sep 26, 2012 1:54 pm

Now,

We also were going to hook up the DCCD Pro controller but we weren't sure it would work properly??
Ok so, for a track car, with acceleration, cornering G, the DCCD controller is activated and locks the diff. But the thing is i don't really want it to work like that. We hooked the to a power source and played with the G sensor and it did confirm that it most probably not engage on a flat section or not when needed. I just want it to engage for the 'Tuff' section to help me pull out when offroading.

Dave mentioned that in his Auto foz, he hooked the diff lock so he could lock it and unlock it at will, bypassing the tcu then allowing it to work as normal.
He also said that engaging the diff lock for a whole day or even an hour+ isn't good and will burn it out.

So we came up with an idea. Hook the DCCD to a momentary switch that goes through a timer capable of 5 sec to 3 mins. We assumed 1 or 1.5mins would be plenty and if not all you do is hit the switch again.

Would you guys think this is a good idea?
Thx
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
taza
General Member
Posts: 1417
Joined: Sun Dec 26, 2010 10:21 pm
Location: Perth, Australia

Post by taza » Thu Sep 27, 2012 3:10 pm

I like the idea, nice work! Does that mean that you would hit the switch and it would lock it 50/50 for that small amount of time? Then when unlocked but be like a standard AWD centre diff or 65/35 split?

You still have a bit of work to do though before we can hit the dunes....

User avatar
Gannon
Senior Member
Posts: 4580
Joined: Sat Nov 05, 2005 10:00 am
Location: Bowraville, Mid Nth Coast, NSW

Post by Gannon » Thu Sep 27, 2012 3:29 pm

taza wrote:I like the idea, nice work! Does that mean that you would hit the switch and it would lock it 50/50 for that small amount of time? Then when unlocked but be like a standard AWD centre diff or 65/35 split?
Yeah spot on, when locked its a 50/50 split, but when open, its 65/36 open diff with no LSD'ness. Its also variable between these two extremes too. You'd be surprised how many Subaru enthusiasts have trouble grasping that.
Current rides: 2016 Mitsubishi Triton GLS & 2004 Forester X
Ongoing Project/Toy: 1987 RX Turbo EA82T, Speeduino ECU, Coil-pack ignition, 440cc Injectors, KONI adjustale front struts, Hybrid L Series/ Liberty AWD 5sp
Past rides: 92 L series turbo converted wagon, 83 Leone GL Sedan, 2004 Liberty GT Sedan & 2001 Outback
------------------------------------------

User avatar
niterida
General Member
Posts: 1158
Joined: Sun Apr 13, 2008 5:45 pm
Location: Geraldton WA

Post by niterida » Thu Sep 27, 2012 3:43 pm

Gannon wrote:Yeah spot on, when locked its a 50/50 split, but when open, its 65/36 open diff with no LSD'ness. You'd be surprised how many Subaru enthusiasts have trouble grasping that.
My understanding is that the torque split is always 65/35 no matter how much lock is dialled in. The 65/35 is derived from a mechanical planetary gearset with differing sized gears that still give the same output speed. This mechanical connection cannot be changed therefore the torque split cannot be changed.
The DCCD simply adjusts the time the LSD plates are locked via a PWM (pulse width modulator), or the LSD'ness as you put it :)
[SIGPIC][/SIGPIC]

User avatar
niterida
General Member
Posts: 1158
Joined: Sun Apr 13, 2008 5:45 pm
Location: Geraldton WA

Post by niterida » Thu Sep 27, 2012 3:51 pm

guyph_01 wrote:Now,

We also were going to hook up the DCCD Pro controller but we weren't sure it would work properly??
Ok so, for a track car, with acceleration, cornering G, the DCCD controller is activated and locks the diff. But the thing is i don't really want it to work like that. We hooked the to a power source and played with the G sensor and it did confirm that it most probably not engage on a flat section or not when needed. I just want it to engage for the 'Tuff' section to help me pull out when offroading.

Dave mentioned that in his Auto foz, he hooked the diff lock so he could lock it and unlock it at will, bypassing the tcu then allowing it to work as normal.
He also said that engaging the diff lock for a whole day or even an hour+ isn't good and will burn it out.

So we came up with an idea. Hook the DCCD to a momentary switch that goes through a timer capable of 5 sec to 3 mins. We assumed 1 or 1.5mins would be plenty and if not all you do is hit the switch again.

Would you guys think this is a good idea?
Thx
I think you just need to switch the DCCD Pro to manual and dial up the amount of LSD Lock you need (which will be 100% more often than not I would guess). You will soon know if you have left it locked as it will bind up and let you know in which case you simply wind the lock off.

I had a look at the DCCD Pro map in automatic mode (for road/track use obviously) and I think it actually looks exactly like a throttle trace would (TPS voltage if you prefer). So for my road car I was going to simply make the TPS double up as the manual switch - no throttle = no LSD, full throttle = full lock.
I guess you could do something similar for offorad use but maybe make full lock come in at 50 or 75% throttle ??
[SIGPIC][/SIGPIC]

User avatar
MTB92
Junior Member
Posts: 204
Joined: Thu May 12, 2011 10:58 pm
Location: Koongamia WA

Post by MTB92 » Thu Sep 27, 2012 8:11 pm

Gannon wrote: its 65/36 open diff with no LSD'ness.

alright, lets run with this... a gearbox that increases HP :D
niterida wrote:I guess you could do something similar for offorad use but maybe make full lock come in at 50 or 75% throttle ??
I am not sure this would be that useful in practise. i think most of the time you need a locked centre is not at high throttle, but rather when you are crawling over rocks or lumpy stuff. i think it could be disastrous having it lock at more than half throttle in some cases.

User avatar
niterida
General Member
Posts: 1158
Joined: Sun Apr 13, 2008 5:45 pm
Location: Geraldton WA

Post by niterida » Thu Sep 27, 2012 8:40 pm

MTB92 wrote: I am not sure this would be that useful in practise. i think most of the time you need a locked centre is not at high throttle, but rather when you are crawling over rocks or lumpy stuff. i think it could be disastrous having it lock at more than half throttle in some cases.
hehehe I am not an offroader so don't have any experience in driving in those situations - maybe you could do it the other way then - make it 100% until 25% throttle and then down to 0% at 100% throttle ??
[SIGPIC][/SIGPIC]

User avatar
Gannon
Senior Member
Posts: 4580
Joined: Sat Nov 05, 2005 10:00 am
Location: Bowraville, Mid Nth Coast, NSW

Post by Gannon » Thu Sep 27, 2012 9:21 pm

niterida wrote:My understanding is that the torque split is always 65/35 no matter how much lock is dialled in. The 65/35 is derived from a mechanical planetary gearset with differing sized gears that still give the same output speed. This mechanical connection cannot be changed therefore the torque split cannot be changed.
The DCCD simply adjusts the time the LSD plates are locked via a PWM (pulse width modulator), or the LSD'ness as you put it :)
Yeah common misconception. If the centre was locked and gave a biased torque split, the front and rear wheels would have to be turning at different speeds, which would chew your tyres or twist your tailshaft pretty quick.

When the diff is open and driving on a straight road, the output shafts are forced to turn at the same speed because the front and rear wheels are turning at the same speed, but the planetary gearset is trying to turn the front wheels slightly faster than the rear.
Under acceleration in open mode, because the rear output shaft effectively has a lower reduction ratio because of the planetary gearset, it gets a higher share of the torque.
Imagine having an open centre diff and a 3.9:1 diff in the front and a 4.44:1 in the rear. The rear has more reduction, thus gets more torque, but the diff gears absorb the speed difference and there is no binding.

In locked mode, the clutch plates lock the front and rear output shafts and force them to turn at the same speed, like any other locking diff.
The clutch plates override the gearset and thus the output torque split is 50/50.

Imagine the above example with the 3.9 and 4.44 diffs and then run a belt from the front to the rear wheels on each side. The torque split from the planetary gears/different diff ratios is still the same bias, but because the locking force is on the outside of the bias, it creates a 50/50 torque split.
Its physically impossible to have a locked diff with a biased torque split AND have the front and rear wheels turn at the same speed.



I hope I have explained that right
Current rides: 2016 Mitsubishi Triton GLS & 2004 Forester X
Ongoing Project/Toy: 1987 RX Turbo EA82T, Speeduino ECU, Coil-pack ignition, 440cc Injectors, KONI adjustale front struts, Hybrid L Series/ Liberty AWD 5sp
Past rides: 92 L series turbo converted wagon, 83 Leone GL Sedan, 2004 Liberty GT Sedan & 2001 Outback
------------------------------------------

User avatar
niterida
General Member
Posts: 1158
Joined: Sun Apr 13, 2008 5:45 pm
Location: Geraldton WA

Post by niterida » Thu Sep 27, 2012 10:51 pm

Gannon wrote:Yeah common misconception. If the centre was locked and gave a biased torque split, the front and rear wheels would have to be turning at different speeds, which would chew your tyres or twist your tailshaft pretty quick.

When the diff is open and driving on a straight road, the output shafts are forced to turn at the same speed because the front and rear wheels are turning at the same speed while the planetary gearset is actually turning slowly within itself.

Under acceleration in open mode, because the rear output shaft effectively has a lower reduction ratio because of the planetary gearset, it gets a higher share of the torque.
Imagine having an open centre diff and a 3.9:1 diff in the front and a 4.44:1 in the rear. The rear has more reduction, thus gets more torque, but the diff gears absorb the speed difference and there is no binding.

In locked mode, the clutch plates lock the front and rear output shafts and force them to turn at the same speed, like any other locking diff.
The clutch plates override the gearset and thus the output torque split is 50/50.

Imagine the above example with the 3.9 and 4.44 diffs and then run a belt from the front to the rear wheels on each side. The torque split from the planetary gears/different diff ratios is still the same bias, but because the locking force is on the outside of the bias, it creates a 50/50 torque split.
Its physically impossible to have a locked diff with a biased torque split AND have the front and rear wheels turn at the same speed.



I hope I have explained that right
Totally wrong - but after a bit more research so am I :(
But lets not hijack Guys thread anymore :)
[SIGPIC][/SIGPIC]

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Thu Sep 27, 2012 11:38 pm

Gannon great explanation but is the idea Dave and I came up with a good one? To have the Dccd lock via a momentary switch and 30sec timer? Would that work for four wheel driving???
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
Gannon
Senior Member
Posts: 4580
Joined: Sat Nov 05, 2005 10:00 am
Location: Bowraville, Mid Nth Coast, NSW

Post by Gannon » Fri Sep 28, 2012 6:19 am

I didnt know about the overheating thing and I have no experience with the dccd pro controller, but im pretty sure with the factory system, ya just set in manual and wind it up to 100%. The later factory system also uses the ABS sensors to measure wheel speed too.

This thread is about learning, so I will continue...

Although this drawing is rather rough and shows bevel gears rather than planetary gears, the theory is the same
Image

The green side gear to the rear is larger than the red side gear to the front, which provides the rear biased torque split because it has a higher reduction ratio, with the blue spider gears providing torque from the engine. The clutch plates lock the red and green shafts together, cancelling the bias and providing the 50/50 split

Image
Attachments
DCCD tech ref.JPG
DCCD tech ref.JPG (88.78 KiB) Viewed 2213 times
VLSD2.JPG
VLSD2.JPG (24.59 KiB) Viewed 2218 times
Current rides: 2016 Mitsubishi Triton GLS & 2004 Forester X
Ongoing Project/Toy: 1987 RX Turbo EA82T, Speeduino ECU, Coil-pack ignition, 440cc Injectors, KONI adjustale front struts, Hybrid L Series/ Liberty AWD 5sp
Past rides: 92 L series turbo converted wagon, 83 Leone GL Sedan, 2004 Liberty GT Sedan & 2001 Outback
------------------------------------------

User avatar
guyph_01
Senior Member
Posts: 2515
Joined: Fri Jul 25, 2008 10:24 am
Location: Wilson WA

Post by guyph_01 » Fri Oct 05, 2012 12:26 am

Still not sure if the idea of the timer moment switch is a good one or not?

Got some work done on the car 2day, Replaced a torn rear cv boot with one of those hardcore heavy duty ones that Allan's cv joint supplies, Happy to report that since i started to use them 6-8months ago they are still holding and the car has been driving as a daily since. They used to rip open every 1-2 months because of the angle i've got.
Also replaced a torn steering rack boot:)

On the other hand, although the car isn't hunting/reving anymore, the check engine light still comes up while driving and goes away when i press the clutch and let the revs under 1 for a few secs. Pretty sure its related to that extra alt wire and high input to the MAF sensor. Also realised that when the check engine light is on, I i put my foot down, the car acceleration is sluggish. It doesn't happen when the engine light if off.

Might still need to look for the proper wiring diagram..
The project, EJ22 --->>> EJ25 Quad CAM:D touring wagon
Image
Thinking of going on holidays on a little paradise island, Check out http://www.dodolidays.com

User avatar
El_Freddo
Master Member
Posts: 12175
Joined: Tue Oct 04, 2005 10:00 am
Location: Benders (aka Bendigo) Vic

Post by El_Freddo » Fri Oct 05, 2012 5:17 pm

guyph_01 wrote:Still not sure if the idea of the timer moment switch is a good one or not?
Personally I say not. Just dial it up manually and if need be have a light on the dash that is your indicator to say that you're in manual mode.

Cheers

Bennie
"The lounge room is not a workshop..."
Image
El Freddo's Pics - El_Freddo's youtube

User avatar
Gannon
Senior Member
Posts: 4580
Joined: Sat Nov 05, 2005 10:00 am
Location: Bowraville, Mid Nth Coast, NSW

Post by Gannon » Sun Oct 07, 2012 10:26 am

According to the technicians reference for the dccd, it requires a PMW signal of between 300Hz and 2kHz at a maximum current of 4 amps.

So what you need to do is get a PMW controller that works in that frequency and adjust its maximum to limit current to 4A. It should read about 7V on a multimeter in this condition.

Although the frequency is a little high, something like this might do the trick
PWM DC Motor Speed Control Switch 12V-40V 10A Pulse Width Modulation 13khz
Current rides: 2016 Mitsubishi Triton GLS & 2004 Forester X
Ongoing Project/Toy: 1987 RX Turbo EA82T, Speeduino ECU, Coil-pack ignition, 440cc Injectors, KONI adjustale front struts, Hybrid L Series/ Liberty AWD 5sp
Past rides: 92 L series turbo converted wagon, 83 Leone GL Sedan, 2004 Liberty GT Sedan & 2001 Outback
------------------------------------------

Post Reply

Return to “Guyph”