Member Profile- 2007 3.0r-b sedan 65Z 6MT

Show off what mods you've done to your car.

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Mon May 16, 2022 11:47 pm

Starter Motor Rebuild

Over the previous few months I had noticed the car was sounding more and more lazy while cranking. Doing some basic electrical troubleshooting there was zero issues I found with the charging system, and the battery is less than 2 years old so I doubted it was this.

Image

A teardown of the starter motor revealed the brush pack was incredibly worn- to be expected really after 15 years and 255k kms! The brushes were so worn they were at the limit of their springs and not making consistent contact with the commutator, so I added a washer in between each spring and brush to get a bit more brush force while waiting for a replacement. This did help, but isn’t a long term solution. Thankfully, although dirty, the commutator itself was in good condition, with minimal scoring or wear, and all the starter bearings were likewise in good condition. Interestingly, it was the positive brushes that were significantly more worn than the ground brushes.

Image

Credit to BillyCorgi for the information, where it was revealed in this thread (http://forum.liberty.asn.au/viewtopic.php?f=16&t=32776#p437320) that the EZ30 starter motor shares its brush pack with quite a few vehicles, and an example listing was shared! With this information I was able to determine the original Denso part number, which is 128099-6310, and this meant I could purchase just the brush assembly for about $25, instead of the ‘Cover AY-Starter’ (23337AA10) for over $130!

Image

As the EZ30 starter motor uses a face commutator, rather than the more common radial commutator (like what’s normally used in a universal motor), it is very easy to replace the bushes! This style of commutator doesn’t require the brushes to be held out of the way while assembling the motor, so the old ones can be removed by undoing the 3 phillips head screws that hold the brush assembly in, removing the old brush assembly and replacing it with the new one! Two of these screws are also the ground for the brush assembly, so the end cover was cleaned properly in this location to minimise contact resistance.

I also took the time here to clean off any old grease and re-grease every surface that required it.

Image

With the new brush pack installed and every cleaned and re-greased, the starter motor was reassembled. The only slightly difficult bit here is lining up the plate between the motor itself and the gear set; this is a bit annoying as the permanent magnets are attracted to the metal plate, but I found a thin plastic trim removal tool helped get everything lined up out of the influence of the strong magnet.

The bolts that clamp the starter motor assembly together from the outside are also the main path to ground for the brushes, so make sure both the heads and threads of these bolts are clean, and ensure the surfaces they touch are clean too!

All of this effort got me precisely one start. After that one start I would just get clicking from the starter solenoid, but no power was transferred to the motor itself. I assume the higher inrush current of the repaired starter motor was too much for the old solenoid. Checking the price of new starter solenoids, as well as the lead time on new parts, made me have a closer look at rebuilding the worn out one.

A basic resistance check showed the solenoid coil was still fine, and as I had just rebuilt the rest of the starter assembly it was not the pivot or plunger getting stuck in old grease that was causing the problem. The starter solenoid is built in a very similar way to the previously torn down AVCS solenoids, which is what gave me the idea to try and take it apart and either clean or repair the high current contacts inside.

With the solenoid removed from the starter motor, a flat blade screwdriver can be used to un-crimp the outer case, and the end with the copper high current lugs simply pulls out. This revealed some very worn out contacts, and the copper bridging piece had significant arc pitting as well.

Image

Unfortunately I couldn’t find the bridging piece as a separate part, and it could not obviously be disassembled to replace. I cleaned it up as much as possible using a rotary tool with an abrasive buffing wheel, being careful to not remove too much material which would impact the current carrying capacity of this part.

Image

I also could not find replacement contacts in a timely manner, but inspection of the contacts inside the solenoid showed they were part of the lugs on the rear, and they only showed wear on half of their surface. These lugs are held into the plastic housing using a thin piece of spring steel as a nut, which can be un-threaded using a pair of pliers. After cleaning up the surface using an abrasive buffing wheel, the lugs were reinstalled, but rotated 180 degrees. This exposed their undamaged surface to the copper bridging piece, and should hopefully significantly increase the life of the solenoid!

Image

Image

The solenoid could then be reassembled and re-crimped, just like the AVCS solenoids above, and then reassembled onto the starter motor, with the plunger and pivot assemble re-greased for good measure.

Image

This work got me my starter motor back, and in combination with the wiring upgrades means the car starts better than ever! It cranks over much faster, and now tends to fire on the first or second rotation every time. I also noted a massive increase in the peak cranking current, jumping up from barely 190A up to a much more respectable 396A.

Image
Last edited by JezzaH6 on Wed May 25, 2022 10:06 pm, edited 1 time in total.
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby Yowie » Tue May 17, 2022 10:06 pm

Bl00dy good work. Cheers for the photo walk-through with explanations.

A few extra possibilities to keep in mind rather than just assuming "sad battery" when things are slow to crank.
User avatar
Yowie
 
Posts: 574
Joined: Tue Jul 21, 2015 10:05 pm
Location: Brisbane
Car: SH Forester XT

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Fri May 20, 2022 2:36 pm

Yowie wrote:Bl00dy good work. Cheers for the photo walk-through with explanations.

A few extra possibilities to keep in mind rather than just assuming "sad battery" when things are slow to crank.


Cheers! It's all well and good to assume it's just the battery, but as these cars get older and things start getting worn out it's worth the extra effort to find out why things aren't working to their fullest extent.

None of these things were particularly difficult to diagnose, and the battery can be ruled out by simply watching a multimeter as someone cranks the car over. If the voltage sags significantly below 9V then likely it is the battery, but if the voltage stays higher then it's time to dig a little deeper. it's nice having tools that datalog, but they definitely aren't necessary for a system as simple as the starting circuit.

A few simple checks helps avoid just firing the parts cannon at these older cars!
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Thu Jul 14, 2022 10:37 pm

AP Racing Big Brakes

To be completely honest, on both street and track, I was really happy with my previous brake setup. Consisting of Cadillac CTSV 4 piston Brembo calipers, Forza FP3 brake pads, Vmax ‘J’ type 326m rotors and braided lines at the front, with the stock Spec B rear calipers paired with Intima SR pads and 296mm Vmax ‘J’ type rotors and braided lines at the rear, the only thing I swapped out for track use was the front pads; I could get about 2 hot laps from the Intima's and 3 hot laps from the Forza’s, but swapping to Winmax W7’s just at the front for track use got me unlimited hot laps (at least until something else overheats).

Image

Even with this in mind, however, I was still left wanting more. With the upgraded master cylinder the pedal feel was *fine*, but driving friends cars with fixed piston calipers on all 4 corners I knew rear Brembo’s were definitely on the cards. I had actually come to this conclusion quite a while ago, but the lack of people prepared to part with just the rear Brembo’s made this hard to actually implement. The aesthetic benefits of a proper BBK cannot be understated in my opinion, but the real reason for wanting more was to experiment with potentially less-aggressive pads for daily driving in the colder months, with all the drivability benefits that comes with this, but still having something that can cope with elevated temperatures and still have access to a decent catalogue of more aggressive options for both track and summer use.

I managed to get my hands on a set of 4 AP Racing calipers, rotors and pads as new, open box stock, for an almost unbelievable price. A check of the serial numbers on the calipers showed they were, in fact, genuine, and for only a bit more than the cost of a set of used front and rear STi Brembo's I jumped at the chance! The fronts are CP9040 6 piston calipers, featuring 27mm, 31.8mm and 38.1mm pistons, and the rears are CP5200 4 piston calipers, featuring 38.1mm and 41.3mm pistons. Both calipers are of two-piece bolted forged aluminium construction.

Image

The fronts are paired with two piece 355mm x 32mm rotors with aluminium centre hats, and the rears are 345mm x 28mm, with steel centre hats for compatibility with the STi 190mm handbrake assembly.

The fronts are clearly a kit sold by ProSpeed Racing (https://www.prospeedracing.com.au/AP-Ra ... /Toyota-86), but the rears are a bit odd. For starters, the CP5200 calipers are only *officially* compatible with up to 330mmx28mm rotors, which means these 345mm rotors are not going to have optimum pad contact, with a benefit of increased brake torque and more cooling. I could not find a kit anywhere for this caliper and rotor combination, and the poor (read: non existent) fitment on my Gen 4 Liberty makes me think these were not actually made for a Subaru (but the inclusion of the handbrake drum means I have no idea). I assume they are a bespoke creation that never saw fitment.

Image
AP Racing CP3215D50 pads (rear) on the left, stock Spec B rear pads on the right!

I really liked the red finish on these, but given I have put a lot of effort into trying to keep this car as subtle as possible I couldn’t leave them that colour. After a lot of deliberation, and wanting something different to my old black brake setup, I decided to go with a bright silver. This colour, I hoped, would blend in nicely with the rotor surface, while still being somewhat visible under closer inspection.

Image

The stock AP Racing finish is paint, rather than powder-coat, which makes changing the colour of the calipers somewhat straightforward. After scuffing up the existing paint and masking off all non-painted surfaces the calipers were painted using many thin coats of DupliColour silver caliper paint.

After giving the silver paint a good month or so to allow any solvents to fully flash off, labels were added to the front calipers with a set of black high temperature vinyl. The rear calipers, with their indented letters, was a different story entirely. I tried many different techniques to mask and paint these letters, but none worked satisfactorily.

Image

To paint these recessed letters I ended up roughly masking the rest of the caliper, and then spraying black caliper paint over and into the letter recesses. This would inevitably end up getting paint on non-letter parts of the calipers, but after removing the masking and allowing the paint to flash off for 5 minutes the excess paint was removed with paint thinners on a very absorbent rag. This was repeated for multiple coats until a decent finish was reached.

Image

After giving this black paint a few weeks to completely flash off, the front face of the calipers was polished to remove any remaining black paint residue.

Image

After seeing the single-pack clear coat I applied to my old Brembo calipers get destroyed in a very short time from both brake fluid and UV yellowing I decided to clear coat these new calipers with 2 pack epoxy paint. Without access to a suitable spray gun I tried out 2 part paint-in-a-can, which worked really well! Whilst I wouldn’t use this to respray a car, for brake calipers the finish is definitely more than good enough, and the chemical and UV resistance of 2 part paint totally justifies the extra cost. The semi-gloss paint I used catches the light nicely, without being attention-grabbing.

Image

The front caliper fitment required new front hub assemblies, as my old ones were modified to fit the CTSV Brembo’s in such a way that they wouldn’t work with the AP Racing calipers. I got a set of front hubs from a local wrecking yard, and given everything would have to be taken apart anyway I replaced everything (new wheel bearings and new mounting hardware everywhere), and I chose to replace the (less than 10k km old) Whiteline bump steer and roll centre correction ball joints and tie-rod ends, as these were already showing significant signs of wear- these I replaced with some much higher quality HardRace products. At this point in time I have removed every Whiteline product on my car as none of them have lasted more than 15k kms, and I wouldn’t recommend their products to my worst enemy, but I digress.

Image

Front caliper fitment was otherwise straightforward, with the new calipers being compatible with the stock banjos, and the adaptor that came with the calipers spaced them out perfectly- although with not much clearance to my 18” wheels! The front rotor dust shield had already been modified when I installed the Brembo's, so I didn’t need to touch these this time.

Image

Fitment of the rear calipers was a lot more challenging. I chose to replace the rear wheel bearings at the same time, as well as the rear differential (more on this in another post), as well as all the mounting hardware for everything that bolts to the rear arms.

Image

The centre hats that came with the rear rotors suits the STi 190mm handbrake assembly, whereas my Spec B had the 170mm handbrake. Rather than go to the effort of a full hand brake assembly conversion, I got a set of WRX to STi conversion shoes. These do not, however, fit the Liberty handbrake, and I learnt (after purchase of course) that shoes do also exist to convert the Liberty handbrake without modification (but these are a lot less common, and a lot more expensive). The shoes only require their pivot point hole to be moved, which was as simple as holding the original handbrake shoe over the new one and re-drilling the hole. There is a lot of movement built into the design of the handbrake drum, so these holes didn't have to be particularly precise.

Image

The mounting bracket that came with the rear calipers did not fit on my car, and after cutting part of the bracket away to allow it to clear the hub it would not hold the caliper in the correct spot. A quick google seems to suggest this bracket may work on the GR WRX, but I am not quite sure.

Image

To get these calipers working on my car I ended up designing new brackets after taking a bunch of measurements. A couple of 3D printed test brackets allowed me to perfectly dial in the fitment, and ensured as good as possible pad and rotor coverage. Some quick maths showed that the PLA brackets would actually be barely strong enough at the expected caliper loading (with a safety factor of 1.06!), but I got the brackets CNC manufactured from billet 6061, which the maths (and basic simulations) showed would give me a safety factor of about 16.5 which is much more comfortable!

Image

I ended up using the services of an online CNC manufacture as I don’t have access to one at home. The two brackets arrived 10 business days after I uploaded the file, and only cost me about $90aud per bracket, shipped, which I am quite impressed with.

Image

The quality of the brackets is fantastic; I opted for a brushed finish; and were dimensionally perfect to the accuracy I am able to measure at home. I only had to thread the mounting holes before I could mount the calipers (I chose not to have the CNC machine do this as it is a time consuming, and therefore expensive, process, and I had the correct taps in my toolbox anyway).

While taking everything apart for the first time I noticed a damaged rear brake line. I fitted these brake lines from HEL Performance in 2019, and in that time the rear calipers had never been removed and the lines never clamped. There was measurable ballooning evident at the damage site, and it looks like possibly an internal fault in the line which caused the damage to the outer coating. Thankfully HEL’s customer service is fantastic and a replacement line was on it’s way to me within a day of initially contacting them.

Image
With all the parts finally lined up and sorted the rear caliper install went smoothly as well- after a lot of preparation work anyway. The modified handbrake shoes, with a new fitting kit, work perfectly, and the custom brackets hold the calipers in the perfect position. The rear dust shields did have to be removed to fit the extra 55mm of brake rotor; this was easily done by using a thin power file to grind away the spot welds. Both front and rear calipers and brackets were mounted up with strong semi-permanent thread lock.

Image

To the surprise of probably not too many people, this setup was significantly worse than the one it replaced. I was expecting the standard AP Racing pads to be not as good as my tried-and-true setup, but I was also worried about the impact on the brake bias that this larger setup would have after having done some basic calculations for piston and pad area etc.

These concerns would be well founded, as I immediately noticed the rear wheels would lock up really easily. For day to day driving with light braking this wasn’t noticeable, but as soon as I had to brake harder it would see the rear wheels get the ABS treatment almost immediately. This could be solved to some degree by locking the centre differential; this forced the front and rear axles to turn at the same speed and eliminating the effects of brake bias, but this is not a solution for either daily driving or when looking for performance, as the locked centre diff causes drive train windup on grippy surfaces and makes the car handle worse.

With the centre diff set to anything other than locked, braking hard into a corner would immediately see the rear wheels lock up, resulting in the ABS pump activating, and upsetting the balance of the car causing oversteer. With the centre differential locked up the brakes felt amazing; the car pulls up crazy hard and the pedal feel and feedback was nothing like anything I had experienced before. However, the locked centre differential causes the car to scrub through the corners and in most circumstances would see the onset of understeer. Worth noting I could reprogram the DCCD controller to react better in this situation, but given the time I had put into programming it previously and how happy I was with that setup I didn’t really want to put a bandaid over the problem.

The solution to this decidedly first world problem was the installation of a set of brake proportioning valves (one for each rear left and right caliper to suit the 4 channel ABS). Mounted after the ABS pump, the two valves are plumbed into the hard lines that go to the rear caliper. A custom hybrid aluminium and 3D printed bracket holds the two valves next to each other right in front of the master cylinder.

Image

After the mess that was the custom hard lines for the master cylinder swap I decided to go with custom braided lines to and from the valves. A high quality braided line should have no real downside compared to a hard line, and allows for direct connections without any adaptors or additional fittings.

Image

Proportioning valves are commonly fitted from the factory to older vehicles, but it is a common modification in the racing community to get braking performance dialled in perfectly. They function as a pressure regulating valve, allowing the brake line pressure to pass through the valve until a certain pressure is reached, where after a set point the valve causes a slower pressure build up than the line pressure build up. This behaviour ensures good day-to-day driveability by not adjusting the brake line pressure under day-to-day braking conditions, while allowing for a reduction in rear caliper pressure (and thus reducing the chance of the rear wheels locking up) under heavier braking conditions.

Image

A system like this is normally fairly challenging to set up correctly, especially as the way I have this set up also allows for left-right pressure adjustment which is very dangerous if setup incorrectly. Thankfully a friend got me onto the Safe-T-Stop system at the workshop he works at, which is essentially a simple brake dyno. This system gives a readout of the braking force at each wheel, and allowed me to very accurately set up the valves for optimum stopping power.

Image
The first test after an initial setup. This test shows the left-right bias at the front and back on the top diagram, and the front-rear bias on the bottom diagram, with the actual braking force in newtons written inside the ‘wheels’ of the top diagram. I was not happy with my previous alignment that’s for sure!

With the valves ‘off’ (ie. not doing anything) the bias was close to 55%F-45%R, which explains why the car was so undriveable. Over the course of about 15 minutes we were able to get the bias to 65%F-35%R. This is a little more rear biased than normal for a street car, but this will ensure the large rear brakes will still be doing their fair share of work. I have now reprogrammed the DCCD controller to lock up harder under braking, without completely locking up the centre diff, and the factory ABS-based proportioning system works well enough, even before the addition of the proportioning valves, that it can still be relied upon to be the safety net that it’s supposed to be.

I am still not happy with the outright performance of the included AP Racing APF404 pads- these just don’t have the hot performance of the Intima’s or the Forza’s, but for a daily-driver focused pad they are very quiet, very clean (not dusty), have very good cold performance and feedback like nothing I have ever experienced before. What they do allow me to feel, however, is the potential that this setup offers and I can’t wait until better pads (and better weather) arrives so I can push these things harder!

Image
Last edited by JezzaH6 on Fri Nov 24, 2023 4:36 pm, edited 5 times in total.
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby Yowie » Sat Jul 16, 2022 7:27 pm

Great write-up. If it's not Post of the Year it'll be in the finals.

That mail-order CNC work is pretty handy.
User avatar
Yowie
 
Posts: 574
Joined: Tue Jul 21, 2015 10:05 pm
Location: Brisbane
Car: SH Forester XT

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Mon Jul 18, 2022 6:59 pm

Yowie wrote:Great write-up. If it's not Post of the Year it'll be in the finals.

That mail-order CNC work is pretty handy.


Haha thanks mate! I basically use this space to keep a good reference of what I've done, and hopefully something in the wordy mess helps people as much as other on here have helped me :lol:

It's definitely a useful thing to know exists!
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Wed Aug 10, 2022 10:40 pm

Caliper Comparison

Front calipers- AP Racing CP9040

Given the forged aluminium construction these calipers are really light! According to my bathroom scales, without pads or mounting hardware the CP9040 calipers weigh 3.5kg, which compares really favourably to to 4.75kg of the CTSV Brembos, or the 6kg of the stock 2 pot calipers. Even with pads and mounting brackets the CP9040 calipers are less than 5kg- this reduction in unsprung mass is not insignificant!

Image

Image

Front rotors- 2 piece 355mm * 32mm, aluminium centre hats

As two piece rotors the new ones are obviously significantly lighter than the single piece steel rotors they replaced, which makes a comparison slightly unfair. Regardless, the new front rotors are featherweights at 7.9kg, compared to the single piece 326mm * 30mm Brembo rotors at 11kg, or the stock single piece 316mm * 30mm at 9.5kg.

Image

Image

The CP9040 setup seems to be all about weight savings, as the pad area is not significantly larger than the Brembo pad. For such a large diameter rotor this means the ring itself is fairly small, and that with a large aluminium centre hat helps keep the weight down. For a 6 piston caliper they are relatively small too; definitely smaller than the ShockWork’s 6 piston BBK.

This gives a total weight of 15.5kg per side for the stock 2 pot 316mm brakes (single piece rotors), 15.75kg for the CTSV Brembo’s (single piece rotors), and 11.4kg for the AP Racing calipers (minus pads for all calipers)- a saving of almost 4.5kg of unsprung mass per front corner which is quite significant (almost 30% less brake mass compared to my previous setup!). This is definitely noticeable even for day-to-day driving, with noticeably better suspension and steering response and better ride quality.

Rear calipers- AP Racing CP5200

These calipers are also forged aluminium and are very light at only 2.2kg without pads or mounting hardware! I was never able to get my hands on a set of rear tbSTi Brembo’s, so I can’t comment on their weight, but the stock single piston calipers weigh in at 2.95kg.

Image

Image

Rear rotors- 2 piece 345mm * 28mm, steel centre hats

The rear rotors are the only area where the new brakes are heavier than the ones they are replacing due to the necessity of the steel handbrake drum lining and the much larger size compared to the stock 290mm x 18mm rear disks. The new rotors come in at 9.75kg, compared to the original rear rotors at 6kg.

Image

Image

This gives a new brake mass of 11.95kg, compared to 8.95kg of the stock rear brakes, for an increase in brake mass of 3kg. As this is the rear end of the car the effect of this is less significant than the front, but it is still not ideal, especially as this extra mass is in unsprung, rotational mass. An upgrade to the latest ShockWork’s dampening and spring rate revision, however, has made a huge difference in controlling this extra mass!
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby nvmylh » Thu Aug 11, 2022 8:35 am

Those rear rotors look nuts compared to the stock!

Are those handbrake conversion pads simply a stock pad with 10mm of material added on top? Hope the heat does not affect them on the track! (though you're running around a track with the handbrake on...).

Hows your engine going these days? Looking at building another any time soon?
nvmylh
 
Posts: 141
Joined: Thu Apr 15, 2010 1:15 pm
Location: Melbourne
Car: '08-H6-Wagon

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Fri Aug 12, 2022 2:15 pm

nvmylh wrote:Those rear rotors look nuts compared to the stock!

Are those handbrake conversion pads simply a stock pad with 10mm of material added on top? Hope the heat does not affect them on the track! (though you're running around a track with the handbrake on...).

Hows your engine going these days? Looking at building another any time soon?


The rears especially are quite an upgrade over stock- or even over the Brembos! The extra pad area, as well as a much thicker and better ventilated rotor, is going to make a huge difference to brake temps when I can get some good pads in them.

The handbrake pads are the stock wrx liners, but with thicker material (it's not just the extra material glued on top). They are then shaped to closely match the 190mm diameter of the larger handbrake drum. I don't think the heat will affect this as it is thicker material, and not just a 'spacer' glued on top of a normal handbrake, but time will tell I suppose!

Engine is going great! Minimal oil consumption, no coolant consumption, good iam and good avcs and avls operation which is honestly surprising given its now exceeded 260k kms, over 120k of which has been hard driving! At this stage I don't really have plans of building something too crazy, but that itch is always there :lol:
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Mon Aug 15, 2022 11:55 pm

Quaife QDH2Y Install

Over time I had been noticing the rear end of my car behaving worse and worse while trying to corner under power. As a 3.0R-B my car came from the factory with a viscous limited slip differential which, although definitely better than an open diff, is prone to wearing out. This would have been accelerated after the installation of the DCCD-equipped JDM 6mt which routinely sends more torque to the rear diff than the factory gearbox, making it more likely for the rear wheels to loose traction and putting more load on the viscous coupling.

Image

Trying to power out of corners on even slightly slippery surfaces would see the car trying to light up one of the rear wheels causing the rear of the car to try and step out, and it was very obvious that the fluid in the viscous LSD was worn out, causing the rear diff to behave no different to an open differential.

To be expected I suppose after over 260k kms of driving and quite a few track days at this point.

There’s a few different options for better LSD’s in Subaru’s, but the majority of the offerings are of the clutch-type. These have the ability to totally lock the rear wheels together, which is a very good thing on extremely loose surfaces such as rally, but they have a few downsides. The main one, especially for a car that is daily driven, is that they tend to be louder while operating, and their low speed driveability is normally quite poor, with clunking and drivetrain windup occurring at low speeds. They also have clutches that wear out, and as they wear they contaminate their own fluid, meaning even more maintenance is required.

This is offset (to the right person) by their awesome adjustability, where they can be tuned to react exactly how you need them to. For a dedicated track car this totally outweighs the downsides, but for a mostly street driven car I didn’t want to put up with this.

I decided on a Quaife Automatic Torque Biasing (ATB) differential, which is a geared LSD. Under normal driving conditions these function as an open differential, with nothing preventing the rear wheels from rotating at different speeds. This ensures good low speed and day-to-day driveability, without it being noisy and with no driveline windup.

Image

As soon as the torque delivered to one wheel is different, however, the mechanism inside creates a torque bias favouring the wheel with greater torque requirements (greater traction), preventing the slipping wheel from spinning and ensuring good torque delivery even on slippery surfaces. The benefit here is that the torque-biasing happens independently to the differential functionality, and the two can happen simultaneously. This means the rear wheels can rotate at different speeds but with equal torque transmitted to the road, giving really good performance without risk of spinning a tyre and without tyre scrub under power on higher traction surfaces.

Aside from the cost the main downside to this style of LSD is that if one wheel is lifted there is an ‘infinite’ torque bias to the lifted wheel, which the ATB diff cannot overcome. Since I am confident in the suspension setup on the car that it won’t lift a wheel (no stupidly large swaybars here…) this is very unlikely to be an issue.

Image

I wanted to stick with the R160 rear differential so as to not have to muck around with rear hub conversions or expensive adaptor axles, but the only R160 compatible Quaife lists is their QDH2Y, which lists 1993-2000 Impreza/Legacy R160 compatibility. Thankfully I have a 99 Impreza, which let me check the axle splines for compatibility. A check of the technical drawings as well showed it *should* fit in the later model rear differential housing.

I got a replacement rear differential assembly from another member so I could still drive the car while going through the majority of the setup and install process. I replaced all the bearings and seals at the same time, but as I was very time-poor at the time I ordered this, and couldn’t find a reliable source for shims, I chose to get a local shop to assemble and shim the differential.

I won’t name the shop as ultimately they made it right, but this was a massive mistake. From supplying all the bearings, seals, torn down and cleaned differential housing and the Quaife itself, it took almost 7 months to get it back! It took multiple phone calls and many emails to determine that, although I was reassured it was possible to be done in-house, everything was shipped interstate to ADS. ADS did a great job, but the time it took for shipping during covid, and them messing up the order details multiple times meant the whole process took a heck of a lot longer than it should have.

Image

Aside from skimming the heads when I rebuilt the top end this is the only other thing I have had a shop do to this car since I started working on it, and this whole experience reminded me of why I’ve tried to learn as much as I have.

I also got a nice billet high capacity rear differential cover, but unfortunately this didn’t end up clearing my exhaust, and I ended up running the stock rear cover. I'm holding onto it though as a new exhaust is in the works!

Image

The only thing I had to do after receiving the assembled differential assembly was transfer the circlips from the old diff to the Quaife. It was decidedly not fun to remove these from the old diff, and even less fun to get them in the Quaife, but some long reach pliers with a grove cut in the tips let me get these in position.

Image

Since the rear shafts had to be removed to do the rear wheel bearings when I did the rear brakes, I did the differential swap at the same time. This was a lot to get done in a day, but an early start and late finish made it doable! Since the differential was preassembled in its housing it was a straightforward process to remove the old rear diff and replace it with the new one.

Image

Since the Quaife doesn’t use clutches, it doesn't require an LSD-specific oil. It was run-in on high quality Castrol gear oil, but after the initial two oil changes the remainder of the run in process, as well as its actual running oil, was changed to Motul Gear 300 to help this quite expensive component last as long as possible!

With the helical front LSD, DCCD and the ATB rear LSD the on-power feel of this car is like nothing I have ever experienced before. At even quite extreme steering angles it is possible to be at or close to WOT on or before the apex with zero wheelspin, no scrub and no sign of over or understeer- the car just grips. While I have described the front LSD and DCCD combination in the past as allowing the car to be driven like a AWD, rather than a FWD, the ATB rear differential takes this to another level entirely. It is incredibly confidence inspiring on even the most inconsistent surfaces, and the level of control this setup offers is nothing short of incredible- it really is something that needs to be experienced to be believed. I can’t wait to try it out on track (or even just in dryer weather- thanks Melbourne winters)!

My AD08r’s had reached the end of their useful life at this point, being heat-cycled out rather than just worn out. After having a few sets of these now (and absolutely loving them), I was also wanting to try out something new. Being impressed with the AR1’s on track, as well as on recommendations by mates, I got a set of Nangkang NS2R’s in 235/40/R18, 180 tread wear.

Image

On the few dry days I’ve had time to get out to the hills, after a slightly longer warm up time these seem to perform about as well as the AD08r’s in the dry, and they do seem to have ever so slightly better turn in response as well. They are worse in the wet, however.

Where AD08s blew me away was performing way better than you’d expect a semi-slick to perform in the wet, the NS2R’s stay true to the stereotype that semi’s aren’t great wet weather tyres. They aren’t as bad as other semi slicks I’ve tried (Federal 595RS-RR’s are amongst the worst performing wet weather tyres I’ve ever experienced), and are still better in the wet than the ‘eco’ tyres that a lot of my normie mates have on their cars. With AWD they are certainly daily-able, and the savings vs AD08s are decent enough that I can live with the lower wet performance. Over about 2000kms so far they seem to be wearing really well for daily driving!

Image
Last edited by JezzaH6 on Tue Aug 30, 2022 2:57 pm, edited 1 time in total.
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby Ric » Thu Aug 18, 2022 2:40 pm

Great read! :)

This caught my attention.
JezzaH6 wrote:The only thing I had to do after receiving the assembled differential assembly was transfer the circlips from the old diff to the Quaife. It was decidedly not fun to remove these from the old diff, and even less fun to get them in the Quaife, but some long reach pliers with a grove cut in the tips let me get these in position.

The circlips failed in my diff, and Subaru told me they couldn't be replaced without rebuilding the diff, so I ended up getting a good second hand replacement.
I've since sold my 3RB, but I still have the original diff, and believe the only thing wrong with it is the failed circlips...
User avatar
Ric
Site Admin
 
Posts: 6429
Joined: Fri Oct 06, 2006 11:08 pm
Location: Melbourne
Car: OLD: MY06 3RB Wagon NEW: MY21 Outbck
Real name: Ric
Profile URL: viewtopic.php?f=7&t=12

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Thu Aug 18, 2022 10:44 pm

Ric wrote:Great read! :)

This caught my attention.
JezzaH6 wrote:The only thing I had to do after receiving the assembled differential assembly was transfer the circlips from the old diff to the Quaife. It was decidedly not fun to remove these from the old diff, and even less fun to get them in the Quaife, but some long reach pliers with a grove cut in the tips let me get these in position.

The circlips failed in my diff, and Subaru told me they couldn't be replaced without rebuilding the diff, so I ended up getting a good second hand replacement.
I've since sold my 3RB, but I still have the original diff, and believe the only thing wrong with it is the failed circlips...


Interesting. I wasn't able to get my hands on a set of new ones in time for this project, so I do hope these re-used ones hold up. I can't really see them having that much load on them and can't really work out what would make them fail- do you know what caused yours to fail?

Getting the circlips in was not fun. To get them in I ended up crushing them as little as possible so they fit in the shaft bore, and then expanding them again once they were in position. I cut a thin groove right at the tip of some long handled pliers to use to expand the circlip in its bore.

Image

Image

A few people claimed they could be mounted on the shafts, which could then be used as a slide hammer to get the circlips in place, but this didn't work for me at all (might work for the stock diff though). The circlip is 805324010, but I have not been able to find these in stock anywhere.
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby Ric » Thu Aug 18, 2022 11:19 pm

JezzaH6 wrote:...and can't really work out what would make them fail- do you know what caused yours to fail?

Nope.
It did happen while I was towing a car trailer, but that shouldn't put any load on the circlip.
Just the passenger side one failed, so the half shaft started popping out of the diff and just hanging on by the rubber boot. I only had a small amount of drive due to the limited slip operating.
(I had been competing in my MGA at Sandown, and left early because I broke the MG's gearbox. Had just left the track, and was turning on to Dandenong Rd when the shaft popped out the very first time. Was not a good day!)
User avatar
Ric
Site Admin
 
Posts: 6429
Joined: Fri Oct 06, 2006 11:08 pm
Location: Melbourne
Car: OLD: MY06 3RB Wagon NEW: MY21 Outbck
Real name: Ric
Profile URL: viewtopic.php?f=7&t=12

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Fri Aug 19, 2022 2:17 pm

Ric wrote:
JezzaH6 wrote:...and can't really work out what would make them fail- do you know what caused yours to fail?

Nope.
It did happen while I was towing a car trailer, but that shouldn't put any load on the circlip.
Just the passenger side one failed, so the half shaft started popping out of the diff and just hanging on by the rubber boot. I only had a small amount of drive due to the limited slip operating.
(I had been competing in my MGA at Sandown, and left early because I broke the MG's gearbox. Had just left the track, and was turning on to Dandenong Rd when the shaft popped out the very first time. Was not a good day!)


That's really annoying. After seeing how they work I'm surprised that even if they do fail the shaft can come out. The clip really didn't want to come out of the old gearbox, and I can't see that job being much easier even if the clip was in multiple pieces :pardon: maybe just an unfortunate bit of bad luck for you. Do you know if the clip is still in your diff or did it come out completely? Since you can get the clip as a separate part I can't see why a competent workshop couldn't replace them (especially as I was able to in my garage without mechanics training lol).
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

Re: Member Profile- 2007 3.0r-b sedan 65Z 6MT

Postby JezzaH6 » Sun Aug 28, 2022 2:21 pm

It's got its own thread in the 3.0R &3.6R Engine Specific thread, but I am running an oil cooler on my car! This is definitely not something for every EZ owner to do, but if the data you've gathered suggests it's a good idea I am really impressed with how my setup has been going so far!

I have been running this cooler now for over 18 months (yes my posting timeline is all over the place- I just write things up when I have the time!) and over 30k kms now. It has been on the car for about 5 track days, and even on the hottest of days pushing the car as hard as possible I have never seen the oil get above 126c- this is a major win considering the size of the cooler! In combination with the 42mm triple core radiator I can now do unlimited hot laps if the ambient temperature is below 30c, at least until brake or tyre fade, and on the street I have absolutely zero issues with engine cooling.

Image

Possum Bourne Motorsport's EZ30 baffled sump plate has also been in the car for about 14 months. Once again this is not something for every EZ30 owner, but I would argue if you are constantly pushing one of these cars hard enough to need an oil cooler your peace of mind would likely benefit from a setup like this as well!

Image

Check out the thread for more details on how I made the oil cooling system: http://forum.liberty.asn.au/viewtopic.php?f=16&t=37220
2007 Subaru Liberty 3.0r-b 6MT
User avatar
JezzaH6
 
Posts: 104
Joined: Thu Jun 11, 2020 6:57 pm
Location: Melbourne
Car: 2007 3.0r-b 6MT
Real name: Jeremy
Profile URL: viewtopic.php?f=10&t=36687&p=450058#p450058

PreviousNext

Return to Post your mods

Who is online

Users browsing this forum: No registered users and 9 guests