VIP GT-BBB

[Yowie]

Sounds like quite the epic.

If you like a VF46 and you're considering your options, I'm happy to give further info on my ported VF46 that has just come out. It has more shaft play than a person would want but it still ran fine and was driven with mechanical sympathy (noting that small turbos have a harder life). If you are spending money on a VF46 rebuild, a ported one could be a good start. No obligation obviously.

Checks upper rear right of block looking for oil leak: it’s the sump breather hose, one of the two primary *block* pcv hoses.
Replaced the cracked pcv pipe with a diy suction hose made from heater hose with a stretched compression spring inside.

Intriguingly this pipe does not appear on partsouq, opposed forces diagrams.
I did find it in the Baja diagrams, in the FSM, and in dealership diagrams that use Subaru FAST EPC. Weird.

Was it a 19mm ID "slightly Z-shaped" hose off the rear crank breather?

Mine (2008) had this but does not show up on partsouq (straight 19mm hose piece to plastic T shown) and even ordering that older style crank breather assembly gets you the new 2009+ "F shaped rubber moulding" based system for $160+ (no thanks).

Because I'm running catch cans I ended up grabbing an Aeroflow (?) silicone right-angle reducer (19mm to 13mm) to come off the crank breather, up over the gearbox (next to the turbo) then (with 1/2" barbed straight joiner) off to the rear catch can.

[bigBADbenny]

Yep that’s the one, the original part is still available, I was just too impatient to order it.

I’ll add it to the aos cc post as well.

What was really epic was the amount of Km being done weekly, and that I didn’t manage to kill the engine with all the service abuse.

Although the engine eats oil, and the pcv system leaks really made the rate of consumption increase, there was no extra oil in the inlet tract.

Thus I’m still *not* using an aos cc system other than the stock configuration with the built in separation.

If I was going to upgrade that, I’d keep the stock pcv system routing, pipe diameters, choke diameters (eg those in the 3way) and intercept the two stock location returns to the inlet pipe with KillerB BRZ/86 3D printed separators, with the collected oil/water draining to hoses with petcocks, for easy draining.

The new KB separators are much like the oem items that bmw etc use, the compact size might allow for shorter hose runs reducing extra drag in the modified system.

I’m considering experimenting with using a second pcv valve in the heads breather return to inlet pipe.
If the valve opens towards the inlet, closing towards the heads/crankcase/sump, the crankcase pressure might trend slightly more negative overall.

Perhaps that could help my really worn out engine, by increasing scavenge during boost conditions.
Or it could increase detonation potential everywhere by negating blow-by gas dilution via restricting fresh air entry to the crankcase.
There’s a third approach using a Nissan type pcv valve that remains partially open in the closed direction, probably it’s a pcv valve for an n/a engine, I’m not sure.

Either way, any pcv system mods must be verified with a crankcase pressure test during operating conditions.
I have three options using a spare oil cap with a fitting installed: Aem boost wbo2 gauge, vgb1 boost gauge or a basic mechanical gauge…

The engine appears to have survived the latest maintenance misadventure, so far.
Hopefully I’ll be back to exploring what healthy datalogs look like after literally years of slightly negative fuel trims…

[Yowie]

I’d keep the stock pcv system routing, pipe diameters, choke diameters (eg those in the 3way) and intercept the two stock location returns to the inlet pipe with KillerB BRZ/86 3D printed separators, with the collected oil/water draining to hoses with petcocks, for easy draining.

The whole "factory restrictions" thing is interesting.

After some experimentation I ended up disregarding deliberate restrictions under the theories:

(a) more than factory blow-by & water vapour anyway with E85 and a worn engine - I would rather all that water and foam out of the crank case than in; and

(b) between a "factory 6mm restriction in a compact system" on the one hand and "one metre of hose plus a tall catch can full of steel wool" on the other hand, the actual impediment to flow might be equivalent.

Performance car builders seem to go the "big hose" path with no deliberate restrictions. We road car uses probably have to pick something appropriate to task.

I’m considering experimenting with using a second pcv valve in the heads breather return to inlet pipe.

Interesting. Plenty of speculation there on the value of air entering the valve covers under some conditions.

On a related note, I'm considering a one-way valve on the rear system (hose to pre-turbo air inlet pipe) so a harder crank vacuum is drawn in non-boost conditions.

KillerB BRZ/86 3D printed separators

Good tip. I haven't seen those before but will look into.

Depending on how the "clean" returns-from-catch-can are on my new "tight" engine, I'm considering adding extra catch cans in series with what I have in order to scrub more oil.

Eg (in theory) the first can can collect the majority of the water while the second (smaller) can might have a better chance of scrubbing remaining oil without being saturated. TBA quality, but there seems to be an abundance of Chinese ripoff versions of the Mishimoto compact catch can (long & short) I already use for the rear position. I've already paid Mishimoto enough (and am broke) so Chinese might have to do for experimentation purposes if I get to that.

[bigBADbenny]

Thanks for reading, I’m of the opinion that it doesn’t matter the aos cc setup, as long as crankcase pressure is monitored, and the performance of the system is verified under all running conditions, eg: crankcase pressure trending negative = good, crankcase pressure significantly positive = bad.

The highly engineered and evolved standard is the oem setup.
On a healthy engine, crankcase pressure hovers between +/-0.5 psi, and the blowby is generally diluted with fresh air.

This should be fine using stock power levels and premium fuel.
As flatirons mentioned in one of their YouTube videos, moving away from standard power and fuel is where the stock setup might underperform.

The issue is there’s a huge difference between the requirements for a high power street setup and a race application.
That’s where pcv system tuning comes into play, and any mods to support higher power and alternative fuels might be verified for efficiency using crankcase pressure as the standard.

TBC…

[bigBADbenny]

Meanwhile, all aos/cc cockamine aside, the allied departments of Shadenfreude and Spring Summer called:

Your AC is broken, again!

Lights on, clutch on, blows warm, expansion valve hisses. FML.
I wouldn’t be upset but I replaced the compressor with a s/h unit maybe a year ago, same symptoms, diagnosis no compression, a $400 job.

Plot thickens, was that around the same time I replaced the intake manifold gaskets… using the “lazy” method without pulling the AC pump?
Was the clanking on startup potentially my replaced AC pump eating itself due to no gas?

Holy moly the penny drop hurts when it’s from such great heights as this.

By my reckoning that’s an $800 mistake with a new compressor, but is it indeed possible to leak refrigerant by simply nudging the AC lines?
Wouldn’t the leaking gas be extremely obvious, leave some UV oil traces or at least make a foul smell?

Ironically, I recall looking at the AC compressor prior to going in for the intake gasket and thinking “don’t touch that, it’s just been fixed” lol!
Or it’s pure coincidence, having used the AC way less than always on over the last year or so.

I’ll ask the experts at our other local garage, they do taxis so have probably seen everything lol.

Meanwhile the engine passes an inlet pressure test, yet stinks like blowby, and warm idle vacuum has gone from -10.3psig to -9.9/9.7…
So that’s a wait and see where the trends go as the ecu relearns.
The Learning Views remain knock learning free, the fuel learning is the sole area of concern.

[bigBADbenny]

No progress, too much rain... till today at least

before the rain set in I did test inlet pressure and found another tiny leak at the heads breather return: note to self, anything less than a Norma type SS screw clamp on the silicone inlet side of a joiner is probably going to leak, inlet tract leaks affect fuel trims.

it's not that one tiny leaks is going to blow up the engine, but one leak fixed leads to the next, the inlet tract and engine have many joins and penetrations, and the point of maintenance is to keep the engine at a minimum standard related to a sane state of tune.

Of course there's another complication getting in the way of perfect motoring: finally at 340kkm the RHS secondary air injection valve has bit the dust, stuck shut.

So this is a fault that might have a pending status, although I'm not seeing it in BtSSM...
I might try to see what's going on in RomRaider or FreeSSM, but the symptom is really rich then lean AFR during cold start to warm idle.

Pending code mean it takes *more* than one drive cycle NOT meeting a minimum standard to trigger the CEL/DTC code.

I like to occasionally make this cold start to warm idle log, around 15 minutes long with no haptic input as its a repeatable test and the parameters can be compared to a page of idle specs in the FSM, the only potential deviations being ambient temperature, fuel quality and the elevation of the vehicle above sea level.

Since the engine gets into closed loop almost immediately, the fluctuating AFR's from the valve not opening are affecting fuel correction and in turn fuel learning.

I'm guessing the ecu uses a compensation table during normal secondary air injection, adding more fuel with the extra air (to make the cats hotter sooner) but one engine bank goes severely rich (due to the failed valve) and fuel correction kicks in, afr goes very lean and then the injection event is over and the engine gets back to stoich, but with fuel learning trending negative with fuel correction trending positive to have total fuel trim slightly negative... *YAWN*

Of course I have the parts and ability to do the secondary air delete, but I also have a spare RHS valve, potentially two!

Using the info in the FSM workshop manual, I'll test the spare valve before installing it.

Somehow I simply prefer aspects of emissions compliance to deletes...

[bigBADbenny]

I used FreeSSM & Romraider logger to see exactly what the rhs bank1 air injection solenoid valve was doing.
The DTC, P2441, is intermittent and every time I’ve had the tmic off recently, I’ve been spraying it with UEC in the hope of helping it free up.

Initially the solenoid valve didn’t actuate directly triggering it in FreeSSM, so I peeped the wiring diagram and tested the stashed spare unit via power & negative using a 12V battery and alligator clip leads: pass!

Then I plugged it into the solenoid harness and cycled it in ECU test/delivery mode… Pass again meaning the solenoid harness is working.
I then put an inspection camera in the solenoid fresh air port, replugged the harness into the faulty solenoid… Pass!

Hopefully the UEC treatment plus a months worth of actuations in 5 minutes of test mode have freed it up from sticking shut?

Ideally I’d put the lower Km solenoid in however it’s looking like the turbo has to be moved rearwards to get to the solenoid mounting nuts on the pipe that leads to the head, so that’s a bigger job for another day, perhaps when installing an upgrade turbo or the spare long block

Hopefully the UEC is working and helping the tired solenoid to operate.
If it plays up again, I’ll try capture the fault in FreeSSM or Romraider logger, eg with the various relevant parameters and switches selected.
In the meantime I’ll keep spraying & using test mode to help the solenoid.

I had a think about what else might cause the afr cold start swings, fuel purge return to inlet pipe?
I’d noticed the return vacuum hose was slightly linked, so rather than pull apart the mount to add a new straight section of hose, I found a stainless steel spring of the correct length & diameter to fit into the return hose, preventing the kink.

There’s nil chance of the spring making its way past the nylon elbow in the inlet pipe, fyi.

That just leaves the potentially restrictive brass elbow and diy suction hose between the pcv 3 way joiner to the pcv return in front of the compressor to be replaced with a nylon straight butt joiner and an oem inverted W shape hose.

[Yowie]

Have you considered deleting the whole secondary air pump system? I'm interested in your reasoning as I'm considering the same.

Also, each to his/her own, but (provided there is some airflow) I wouldn't worry too much about restriction or otherwise in the crank gas breather system. It seems to work with OEM tiny restriction holes, 13mm pipe, catch cans of all different designs, etc. as long as there is some flow and air is going where is should (with no or minimal leaks).

[Having said that, I smooth the internal sharp edges on my brass joiners on general principles, so hypocrisy noted. ]

[bigBADbenny]

No delete planned, I have found another set of spare solenoid valves in the parts pile, so at this rate I have enough spares for the foreseeable future, especially as the uec treatment has solved the issue for now: with 340kkm on those!

I’m stock tuned and the sole delete is the rear o2 which is handy for external sensor logging via the ecu, but also easy enough to reinstate if required with a addition of a second cat, new sensor and a retune.

So we’re road tripping interstate, the lib is using a heap of oil, otherwise going great and returning clean learning views whilst freeway cruising, but showing negative fuel trims in the upper airflow ranges in town.

Its potentially not a boost or vacuum leak, but I saw symptoms of a compression issue:

Rolling into town after the long trip, there’s a huge downhill section of road that suits engine braking to avoid overheating the brakes.

Getting back on the throttle, I felt a familiar feeling, a slight hesitation followed by a puff of smoke from the exhaust.

Except this time the puff of smoke was a thick white smokescreen!

Luckily I was taking it easy with no one following, because that would have been quite dangerous and alarmingly to drive through.

So now I’m using the clutch when lifting the throttle and avoiding engine braking altogether, a pity because that’s the primary benefit to me of a manual car: for the control and the sound mainly.

So clearly the engine has a severe compression issue, one which should be easy enough to diagnose by the usual methods to find if the cause is rings, valve stem seals or head gaskets (no coolant loss atm).

I’ve helped it all I can through datalog analysis, health checks and repairs, and apart from oil use, it runs well.

But the fix for the actual fault means the spare long block will finally go in along with a fresh clutch, stock flywheel and some other basic upgrades.

Or potentially soldier on for another year or so and hit 400kkm for the hell of it

[bigBADbenny]

Iirc at around 240kkm, the re003 were used up, luckily my local tyre joint Donellans in Collingwood had an awesome deal going on the update to the well regarded PS4…

I finally got a HUGE road trip on them, footy season being done, Melbourne Adelaide Robe and back.

Here’s some feedback so far: they definitely don’t get in the way of grand touring fun:

I can only give feedback on them in sedate touring & city driving vs the used up RE003…

Slightly quieter, slightly less alarming when hitting standing water, overall awesome as a daily driver tyre. My set in 235 40 18 was just under 1k fitted at Donellans Collingwood, $200 more than a new set of RE003.

Given the positive reviews and decreased tyre wear compared to eg PS4, they were the obvious choice.

I’m on +35 8.5 with caster adjust lca’s and rear upper outer camber correction, stock alignment, Donellans were happy with the wear pattern on the old tyres so they only tweaked my steering wheel centering.

I asked about tyre rotation and apparently the PS5 can be rotated side to side as well as front to back. Need to fact check that, but if so that may help prolong their life.

[bigBADbenny]

On the subject of pcv systems, I’ve been reading every relevant technical article online including Wikipedia.
Its interesting that all the information I’ve found so far is fairly vague and generalised.

This means that one might need to revert to looking at the absolute fundamentals when analysing the various systems and their intent: without access to a manometer to do real world tests on both stock and upgrade pcv systems.

Here’s some of what I’ve found out so far:

The pcv valve is a controlled vacuum inlet tract leak that draws around -1 to -3 inches of mercury (will check that stat), peaking during engine braking, holding steady during idle and tapering to zero as the engine is under load with the throttle plate open.

The spring inside the pcv valve controls the exact pressure at which the valve opens and closes, in conjunction with the pressure in the intake manifold.

Naturally aspirated engine pcv valves (some or all?) do not close completely but remain partially open when closed, unlike forced induction engine pcv valves which completely seal with positive pressure in the intake manifold.

All stock pcv systems use a fresh air port to enable the pcv valve to draw a vacuum on the crankcase.
I didn’t find out the specific information as to why this is so, but the reason is probably to:

Dilute blowby, so undiluted blow-by doesn’t affect AFR too severely.

Restrict maximum negative crankcase pressure, which can theoretically pull too much oil away from the piston rings, aka wipe the cylinder bores.

Provide a pressure failsafe in the event of the pcv valve sticking shut or the main pvc vacuum tube becoming blocked.

So in the instance of a dual can forced induction aos/cc setup, if the correct pressure differential exists between the fresh air breather and the sump breather returns to the inlet pipe, the catch can on the breather should only fill with oil in the event of either heads oil surge or a restricted sump breather, or stuck shut pcv valve.

This point is backed up by the pcv system flow diagram in the FSM, and in the general descriptions online of how pcv systems operate when working properly and by the failure modes, eg excessive oil in the inlet and or on the engine air filter.

Its only when looking at diagrams for aftermarket aos/cc setups that the fresh air breather flow path is seen as two way, in my opinion a reversal of flow away from the intake manifold can be defined as a failsafe mode when the sump breather or pcv valve is blocked or restricted.

[rexhunta]

If you can make a setup run, i am interested in looking at it as well. ATM i've got the 2 catch can setup and gives so many extra lines.

[Yowie]

Make sure you consider that this whole area may not be the finely-tuned orchestra you may be assuming. It could just be plain old non-critical emissions compliance stuff with flow factors dictated by other concerns (cost, packaging, manufacturing ease, etc.).

OEM breather gear changes over time (see the hard plastic T versus the soft rubber T that changed on an EJ25 circa 2008). Are they more finely tuning it for more perfection (despite the wide range of engine running conditions & wear states) or looking at cost factors?

Performance builders just run "big enough" breather hoses. Serious performance builders try to run a hard crank case vacuum via dry sump suction or (for naturally aspirated off-road applications) a pipe connected to downstream exhaust.

So in the instance of a dual can forced induction aos/cc setup, if the correct pressure differential exists between the fresh air breather and the sump breather returns to the inlet pipe, the catch can on the breather should only fill with oil in the event of either heads oil surge or a restricted sump breather, or stuck shut pcv valve.

Or, per Occam's razor, the explanation of fluid in the breather-side catch can is that, under boost conditions, blowby pressure comes out "all the holes at once"?


The idea of a hypothetical "correct pressure differential" seems far-fetched to me when you consider not only the wide range of engine running conditions (22 inches of mercury vacuum to 18+ PSI boost) but the wide range of factory configurations. Does an OEM TD04 Forester XT have a different OEM breather system than an OEM VF52 Forester XT "S Edition"? If not, I suggest it's non-critical flow and "does it work ok?" is the only question being asked by the manufacturer.

[bigBADbenny]

It would be great to plot the full evolution of various pcv systems over various generations, and also look at the choke sizes of different pcv valves from engines of different capacity from other manufacturers.

Also plot the evolution of the location of the inlet returns.

And to map out all recommended routing schemes for the popularly available aftermarket aos cc/systems.

But I simply couldn’t find any reference anywhere at all that the fresh air breather was anything but a fresh air supply to the crankcase, and that blowby exiting this port into the inlet is a fault or failsafe mode, according to the design intent of stock pcv systems.

Personally I think the placement of the pcv return near the fresh air breather as seen in some aftermarket inlets is a feature for convenience of install and an archaic hold over in the instance of attempting to retain some of the stock pcv system arrangement.

An inlet like this is probably more suited to a three into one return aos/cc setup, with or without pcv delete, the latter being a race cars only setup, due to making positive crankcase pressure during idle. From memory the AVO install instructions mention blocking off one or other of the returns.

Part of the issue with the aftermarket inlets and aos cc systems is that the install instructions are all procedural, there’s little in the way of diagram maps, eg Perrin and IAG.

So my line of enquiry is to get to the stock intent, then try nut out the rest.
Part of this is thinking about the possibility of using a tunable, scalable pcv system that would ultimately completely immune to fuel type, boost & blowby pressure.

So without using a crankcase vacuum pump, potentially not even a pcv valve, yet allowing some tune-ability of crankcase vacuum and failsafe modes.

[bigBADbenny]

If you can make a setup run, i am interested in looking at it as well. ATM i've got the 2 catch can setup and gives so many extra lines.

CC5F8B99-07EB-44F6-950B-D80CE762D915.jpeg (74.89 KiB) Viewed 1023 times

Not really, here’s a setup that retains the stock arrangement, so the S pipe from sump breather port to the three way is extended to one can and back, the fresh air breather’s lines are extended to the other can and back.

Since the can on the fresh air breather is there as a failsafe for a blocked pcv or sump breather, or oil surge in high G corners, one could use a can on or opposite each head for three cans total.

There’s other possibilities like running dual cans inline for extra, extra redundancy.

3A9D5400-A853-4D32-AB04-CAF04D55BBE1.jpeg (147.42 KiB) Viewed 1023 times

Just don’t be tempted to use the balance ports… unless you want to delete all other ports and use these in a three into one race arrangement (it’s been done lol) but I’d still go the sump breather over the crankcase vent.