Bob Rawle

Modifying the oil pump pressure relief valve is good for a couple of reasons, first off a little extra pressure in the system is good although overdoing it can blow seals out, more importantly the standard piston in the relief valve is prone to stick, it has a machined annular groove in it which has a sharp edge to it, as its running in an aluminium bore then there is a risk that the piston can pick up and stick especially if prolonged high revs are used, when the revs drop the piston stays where it was. So now you have the engine running at lower rpm and the pressure valve is stuck open, pressure is now much lower than it should be under the conditions of the moment. Again this can lead to oil film breakdown and pick up. Solution is to polish off the sharp edges or replace the piston with one that does not have the groove. These are the oil pump mod that is normally done.

Do I sustain high revs ? I have wired my check engine light as a shift light, its set to 6500 rpm and thats when I change under almost all conditions, since I have max power at 7000 rpm then this sits me nicely in the meat of the torque curve (max torque at about 3800rpm) ... no point pushing higher unless forced to by conditions, although the redline is at 8000 for the later cars I would certainly not advocate running a car for any length of time at that level unless it has been specifically set up to do so and that lets out all road cars IMHO. Again, we have high performance cars but they are not designed to sit on the red line for any length of time, the only real change that allowed the later cars to rev higher was to the valve train, be safe and watch the rev counter.

The choice of oil filter relief pressure is surely one of the reasons that Subaru are so picky about oils, under cold conditions the relief valve could open but only if the oil viscocity is low enough, so a 10-60 oil is more likely to open it than a 5-40. Interesting that Subaru specify the 5-40 for normal use, they aadvise against the 0-40 as they haven't tested it yet that would be least likely to open the relief valve.

Observing oil pressure I noticed that my car has 6 bar normal pressure and 2 bar idle pressure when hot, as revs increase pressure increases proportionaly, at 3000 rpm it makes about 5 bar, at 3500 its 6 bar, at 2500 its about 4 bar. When cold, idle presure is 6 bar at start up and progressively drops as it warms up but it only settles down to its hot settings after the oil temp is above 80 degrees, hence my waiting for that temp to be reached before boosting the car.

So maybe the choice of oil viscocity also has an impact on this. Some oils also carry a higher level of detergent in them than others, an oil like this could dislodge debris from the engine internals, cold starting could allow this debris into the bearings, a hot spot could occur due to pick up, bearing failure could then occur but ... that would be just as likely to happen to any of the bearings except that the middle crank bearing oil feed is bigger than the others as it has to cope with two big ends, this would explain why No 2 or 3 bearings seem to be prone, given that logic says that flow is greater to the centre main and that would explain some of the circumstances that may be occuring to cause these bearings to fail. I normally use a 5-40 oil as that is approved by Subaru, I have tried other grades but found the engine much noisoer when cold. Thats not to say that I think that a 10-50 is a problem, far from it, Subaru are pretty picky about the oil spec that should be used though and they will have calculated the oil filter relief valve pressure based on their oil recomendations. After market filters may not have the same attention payed to the relief valve setting.

As a thought would those who have suffered engine failure be interested enough to participate in some research ? If the circumstances and history of the vehicle up to the failure were known then it would be possible to analyse the data and look for trends, info needed would be make, MY, mileage, sevice habits, oil used, when last serviced, circumstances of failure, events leading up to failure (if not incriminating of course) and so on. Info would need to be detailed and quite specific.

I'd be happy to do the analysis and provide the info to all interested parties. Seems to me that is something we could be doing to help ourselves.

ric

MRT (www.mrtrally.com.au) offer a fuel rail kit and wrote me the following in an email:

"The fuel rail kit is AU$450 exchange or AU$650 outright and includes
everything you need to replace the fuel rail on the effected side of the
engine. It is designed as a direct bolt on replacement.

You don't mention what other modifications you have made to your car. This
modification is really only needed on cars running high boost, or cars used
in racing/rallying. "

greetings
ric

bob

Bob Rawle/
Very good idea Bob, I only hope those who have had troubles will give you the info you require. This could be of great benifit to us all.
Please assist Bob Rawle all you can.

rsquire

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:<HR>As a thought would those who have suffered engine failure be interested enough to participate in some research ?<HR></BLOCKQUOTE>

An excellent idea Bob, I'll get "my Drivers" onto it when they return from Germany.

Would it be best to email you the detailed data for your examination, rather than using the BBS

Richard

carlos_hiraoka

Does anyone run a "modifed oil pump" ?????

Carlos H.

CraigH

Yep, I'll send a mail too if that's ok.

This may be simplistic (as I'm a simplistic person ), but if a high proportion suffer from these failures after a service, how many of the cars have had the engine flush favoured by most Subaru dealers?

Could this not be dislodging contaminents also?

Adam M

Bob, there is one thing that doesn't quiter make sense to me.

Unless there is a specific failing in the subaru engine, the fault you are pointing your finger at seems to be generic to all cars.

As in, hot spots and debris deposited due to poor practice in oil changing. Surely the symptoms of what you describe would cause any engine to fail?

Can you explain why this diagnosis is specific to subaru please?

Adam M

And one more thing,

I (actually pat) has just installed my oil temp sender in the same position as bob by the sound of it (v close to number three gallery) and am wondering what temperatures people are seeing in normal driving.

I was shocked to see mine rest at low load and on cruise between 90 and 95. After a swift boot up to about 1.1 bar with my foot planted for about ten seconds, oil temp rose from 90 to 100. This surprised me as |I have my warning light set at 105.

This sounded a bit high to me.

Can anyone post there normally viewed temps and the position of their oil temp sender please.

RaymondH

This has to be one of the best and highest quality threads that I have read for ages. It reminds me what Scoobynet is all about when I was getting fed up with all the rubbish being posted. Anyway,

Adam - I have my oil temp sender in the same place as you (ie: under the turbo if I've got the cylinder designations correct)and I get very similar readings. I have had the instruments fitted for a few months now and it is surprising how much quicker the oil heats up at this time of year compared to in the colder months. I find that my engine is running &gt;5 degrees hotter now than it was in winter but I think that the most I have seen is 110 a few weeks ago on the SW Scotland run. Also, my oil pressure echoes the figures given by Bob earlier on.

Raymond (MY98)

Bob Rawle

I think it best if anyone wanting to mail details to me did so off line, not everyone wants this info to be public for a variety of reasons and I would not be using owners names (not needed in the analysis) and so confidentiality would prevail. Use the address in my profile.


Specific to Subaru or generic to most cars .... that depends on several things, all cars don't necessarily have the same oil distribution and filter type that Subaru use, some have the bypass valve in the oil pump only, for example, and yes, of course the routine for oil changing should be followed irrespective of vehicle, its only common sense in that respect although maybe not obvious.

The real point, I guess, is the fact that most people assume the Impreza is an infallible machine which it is definately not, it needs TLC as much as the next car. And I am not putting this forward as the reason for failure, merely attempting to open the debate and discussion up into an area that I think is worth examining.

Adam, it sounds like your oil temp's are pretty much in the norm, my cruise is the same, by experiment if I hold zero boost and let the revs build then oil temp runs at 98 deg C. On boost it just creeps past the 100 and swiftly drops back off boost. All this is normal. Maybe your base info led you to believe that cooler temps should be expected, having confused myself, in the past, into believing that sump temps were in the 60's it soon became clear that certain guage sensors can be artificially cooled by air flow under the car. The current car has had the temp sensor in the sump (via sump plug) and in the oil gallery as yours, the oil gallery position shows faster reaction to boost, for example, but no real difference in temperature levels as such.

Since the oil is "cooled" by the engine coolant then we should expect oil temp to be at least that (80-85 deg C) when moving and we should also expect oil temp to increase as coolant does.

Oil needs to work at temperature, between 80 and 100 is probably the ideal, less than 80 is too cool IMHO, for perspective my Vauxhall turbo used to run at 105/110 deg c oil temp under cruise, on boost it ran at circa 120 deg C, that was with an oil cooler fitted as well. The Impreza does not seem to run with excessive oil temp as normal.

owbow

just to ask a stupid question maybe...

i'm new to all these electronics and turbocharging etc...but keen to learn!

if the exhaust system or induction side of things are changed, encouraging increased flow through the engine of air, we used to have to rejet our carbs 'in the old days' to give more fuel.

would all these advanced modern electronics, airflow sensors etc not be able to increase the fuel flow to match the increased air flow? and would one of these 'link' systems i hear so much about do the same without being manually told to do so by remapping?

oh yeah, hello to everybody who was at the Ring, where i did a lap in the 22b. what a car!!!

Owain.

MorayMackenzie

rsquire,

I suspect Anders meant "breathing parameters" rather than induction parameters.


A question to clear the air:

Is it the case that the exhaust system in use on the car at the time of the engine failure is _not_ the same exhaust system that was fitted when the ecu map that the car was running at the time of failure was finalised?

Moray

bob

Now we are starting to worry those who have had downpipes and other exhaust changes made to their car. Should they all now go and get an aftermarket ECU to allow for the extra flow? Or will the STD ECU be able to handle this?

Have any of you considered Richard has told Bob about the other exhaust mods that was done after the mapping?

Anders

Since I have been mentioned earlier I would like to clarify:

1, my car had 29,000 hard miles including high speed continental runs just under the rev limmiter. Including 112 miles until it ran out of juice on the way down to Catalunia (of note the last 10 miles were at 56 mph as it became obvious we were running out of fuel )

2, my car lasted 600 miles after the Powr Engineering rechip.

3, David Poor stated it was not his products fault but a fault of the oil suply to the number 3 bearing.

4, I needed a new turbo as the turbo seals had fried, in the heat.

5, I needed a new manifold as it had gone scarlet with the heat.

6, There was detonation on the pistons.

7, The valve stem seals had fried due to the heat.

In addition the number 3 bearing was smashed with a bent crank both due to localised oil deprivation in number 3, one hypothesis is that detonation on number 3 with associated heat had led to a disruption of the film of oil for the number 3 bearing

Anyway David Poor told me he genuinely believed his re-chip was not at fault. I just thought it was a bit off that they could charge me £700 and then the car fail after 600 miles and then I discover that the "don't worry I will look after you" meant that he couldn't tell me how much it was going to cost when I thought it was all being done for free!

Trying to be open minded I have to say that my cars failure could have been coincidence, as could Mr Cartwrights and Mr Martins 22bs!

Richard Squires car ran faultlessly until induction parameters were changed without the map being adjusted and then banged off the rev limmitter

This is so sad

I just think that it takes ***** of steel to offer car tuning upgrades then cars go bang and then say:"not our fault its an inherent problem with the car"! Thats why you go to experts in the first place isn't it?

BTW I always liked Mervyn

rsquire

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:<HR>Richard Squires car ran faultlessly until induction parameters were changed without the map being adjusted and then banged off the rev limmitter <HR></BLOCKQUOTE>

Being, that I was there at the time, please let me state what did or did not happen with my car.

Just for your information NO induction parameters had been changed.

Richard

*edit* Bob we'll email you offline with the full details that you requested. I fervently hope that this important thread does not become derailed.

[This message has been edited by rsquire (edited 03 July 2001).]

bob

This is a great thread now lets not let it go astray. It would be a shame to have a lock put on this one after we are just starting to achieve something.
Any info you have concerning a No 3 big end failure please send to Bob Rawle. You can get his email addy from his profile.

R19KET

Richard,

I think Anders is referring to the change of exhaust parts.

If one changes an exhaust, in an attempt to gain more power, and after the changes the car does feel quicker, it's likely that the engine is flowing more air. More air, needs more fuel.......

Mark.

rsquire

Mark,

I disagree totally.. Induction is pre the engine..

Anyway enough of that nonsense, what we do need is to keep this thread on track and keep the information going to Bob Rawles ..

Somewhere in here are facts that are important to ALL Subaru owners be it... 22B owners, STI owners, UK turbo owners, or even humble Naturally Aspirated North American owners..

Richard

Adam M

I dont want to express the wrong concern but would anyone object if this thread were moved to drivetrain as I think it should be. Obviously i dont have the poor to do this, but in writing this, stef or moray may agree with me so I thought I'd make the point.

This is a very important subject, and the general forum is diluted with other "less" important subjects which will push this one out of contention. In Drivetrain, it may last longer.

R19KET

Richard,

I'm not certain what you're disagreeing with, that Anders was obviously referring to the exhaust, or that a higher flowing exhaust will lean the engine out ???

Either way, it's VERY relevent, if your engine let go due to running lean, over heating, detting, and then taking out a bearing. It has to be concidered.

If we want to get to the bottom of the problem, we need the FACTS, and not some excuse to make someone feel better about the situation. How is Bob supposed to analize the data, unless the data is accurate ???.

Mark.

MorayMackenzie

Guys,

I really hope that rehousing this thread doesn't kill it, it's one of the interesting ones that simply don't seem to happen often enough these days.

Moray

moved to drivetrain

Adam M

Sorry richard but I have to agree with mark.

The circumstances of how your car let go are crucial to the information provided to bob.

Didnt you say you had it on video too?

Would have been great to have had a video in the car filming the defi gauges at the point of breakage.

Shame they weren't in at the time, and that the camera wasnt either.

Adam M

Bob,

the standard ecu can better cope with changes to the exhaust and induction because it runs rich. If a mod has made an improvement it will require more fuel, and some of the rich mixture can go some way to realising this benefit.

If you already have an after market ecu, it may be running slightly rich for safety (bob will normally do this) but much less so than the standard ecu, because it has already been optimised to the mechanical hardware of the car.

A change such as this should be checked to make sure it doesnt cause too much difference to the fuelling, that is why we should all have lambda links.

So no in answer to you question, on a standard ecu, there shouldn't be a problem, but it is worth getting a lambda link on there just in case.

Ultimately if you need some kind of ecu modification it will mean buying an aftermarket ecu or the like, but it will be a good thing as it will mean your exhaust or air filter mod has done something.

and to think, people used to question the need for a lambda link .

Jay m A

iTrader: (2)

Top thread.

Adam, when you refer to standard ECUs running rich, were you refering to the UK/EU ECUs or do you mean all subaru ECUs?

The reason I ask this is that I'm running a OE ECU on a MY95 WRX with a full decat exhaust and an uprated panel filter and am rather concerned about it running lean. I believe these ECUs were mapped for 100 RON fuel, the knock senser hopefully catering for the lesser fuel we get over here. I run on super and millars always.

I am going to get a link/unichip/whatever when funds allow, but not until I get an induction kit first but the question is:

Is an import WRX/STi running lean in with breathing mods, i.e. exhaust/induction?

After reading this thread I'm sure theres a few people out there who'd like to know the answer, if its possible.

Justin

R19KET

Bob/Richard,

My comments re-accurate info, were partly directed towards the 22b (Richard your posts do imply the failure was due to an inherent fault) but to ANYONE sending Bob Rawle info.

None of us would like to think that it may just be our own daft fault, that the engine let go. So facts tend to get manipulated, just like when we have an accident....

Mark.

bob

Adam/
I was reading a thread the other day stating Mike Woods told a lad thet If he fitted a catless exhaust to his P1 the car could det. He went on to say he would also lose his warrentee in the case of an engine seize.
Is this a marketing ploy or do you think some ECU's cannot handle the exhaust changes?

R19KET

.

R19KET

Bob,

Firstly, as it's ILLEGAL to remove the cat's from our cars, we can hardly expect Mike Wood to be advocating it.

We also know that removing the cat's, and improving flow, will lean the engine out, and will also increase overboost. Now, with a STOCK ECU this "should" be fine, but there are no guarantees, particually with the STi's where there is also an octane issue. It's not uncommon for decat'd cars to suffer from overboost, this in turn can cause det'.

Adam has already said that the stock ecu runs pretty rich, and can handle most basic mods. DON'T expect an after market ecu, such as the Link, or Motec to do the same. ALL of the ECU's run "open loop" fuel control, when on boost (uses a specific map) and WILL need re-mapping if engine mods are carried out.

Mark.

rsquire

My car showed NO signs of det or running lean when it was stripped down.

There was sign of "very old" det and this fits in with the fact that prior to the Link being fitted the car was detting when run on SUL.. however as this was over 30,000 miles prior to the engine failing I'd say it was irrelevant..

Therefore, I cannot see that changing the muffler (which was the only change made) from one brand of high flowing muffler to another brand of high flowing muffler is in any way connected.

Richard

[This message has been edited by rsquire (edited 03 July 2001).]

MorayMackenzie

When I changed from a PE T75 high flow cat centre section to a Magnex centre section it made a significant noticible difference to the car's performance and when I managed to persuade PS to check the fuelling they did find it required a bit more fuel in the midrange parts of the map. Hence, changing even one section of the exhaust system to a different design is very likely to require changes to the map.

[This message has been edited by MorayMackenzie (edited 03 July 2001).]