Theories on crankcase vapour and pre-detonation...

[Gannon]

I have recently turned my boost back up to 10psi in search of more power, only to find my engine barely responded at all, feels about the same as what it did at 7psi.

... a quick rundown on what i already have

EA82T fully reconditioned 2 years ago
4 plug ecu
Custom up-pipe
TF035 turbo
Custom 2.25" bellmouth downpipe and exhaust with new 2.25" cat
Liberty WAIC

I have a air/fuel ratio gauge that reads 'full rich' whenever on boost
New fuel pump and fuel pressure regulator
Always run on Vortex 95 or BP 95 (also tried E95 and makes no difference)
Timing set at 20-22deg

So with all (that i can think of) other bases covered, i made a guess that the knock sensor is forcing the ECU to pull timing

I unplugged the knock sensor, went for a squirt, and it pinged straight away.
so i plugged it back in and went searching for anwsers.

Now... what makes an engine ping?

With intercooling and good fuel covered, what else is there???

I have noticed recently that i have a large amount of oil coating the insides of my turbo, intercooler and related piping, and the intake plenum.

My PCV system is pretty much as factory had it.
The drivers side cam cover feeds into the pre turbo pipe. The rear crankcase and passenger side cam cover T and feed into the PCV valve. The valve was replaced recently.

Is all this oil normal (i do give the engine a fair thrashing)

So ive come up with a solution to try to fix it

A custom catch can




It hasnt been welded yet, but hopefully i can get it done this weekend by a fellow board member who has a TIG

If anybody likes, i can do a full write up about the catch can in another thread

If its not oil, then what else is could be causing it?

[SCOOBIDOO]

the ea82t already has an inbuilt catch can at the rear of the block on all efi engines....
when you rebuilt your engine,did you replace the pistons? no i am guessing because you refer to it as being rebuilt and not reconditioned...your detonation is due to the blowby you have in the engine...the ring grooves in the piston are worn and will not hold your rings square in the groove,and now the edge on the ring is no longer square to the bore face. the top and bottom edges are being rounded off to a curved shape....this means there is very little of the ring face touching the bore..and when you ask for more power it blows by the ring instead of sealing.....
your catch can wont change a thing....its extra unnecessary junk in the way...i dont run one and never have any oil problems at 12psi.........oh and yes i have a tig welder.

[Gannon]

I did have the whole engine rebuilt (first post edited to show)

The 3 baffles at the back of the block are a cost vs effectiveness attempt at a catch can.
Im just taking it a few steps further to improve things.

Thankyou for clarifying that oil vapour can cause detonation, i made a post about exhaust gaskets, pop over there and tell me ive done something wrong there too.

[steptoe]

I feel i can answer part of 'what makes an engine ping' especially when you say you disconnected the knock sensor. On the 3 plug ECU the KSModule advances timing as far as it is designed to advance and uses signals from a healthy knock sensor to retard timing. You go and disconnect the retard device from the circuit gives unit the go ahead to advance as much as it can. I take it 4 plug ECU sytem works same way keeping all workings in house and not employing the outside KSM

Drain to what breather ?

If I was doing custom catch can I would do three inlets, one for LHS rocker cover, one for RHS, and one for the little girl... the rear block jobby and let oil drain go to oil fill tube, then just a filtered lid like a SCA breather $10

And you go up the boost wouldn't one think to retard your initial timing correspondingly ?? I dunno , never increased boost

[discopotato03]

Oil vapor can cause detonation because it drops the effective octane rating of your fuel .
I'm not at home ATM and can't remember the crank case plumbing system on EA82's , other than on mine the drivers side cam cover vents into the AFM to turbo inlet air duct .

What oil separators do is basically strain the oil from the crank case gasses and give it a drain path back to the sump .

If you engine re builder was on the ball he/she should have measured the ring to ring land clearance as part of the job . Rings seal by compression pressure getting behind them and pushing their outer surfaces out onto the cylinder walls , it ain't just the ring tension doing this .

The higher you rev an engine the more "windage" it gets meaning an increasing cyclone in the crank case which carries more oil in the gasses in there .

If I was going to have an oil/air separator I'd just have it where the plumbing goes back into the pre turbo inlet hose . The reason being that unless manifold vacuum is reasonably high the PCV valve should be closed .
I think there is a greater chance of oil vapor getting into the inlet tract via the drivers side cam cover breather than the PCV valve .

My rally people used to build high rev Khana buggies and they had to fab oil separators and learnt how to make them work , must ask about them next time I'm there .

Gannon as an experiment you could try using a small amount of 98 ULP when your tank is nearly empty as the rise in octane can be felt as less retarded timing - provided the knock sensing gear is pulling the timing back .

Different people do things differently and I like the idea of higher compression ratios and NA cams because they allow engines to make better part throttle torque off boost .
Provided they can do this you can then go a little larger on the turbos turbine housing which moves the boost curve up on the engines rev range (and removes some exhaust back pressure pre turbo) which causes detonation as well .
TF035's paint you into a corner because they have a pretty small diameter turbine and an enlarged waste gate bypass valve .
To raise boost pressure means delaying the waste gate valve opening in order to feed more gas through the turbine to accelerate it and its compressor wheel .
So it the turbine is a bit restrictive and the gate opens later pressure rapidly builds up between the exhaust valves and turbine housing . Pressure and temperature rise together and hot exhaust back flows into your cylinders in the valve overlap phase . Its double trouble because these gasses pollute AND preheat the "fresh" incoming air/fuel charge and raise the detonation threshold as well .

My engine builder keeps telling me that there are only two ways to increase the base engines performance without pushing power higher up the rev range . The first is compression ratio and the second is increased capacity .
Promise me that if you ever build another EA82T short you'll get the decks machined to jack the CR up to at least 8.5 to 1 .
If I had my time again I'd have 9 to 1 to better suit the MPFI NA camshafts but 8.6 will have to do .
The golden rule is that its greater airflow not boost pressure that helps increase power output .
Boost pressure is after all really just resistance to airflow through the inlet tract , if you reduce the restrictions you don't need as high a boost pressure to move the same volume of air .

Cheers A .

[Gannon]

when you say you disconnected the knock sensor.

I did it to see if the engine was either detonating and retarding timing, or i had a dicky knock sensor that was retarding timing without any detonation present

Yes the 4 plug ecu works in the same manner as the 3 plug, base timing is fully advanced, then pulls timing as it needs to, only i believe the 4 plug is faster and maybe has a memory (can someone verify this?)

Drain to what breather ?

If I was doing custom catch can I would do three inlets, one for LHS rocker cover, one for RHS, and one for the little girl... the rear block jobby and let oil drain go to oil fill tube, then just a filtered lid like a SCA breather $10

Mine will have 3 inlets, like you stated above, but seeing as it will be mounted above the engine, when i switch it off, any oil in the bottom of my container will simply drain back down the easiest path,... the one to the back of the block.

What do you mean by little girl?

you go up the boost wouldn't one think to retard your initial timing correspondingly ?? I dunno , never increased boost

I figured that if the knock sensor is doing its job, it will always retard to just below the point of detonation. Ive never heard pinging so i guessed it was ok.

If I was going to have an oil/air separator I'd just have it where the plumbing goes back into the pre turbo inlet hose . The reason being that unless manifold vacuum is reasonably high the PCV valve should be closed .
I think there is a greater chance of oil vapor getting into the inlet tract via the drivers side cam cover breather than the PCV valve .

I am still having one of the outlets going to the pre turbo pipe, cos yes, most of the air will travel this way but figured id keep the PCV valve path as well, it wont do any harm.

Gannon as an experiment you could try using a small amount of 98 ULP when your tank is nearly empty as the rise in octane can be felt as less retarded timing - provided the knock sensing gear is pulling the timing back .

Next tank is gonna be 98. And the knock sensor has to be doing its job cos it never pings while its plugged in

Thanks for you input fellas, its much appreciated

[SCOOBIDOO]

Promise me that if you ever build another EA82T short you'll get the decks machined to jack the CR up to at least 8.5 to 1 .
If I had my time again I'd have 9 to 1 to better suit the MPFI NA camshafts but 8.6 will have to do . .

Please correct me if i am wrong..
my calculations say that there must be at least 3mm + per side machined off to acheive something even close 8.5 cr.
Could you please fill me in on how this is possible?

1/do you know how many mm,s have to be removed to acheive this raise in compression?
2/once the block and or heads are decked,now the cam axis is closer to the crankshaft,by more than 3mm per side... how do you address the positive cam roll on the distributer side..and the negative cam roll on the turbo side?
this would now mean one side of the engine now has more than factory spec cam timing..and the other less.... and supplying uneven power output from both banks of cylinders..
3/ and lastly how do you get to bolt the intake manifold back on?
Keen to hear your response.

[discopotato03]

I don't have the answers to how the original engine builder went about it .
The engine came complete out of a running Vortex and I don't remember seeing any mods to it when it was in a thousand pieces .
All I can tell you is that the pistons are not very close to the tops of the bores in standard form which is really bad from a squish or quench , as some people call it , point of view .
I doubt very much that a combined 6mm (almost 1/4 inch) has been removed from this engines cases , the std bore size is 92mm and you don't have to go very far down a cylinder that size to remove quite a few cc's .
You are calculating on swept volume vs clearance volume aren't you ?

If you think about it the cam timing is pretty straightforward , you set it up as per normal and if the marks all line up no worries , its the tensioners reeling in any additional belt length gained by the cams being slightly closer to the crank center line .
Now if for any reason they won't line up (between belt teeth) all you have to do is weld up the pin hole in the pulleys and re drill them where they need to be .

Cheers A .

Late Edit , I am wrong about the pistons being down the bores at top dead center . I now think that my engines cases had a bit milled off the tops of the cases and the piston crowns are a little way up inside the head gaskets volume .
EA82's being a non interference engine must have lots of clearance between the valve heads and piston crowns in the overlap phase .
Anyway will confirm today the facts about my engine build .

A .

[steptoe]

the above gets one thinking don't it ?

The difference between the 3 plug and 4 plug ECUs as far as knock control may be something.I know the 'base' timing for the 3 plugger is 20 degrees BTDC at 750 ~800 rpm, but isn't the 4 plugger a set the 'base' timing a different way other than above and once it is set the ECU takes over total control of spark timing over what I reckon is a greater range than the 3 plugger. I think the 3 plugger separate KCM can only advance or retard the timing by a lesser number of degrees compared to 4 plugger ecu. 3 plugger uses 1. operator set initial timing, 2. a KCM with knock sensor and 3. advance retard dizzy diaphragm, where as 4 plugger takes over the lot don't it ?

[discopotato03]

In theory the 4P system can discharge the coil at any point because its not limited by mechanical systems .

I'm not 100% sure but I think the ignitor/signal generator gadget is different within the 3P turbo distributor and can only redard the timing so far .

Overall the 4P system is better but without aftermarket computers your stuck with what Subaru gives you to deal with what it thinks is a std spec engine .

A .

[SCOOBIDOO]

All I can tell you is that the pistons are not very close to the tops of the bores in standard form which is really bad from a squish or quench , as some people call it , point of view .
.

you should stop misleading people on forums....i reconditioned an ea82t over the last week with all stock parts..the pistons sit level with the deck of the block.....

[steptoe]

Disco may be tired (at the wrong end of weird hours shift) praps?

I have not seen this myself either.

What sort of distance from deck to top of piston at TDC have you found Adrian ?

In 1/1000" or mm OK

[shuffbag]

is this what you are talking about



its pretty level

i.m also putting together a ea82t.

[Gannon]

well i finished my separator can (big thanks to Ben for TIG welding it for me, very professional finish)



I havent finished fitting it yet (i ran out of daylight and the mozzies were out to feed) but you get an idea of how it works. I didnt get a picture of the filling i used, but its just a few packets of those steel pot scrubbers

I'll post more tomorrow

[SCOOBIDOO]

Disco may be tired (at the wrong end of weird hours shift) praps?

I have not seen this myself either.

What sort of distance from deck to top of piston at TDC have you found Adrian ?

In 1/1000" or mm OK

Maybe jonno....down here what we all call it is a dickhead...
who else installs a catch can to hide what rooted pistons rings are not doing?


Maybe ben can tig him up a bucket to hang under his chin ......

[discopotato03]

Very quick to be abusive aren't you Scoobidoo , a wise person once said that ego is often in inverse proportion to actual ability . By the sounds of things it fits you like a glove .

I concede I may be wrong with piston heights , maybe I was mistakenly thinking of the pistons being proud of the case decks and a little way up inside the head gaskets volumes .
Obviously if you mill a block (cases in this case) and the pistons were flush they end up being proud or slightly above the decks faces .
No matter , I will confirm this when I speak to Arthur Jackson in the near future since he rebuilt the thing .
Its been sitting out at Wilkins Motor Sport where the tear down and reassembly took place , Stuarts brother Graham (Wilkins Engineering) does most of their engine machine work .

Now you could be really stupid and challenge their experience and abilities but considering that they were born and inside engines before you could count fingers I don't think you'd have a win .

I always remember Joss Acklands famous statement in Lethal Weapon 2 , in that magnificent South African accent : "Who's the dickhead now ?"

A .

[Gannon]

Are you quite right there? Are you having a bad day? Did somebody steal your last penny?

If you think what i built is a waste of time, so be it. Im happy with it, and im sure there are other members are interested in what ive done. I have a problem and this is my solution.

who else installs a catch can to hide what rooted pistons rings are not doing?

Who else installs an extra injector to hide what the factory ecu cant do?

My catch can really isnt that different to your band aid...

fit the extra injecter in the centre of your pod filter

Countless car enthusiasts install aftermarket BOV's, pod filters, grounding kits and 4" exhausts... and we all know they do diddly squat.

Im open to constructive criticism, but if you are gonna make unpleasant comments, be it directed at me, or other members posting in my thread, keep them to yourself!

[Ben]

Maybe jonno....down here what we all call it is a dickhead...
who else installs a catch can to hide what rooted pistons rings are not doing?


Maybe ben can tig him up a bucket to hang under his chin ......

Alright champ, in the last 3 years that you have been a member here you have done a great deal of arguing, misleading of your own and generally being a pain in the arse. (Anyone can feel free to look back through his posts)

You clearly have half an idea what you are talking about, otherwise you wouldn't be able to run your vehicle checking service. You have also demonstrated this in actually helping some members. Although you seem to be very selective in this.

Please start being more constructive in your criticism otherwise you'll not be able to post your crap any longer.

[steptoe]

Gees, just wonderin' if SCOBIDOO has had his PC hacked, coz where I come from and where I am now, calling someone a Richard Cranium or its variants ain't a healthy way of living.

I turned one of my air boxes into a catch can , occasionally cleaning its base out) in what i felt was an emergency situation up on the highway far from home when I blew off another oil cap. It has stayed that way for another 40,000km (motor done 430,000km now) although I replaced the squash ball oil cap with the real thing after a while. Just figure Gannon is doing a better job than mine. BTW what is that upturned biccie tin above the turbo ?>?

[Gannon]

what is that upturned biccie tin above the turbo ?>?

That "up-turned biccie tin" is the factory RS Liberty turbo heat shield. It was the only heat shield i could get my hands on