Building a 10,000 rpm NA engine.

[oab_au]

Oh Tony you shouldn't do this.


Do you realise how much time and money these blokes put into their rigs, to be able to drive across the desert, and climb these sand dunes?

You have just crushed their whole lives work.

Harvey.

[Dessertrunner]

Harvey,
We were traveling with 2 other 4 wheel drives and everyone was to agree on the,
"Quote of the Day"

When we climbed the dune there were about 40 people around and were the rest of our group were at the base of the dune the quote of the day was heard said, it was,

"F_ _ K the 4 wheel drive what we need is a Subaru Sports car".

Its pretty funny.
Tony

[Dessertrunner]

In light of all the discussion about what to do to develop more power we need to advance this a bit. Something Bazza said is going around in my head and also Harvey made the simlar point but in a different context.

Bazza mentioned that on one of the Subaru engine they reduced the size/dia of the intake ports in later models of the same engine. This was done for better torque and I am guess better cyclinder fill.

Harvey made the point on a number of occusions that have the ports intake and exhaust just big may reduce power and that there is a need to have some exhaust pipe on the manifold to enable the gases to pull the last little bit out of the cyclinder.

Okay so were am I heading,
We seem to ASSUME that we need to port our heads to get more airflow,
We assume we need more lift on the valves,
If the ECU if changed we will see a massive power increase,

When in fact we have no idea if there is any gain to be had, Harvey keeps making the point that the engine are very effecent and have high fill ratio, so is all this changes we are going to make going to gain anything or infact reduce the power.

Maybe what we need to do to get more power is to just change the cams and thats it unless you won't to increase revs like I do.

Clearly we need to set up the engine dyno and put a stock standard engine with a standard ECU to get some base line numbers and then change out some parts. Matt has been knid enough to offer to loan me the throttle bodies so we can check that out as well.

So whats every one thing.
Tony

[svxfiles]

Originally Posted by Dessertrunner

Okay so were am I heading,
We seem to ASSUME that we need to port our heads to get more airflow,
We assume we need more lift on the valves,
If the ECU if changed we will see a massive power increase,

When in fact we have no idea if there is any gain to be had, Harvey keeps making the point that the engine are very effecent and have high fill ratio, so is all this changes we are going to make going to gain anything or infact reduce the power.

Maybe what we need to do to get more power is to just change the cams and thats it unless you won't to increase revs like I do.

Clearly we need to set up the engine dyno and put a stock standard engine with a standard ECU to get some base line numbers and then change out some parts. Matt has been knid enough to offer to loan me the throttle bodies so we can check that out as well.

So whats every one thing.
Tony

I would to point out something about the amount of airflow that our engines are capable of.
Please read to the end.
I have a 1750 CFM axial vane supercharger bolted to an otherwise almost stock SVX 3.3.
With some very minor porting, (pictures in my old locker)
stock cams,
and a free flowinng exhaust system including equal length headers,
I was seeing up to 11PSI.
I lent my supercharger to a friend with a 5.2 L, 2002 Suburban with stock cam, heads, headers and free flowing exhaust.
His stock crankshaft pulley was bigger in diameter than my 6.5 inch custom made one.
He saw 15 PSI!
His bigger engine should have made less PSI as it is a bigger room to fill.
His crankshaft pulley was bigger to spin my supercharger faster, pumping more air.
My conclusion, our stock engines can flow a hell of a lot of air!

[bazza]

Originally Posted by Dessertrunner

In light of all the discussion about what to do to develop more power we need to advance this a bit. Something Bazza said is going around in my head and also Harvey made the simlar point but in a different context.

Bazza mentioned that on one of the Subaru engine they reduced the size/dia of the intake ports in later models of the same engine. This was done for better torque and I am guess better cyclinder fill.

Harvey made the point on a number of occusions that have the ports intake and exhaust just big may reduce power and that there is a need to have some exhaust pipe on the manifold to enable the gases to pull the last little bit out of the cyclinder.

Okay so were am I heading,
We seem to ASSUME that we need to port our heads to get more airflow,
We assume we need more lift on the valves,
If the ECU if changed we will see a massive power increase,

When in fact we have no idea if there is any gain to be had, Harvey keeps making the point that the engine are very effecent and have high fill ratio, so is all this changes we are going to make going to gain anything or infact reduce the power.

Maybe what we need to do to get more power is to just change the cams and thats it unless you won't to increase revs like I do.

Clearly we need to set up the engine dyno and put a stock standard engine with a standard ECU to get some base line numbers and then change out some parts. Matt has been knid enough to offer to loan me the throttle bodies so we can check that out as well.

So whats every one thing.
Tony

Subaru introduced tumblers, exhaust pumps, AVCS and small port heads for multiple reasons - the MAIN one being emissions. The stock tune shows holes everywhere in the AVCS and timing to ensure the emissions are always healthy and the AFR's get leaner over the years as does the cross over point from closed loop to open loop. However when you fix all this tuning done to keep emissions good, AVCS is then a brilliant tool for tuning and bringing boost on earlier etc.

I would say these engines need both cams (basic regrind) and exhaust headers (like Jack's) to get them revving cleanly to 7000 rpm and pulling all the way there. Currently as I've said and experienced in your Desert Monster and my own setup is they tend to fall over around 5600 rpm with the stock cam and the torque and power drops off above that.

[Dessertrunner]

Originally Posted by svxfiles

I would to point out something about the amount of airflow that our engines are capable of.
Please read to the end.
I have a 1750 CFM axial vane supercharger bolted to an otherwise almost stock SVX 3.3.
With some very minor porting, (pictures in my old locker)
stock cams,
and a free flowinng exhaust system including equal length headers,
I was seeing up to 11PSI.
I lent my supercharger to a friend with a 5.2 L, 2002 Suburban with stock cam, heads, headers and free flowing exhaust.
His stock crankshaft pulley was bigger in diameter than my 6.5 inch custom made one.
He saw 15 PSI!
His bigger engine should have made less PSI as it is a bigger room to fill.
His crankshaft pulley was bigger to spin my supercharger faster, pumping more air.
My conclusion, our stock engines can flow a hell of a lot of air!

Don't have the answer bill but would say that the other engine is only 57% larger so it won't require a much bigger drive pulley to match that.

Its a interesting point you make.

We have to get the engine dyno going ASAP to put some meat around all this stuff.
Tony

[SVXRide]

Tony/Harvey,

It would be great to have Harvey's EG33 computer model "tweaked" based on actual engine dyno results.
Example -- increase intake runners by 3 inches in length. Model predicts X, actual dyno results show Y. Go back and tweak model so that it predicts Y.

Just a thought for the path forward.

Bill

[Dessertrunner]

Yes Bill you are right,

Unless we start somewhere we are never going to know. As I see it the engine that I got from Mike has cam specs that I am sure Harvey has and the compression is known as well so based on that we should be able to drop it in the dyno and see what is does. THen compare to Harvey's simulator.
So now the ball is in my court to get the dyno running.
Tony

[bazza]

Originally Posted by Dessertrunner

Yes Bill you are right,

Unless we start somewhere we are never going to know. As I see it the engine that I got from Mike has cam specs that I am sure Harvey has and the compression is known as well so based on that we should be able to drop it in the dyno and see what is does. THen compare to Harvey's simulator.
So now the ball is in my court to get the dyno running.
Tony

I asked this from Harvey back in June before any of the recent issues occurred but he wasn't interested. Unfortunately Matt D lost his data file or I'd have my simulator up and running and calibrated by now. At the end of the day it doesn't really matter though - the dyno is going to give 100 times more useful data mainly because it's real. We could do stock base runs, then boosted runs, then play with the trumpets and then the headers and then add cams. The amount of data we are going to get is massive and it's going to be very useful. I actually have a basic excel simulator (which I've calibrated) with both dyno runs and drag passes that will give you airflow + power based on engine VE so we may be able to enhance this nicely and share with all with a new EG33 version.

Are you able to stick the stock engine into your computer and see what it comes up with at 10 / 15 psi with the stock manifolds. Would be interested to see how close it is to reality. Turbo is a GT35R, from here: http://www.turbobygarrett.com with a 0.82 rear.

[SVXRide]

Originally Posted by Dessertrunner

Yes Bill you are right,

Unless we start somewhere we are never going to know. As I see it the engine that I got from Mike has cam specs that I am sure Harvey has and the compression is known as well so based on that we should be able to drop it in the dyno and see what is does. THen compare to Harvey's simulator.
So now the ball is in my court to get the dyno running.
Tony

Tony,

Bravo! Here's to dyno runs and correlation to computer models!

Bill

[SVXRide]

Originally Posted by bazza

I asked this from Harvey back in June before any of the recent issues occurred but he wasn't interested. Unfortunately Matt D lost his data file or I'd have my simulator up and running and calibrated by now. At the end of the day it doesn't really matter though - the dyno is going to give 100 times more useful data mainly because it's real. We could do stock base runs, then boosted runs, then play with the trumpets and then the headers and then add cams. The amount of data we are going to get is massive and it's going to be very useful. I actually have a basic excel simulator (which I've calibrated) with both dyno runs and drag passes that will give you airflow + power based on engine VE so we may be able to enhance this nicely and share with all with a new EG33 version.

Bazza,

I work with NASA scientists whose careers have been wrapped around modeling the various aspects of the Earth and one of the things I've learned is that every computer model/simulation has its own "quirks" and that they all benefit tremendously from the input of real data ("Cal/Val"), be it from ground/air/space-based assets. I look forward to seeing how your simulations and Harvey's match up with/benefit from the input of dyno and drag passes!

Things are starting to get interesting, aren't they!

Thanks for all your hard work and efforts. I've found myself wondering how much it would cost to fly down to Australia and take an extended "SVX racing" holiday

Bill

[Dessertrunner]

Hi Bill,
I have a big house no kids plenty of spare rooms so you are welcome any time to come and stay. Plenty SVX things to bury your head in.

You got me thinking so I will move this to a new thread.
Tony

[icingdeath88]

How beneficial would it be to run a pump like this pulling through a pickup in the stock oil pan, to a reservoir feeding the stock oil pump?

Kind of like this: http://www.subaru-svx.net/forum/showthread.php?t=60915

I could cut off the end of a stock oil pickup where it connects to the pump, and have it welded through the side of the oil pan, and put fittings on the cut end somehow, and have another line that returns oil back to the stock pump.

Like this picture:


What do you guys think? Stupid idea trying to cheap-out on a dry sump?

[Dessertrunner]

Changing to that pump won't give much benfit and will creat to many headaches. The concept is good but,
To get around the cavitation problem you need to have a seperate oil tank and to do that you have to have a minium of 2 stage pumps, one suction one pressure. The other issue is there is not much room between the front of the block and the timing belt to get the suction past everything.

For the mid engine car I am building I have lowered the engine and gearbox by about 100mm and also aim to lower car by about 50mm. My aim being to drop the CG of the car lower.

Okay Bazza and I have been having a debate about the dry sump. I personally think that we should use a 2 or 3 stage pump and forget about pumping oil back from the heads. One of the stages needs to be pressure.

In the mid engine thread I will post a image of the engine block sitting in the subframe with the new engine mounts etc and you can see how low it is, not sure were I will discuss the simpler dry sump set up but will explain it in the near future.

Tony

[Dessertrunner]

My question,
I am flat as a lizard drinking and don't have time to look so can anyone tell me the following,
Size of standard injectors?
Fuel pressure in PSI?
If possiable maxium duty cycle of our injectors?

Other updates,
- Rear drive lock for 5 speed transmission for dyno arrived so I have to get on that soon so we can run some tests.
- My mind is thinking and I wonder how much pain a standard SVX engine can take, in other words can the standard valves, pistons & rod run a lot higher revs that we have all assumed possiable before. I need to find out so it maybe time to distroy a engine.
- Anybody know how many stage pump in a porsche engine and if some one has a oil flow diagram it would be very help full.


Any ideas on engine loads for a turbo engine, in other words its a 3.3l engine what boast number do I need to get to acheive a 6.6l equlivant. If this sounds stupid, go with me for the minute.

Tony