the airbox is great.

 

Also I have one of those CF airboxes for sale for TWO MILLION DOLLARS!

 

I'll give you about tree fiddy

If/when you disassemble that oil feed, let us know what you find. If you pop the head, please measure the restrictor in the block.

 

Ive never seen one of those airboxes trade for under 2k

Dann

 

I've got a '97 block sitting under my bench, I'll dig it out tomorrow and measure the restrictor and report back.

 

Ah man, I was just discussing this idea with the old man on our build. Of course we had the exact questions, comments, concerns that have already been discussed here.

Really interested to see what comes out of this. Thanks for the effort and info!

 

At least on an NA6, do it every time.

Dann

 

Got a chance to drag the spare '97 block out today, the restrictor measured .094"... same as the '99 measured earlier. Doesn't seem like they changed anything between HLA and solid lifter blocks. Now we just need someone with an '01+ block to measure.

 

So if this restrictor is .094" across the board, is this a good thing?

Is that enough to feed a turbo and keep the head lubricated?

A turbo oil feed line seems pretty big in comparison, but the turbo has a restrictor as well correct?

How much oil actually flows through a turbo anyways?

 

A couple of litres a minute is plenty.

Dann

 

This post will likely be lengthy. I have my tests run, and this will include the results. I will also discuss how we might utilize the information, and give some subjective opinions (redundant, perhaps).

I came off the rear head port with a 1/8 bspt to -4AN fitting, then to a 20" long flex. I flushed this into a cup. Didn't see any contaminant, but I didn't filter for it either. Next I attached a gauge and got these readings:
30 psi @ 1kRPM, 40@2k, 44@3k, 44@4k. Unfortunately I did not look at CLT.

Next I added a Tee, right at the head fitting, and ran a second -4AN of 36" length to a -4AN to 10x1.5 turbo inlet that has a 0.060" built in restriction. With this flowing into the fill hole in the head, I got the following:
19@1k, 32@2k, 36@3k, 36@4k. Roughly, the flow rate measured 1.0 L/min at idle

Then I added my TDO4 (WRX) to the end of the long line:
20@1k, 31@2k, 34@3k, 35@4k. Roughly the flow rate measured 0.32L/Min at idle. See attached You-Tube video of that flow at idle. https://www.youtube.com/watch?v=5AGt...ature=youtu.be

Later, I realized that I should have waited until the oil was at full temp to run the test. So, with CLT at 195F, I have, with gauge only:
17@1k, 26@2k, 29@3k, 30@4k

With the 0.060 restrictor at the end of a 36" long -4AN (note, the gauge is at upstream end):
10@1k, 18@2k, 21@3k, 22@4k

I did not repeat with the turbo.

Never, in these tests, did the valve train sound starved.

For me, I will refer to the attached MHI sheet, compliments of someone on a DSM forum (they and the Subi forums have this same discussion, by the way):

What I think I am seeing, is that, with the TD04, I should barely be able to run the head port, without the added restriction. The flow with restrictor and turbine was low, but the pressure is within the spec.

Kind of like this:
At no flow: Idle, 17psi = 1.2 kg/cm^2 > than 0.8 minimum
At no flow: 4K RPM, 30 psi = 2.1 kg/cm^2 which is right at the lower limit of 2.

At higher than expected flow, the pressures will be lower. However, in practice, I think not much lower as the flows with the Turbo attached were considerably less than the restricted line without the turbo.

That is what I plan to do. Run from the head port, -4AN of about 20" to inlet of the TD04, un-restricted. It may be a month, but I will report back.

From what I gather, if one is running a bearing unit, the head port should give enough pressure for a low flow, and a 1mm inlet restriction.

Unfortunately, the data was marred by not all of it being taken at full warm-up. But the primary take-away is the 195*F, no flow readings. If that is enough pressure to drive oil into your turbo, and EDIT: if the flow is not too much (EDIT: LESS than the usage of the 0.060" open restrictor), then you should be OK.

 

FWIW: The initial diagram posted by Eo2K, is a B6. Cam angle sensor pickup on the intake cam. What is mildly amusing, is the listed oil capacity of 3.8 quarts. IME, B6s take about 3.8q, although 3.6 is frequently listed. Ditto for the BP, 3.8q listed, 4.0q is closer to reality, on a level surface, filling to the top of the "F" line.


Unused oil/coolant ports on the B6/BP (and most engines for that matter), are just that, unused extras. Not surprising they're not included on factory diagrams, as they're essentially not there, as far as the factory is concerned.




Cool idea though, I like the idea that the boys at mazda already used this for a turbo setup. Even from a "low production, high output" setup, the fact that the factory used it makes me feel as though it is a legitimate solution. UOA would be great, but someone has to do it first.

 

I'm kinda confused a bit with that post. Will re-read many times over to hopefully understand correctly.
What I'm getting is pressure at the back of the head dropped roughly 20psi after you tapped the back of the head with a .060 restrictor at the end?
But did your overall engine oil pressure drop after you added the t?

 

The first numbers are dead headed on the head port. The second numbers are open flow through the restricted hose. The third with the turbo on the end of the restricted hose.

He lost 8psi hot or cold (in the head) by opening up the flow to the restricted line or turbo (at 4k).

 

Yes, you understood correctly.

I did not have a gauge on the main gallery (Oil Pressure Sensor location). However, the change there would be less with this set-up than with a block feed at the same turbo oil flow, as the head port is downstream of the now famous 0.090 restrictor.

The question to be resolved is only the starving of the valve train, not the main and rod bearings.

This is my generalized take: Mains and Rods have tremendous pressures due to reciprocation and combustion. Cam bearings have considerably less as it is the forces of the valve springs (and some inertia) that they must support. Turbo bearings are the least as the forces are somewhat balanced and the shaft simply spins and transfers torque from turbine to compressor.

What my test did show, was how low the head pressures should get if the turbo oil feed was fairly high (higher than needed in most cases). What I do not know is what kind of oil pressures are needed to support the cam bearings, and be able to squirt a little oil onto the lobes.

Though we did not hear any indication of starving (clatter), I don't know that this tells the whole story.

 

And I don't understand why there was so little difference between the PSI with and without the turbo, as the flows were so different. My assumption is that the oil got hotter as that part of the testing proceeded.

However, I believe that your explanation is exactly what I wished to convey.

 

Very cool, thanks for that.

I thought valvetrain starvation was the whole point of this test, apparently not.

But you did prove that there is a very significant reduction in pressure with the turbo feeding off the head rather than the block, so that kinda reassures my paranoia from earlier.

Though I'm not writing it off still obviously.

I wish there was a sure thing answer to this issue lol

 

There is. VVT pump makes everything better.

 

On a regular bp running it sure, but on a vvt car?

 

On an unrelated note, I'm a bit worried now since I planned on sourcing my AN-6 feed from a oil filter sandwich adapter.


...looking at the diagram, that's pre-filter and any junk floating about goes straight to the turbo