An In Depth Analysis of The MZR DISI Head

[Mazdazilla6]

That's usually the case with most applications. Duration and lsa changes will almost always pick up more gains than just lift. But there's never a drawback to lift other than needing proper clearance/upgrading parts.

Lots of people advocate that you shouldn't run cams with overlap on turbo applications because you "lose boost". The reality is that you won't suffer from overlap on a turbo application until you run LOTS of overlap. What you end up doing is sacrificing efficiency of the turbo in order to gain more flow. By spinning it faster you can hold your pressure ratios the same as you did before but since wheel speed goes up efficiency goes down. In the end you'll still pick up sizable amounts of power. Superchargers on the otherhand are a different story because their speed is fixed so any intake charge pressure you lose out the exhaust can't be gained back.

[Enki]

True but we have something else to worry about: Spraying DI fuel directly into the exhaust stream. That's the main reason why I've designed my cams with the smallest amount of overlap possible; not really sure when the injectors start the firing window so better to err on the side of caution.

[Mazdazilla6]

For your application that's understandable. I won't be spraying that late since I can get enough fuel from my PI and 6th port.

[Enki]

I'll be using 6th port myself, but on full corn it will get to 100% IDC pretty quickly and that's why I'm concerned about spray window.

[Mazdazilla6]

I think you might be okay if you run some large (2200cc) injectors.

Note: might = total guess off the top of my head.

[Enki]

I meant spray window for DI in relation to cams and shit.

[Mazdazilla6]

Yea I meant if you run large enough injectors on the 6th port you might not have to spray super late.

[Enki]

Right, I get that. I already have a pair of 1500cc ones which should be more than enough for my goals.

[Mazdazilla6]

Post #2 updated with exhaust flow numbers. Will add examinations/notes/etc at a later date.

[Enki]

Lol exhaust flow numbers lost 30+ hp on the sim...very sadface.

[Enki]

Averaged flow data for those that want it:

[Mazdazilla6]

Added wall of text to post 2.

TL; DR: Be happy with your stock head. Or don't be and actually read what I wrote.

[Enki]

So the TLDR for headwork is dump the bump on the intake side, go full retard single runner if you want 4 digits and gasket match the hotside....Sound about right?

[Mazdazilla6]

Originally Posted by Enki So the TLDR for headwork is dump the bump on the intake side, go full retard single runner if you want 4 digits and gasket match the hotside....Sound about right?

I don't want to jump to conclusions yet, but for the most part yes. That bump on the intake side may not be as detrimental as you think. What really chokes it up is the sharp cliff on the short turn. Also, keep in mind that single runner stuff isn't just for uber high power builds. Going single runner reduces pumping losses with minimal loss on intake charge velocity. As for the exhaust side that's pretty much right, though I'd use the term gasket match lightly. Depending on your manifold design, not all 4 ports may want the same exit shape.

[Enki]

So the small radius in the intake port to valve area is probably the most important part to smooth, not the big lump that looks like it's in the way of everything. Got it.

Also anyone running a FR mani has a gasket matched mani already, and could make use of a gasket matched head. Sorry for being vague on that; speaking generally about what I know about (my build, similar parts on other cars, etc).

[Mazdazilla6]

Originally Posted by Enki So the small radius in the intake port to valve area is probably the most important part to smooth, not the big lump that looks like it's in the way of everything. Got it. Also anyone running a FR mani has a gasket matched mani already, and could make use of a gasket matched head. Sorry for being vague on that; speaking generally about what I know about (my build, similar parts on other cars, etc).

Yes. The short turn is always the most important and sensitive part of a port.

The reason I say lightly is you may not want to exactly match the gasket. Depending on how your runners are angled in the mani inlet will determine if it'd be more beneficial to gasket match a portion of the head into order to direct exhaust flow more efficiently. Instead of thinking of it like matching up the ends two pieces, think of the head and mani as one whole piece and what its most efficient shape would be for the runners.

[Enki]

Pretty sure they are straight, with bends shortly thereafter. Not angled.

Edit: At any rate, it's too late to change shit as far as that goes.

[Mazdazilla6]

More results, gasket matched the exit partially with a tad bit of floor work. This is what I'd consider a basic port job for our head. This is all of the low hanging fruit on the exhaust side, now I'll be going a bit further and trying to get flow @.400" as close to 200CFM as I can. I've got a couple more tricks that I think might work but we'll see. After I'm satisfied with the exhaust side I'm going to go back to the intake and touch a few more spots and then probably do the real deal on my usable spare head. I might try to quantify gains from a 5 angle but may not have time for that.

Capture.PNG

Capture2.PNG

[zenit]

Originally Posted by Mazdazilla6 I might try to quantify gains from a 5 angle but may not have time for that.

Do your career racecar engine building buddies have any opinions on the longevity of 5 angle vs 3 angle job?

[Mazdazilla6]

Multi angle seats won't affect seat life as long as you aren't reducing your seat face width or running an interference angle (ie seat face angle of 45° and valve seat angle of 46° etc). IIRC as long as seat faces are .020" or greater they'll last a while, also depends on the size of the valve. Larger valves need larger seat faces to avoid premature wear. Keep in mind there's other ways to pick up flow in that area like putting backcuts on the valves or changing/adding the throat angle.

[Enki]

Interesting; I as told that doing a gasket match on the hotside won't net any flow at all.

Edit: Is the gasket match data for gasket match only or on top of the smooth + thin data?

[Mazdazilla6]

Originally Posted by Enki Interesting; I as told that doing a gasket match on the hotside won't net any flow at all. Edit: Is the gasket match data for gasket match only or on top of the smooth + thin data?

Whoever told you that either has a very poor understanding of how airflow works or didn't even spend 10 seconds looking at the head. Stick your finger in each port and feel the first 1/2 inch of the port all the way around. It'll become pretty obvious how gasket matching that area picks up as much flow as it does.

On top, done in stages, so the chart I posted reads chronologically.

[Enki]

Well shit, it was someone known for their headwork (I guess) that completely ignored the fact I wanted the head gasket matched the first time and shipped it off without even touching the exhaust side.

[Easter Bunny]

the answer is ALWAYS gasket match.

[Tokay444]

I would have assumed so as well.
I guess I should have specified and paid extra for it.

[Mazdazilla6]

I haz teh cutaways. Pics to come after I get them cleaned up.

[Atomic41]

waiting news from you sir

[Fstrnyou]

2 years later and no cutaway pics posted. I guess I’ll post one then. It’s of no significance though. The cut isn’t aligned with any particular port so you can’t really get a good idea of the shape like you can the mold created by Mazdazilla6.

As mentioned earlier in this thread, the intake ports are angled, which means the cut would need to be equally angled to get a good cross-sectional view of the port.
With the exhaust ports merging to a single runner, it would be even more difficult to get a cut that reasonably represents the port shape.

[Fstrnyou]

Back again and hopefully I won't muddy up this valuable thread.

In the OP, @Mazdazilla6; said the testing was done at 25" H20, but @Enki;'s summary of the flows shows 28" H20. I'd like to think the screenshot "should" say 25".

Okay. I did some porting to my head.
Intake: Mostly just cleaned up the cast roughness in the runners, did a mild blend and rounding of the intake port inlets because i'll be going to a FoST manifold which is undivided, and I removed quite a bit of material out of the bowls..mostly the small runner bowl. Considering I didn't open up the runners at all aside from smoothing the rough cast and the flow went way up, I can only assume that the velocity increased as well which is fantastic!
Exhaust: same rough casting cleanup work without any heavy removal in the bowls other than smoothing out the transitions. I did go with +1mm exhaust valves but it was for nothing because i didn't open up the throat just below the valve seat. I really don't feel like taking the head apart to do it now. It's all pretty from the machine shop. Looks new.

NOTE: I need to confirm what vacuum value the flow test place used. I'm guessing 28", but I need to be sure. I found something online that says how to convert my new flow numbers to 25" and it claims to be mostly accurate, so I'm posting that graph with that assumption.

Edited to add pic of clean and assembled head.