Shoptalkforums.com

Hopefully someone will log on and tell me how to post these pics. I tried by clicking on the ADD ATTACHMENT box, but noting happened. Anyway here's the text to go with the pics.

Here is a cutaway I did some time ago. I used this info. to resolve a couple of issues including, but not limited to:
1) How deep heads can be flycut before hitting the exhaust stud. This head had a damaged stud boss that someone else performed a piss-poor repair on. I therefore welded the boss shut and remachined (less tapping) to O.E. depth before slicing. From this I concluded 2mm's is about the limit without welding and machining for a shorter stud. That leaves about 2mm of material. I have since come to the conclusion that some heads have suffered core shift and this can lead to even thinner cross sections. I have busted through at 2mm's on a few heads.
2) How large the bowl area can be opened up before busting through the casting. This was a necessary step in determining how large of an intake valve can be fitted to these heads without making sacrifices to the bowl profile, which is very critical and not very forgiving. This photo is of the bowl opened with one of my custom cutters to the absolute limit of reliability. A 45mm valve will work fine with this profile. The casting will not support a larger valve, the material simply isn't there. And even if there was, when you move upstream there isn't enough material to open the port. An effecient port cross section is around 80% of the valve diameter. D-ports help cheat this figure, but since the bowl can't support the shape, what's the point?
If you look closely you will see that the turning point of the port is moved considerable away from the valve in the modified bowl, as compared to the stock bowl, and left sharp. Mixture quality is greatly improved, as is flow with this shape, as turbulence, and all of its pitfalls, is virtually non-existent. This is the shape on the heads (44x38) Jake used to make nearly 180HP from a 71x96 with 10.0:1 on pump gas with 20 degrees of timing! And that engine has a very flat torque curve to boot! On racing gas with more timing this engine could approach 200hp.
Don't have any cutaways of the modified ex. port. but my approach there is about the same. The difference being that I don't lengthen the long side as there are no concerns of storing fuel charge for the next valve event. However, the same effort to reduce port turbulence, by keeping the short side turn sharp and crisp, is there. The only thing I do to the turn here is to KNOCK OFF THE SHARP EDGE BY HAND WITH A PIECE OF EMORY PAPER. I do this to eliminate stress risers that can lead to cracks.
There you have it.
Len Hoffman
Owner
Hoffman Automotive Machine Inc.
[/url]

Len, email me the pic and I'll have Kathy post it for us....

Here are HAM's pictures

I guess the other two photos didn't make it.
Len

My mistake. All three photos are there, albeit a little blurry. But I think everyone can see what's going on.
Len

Wow, very informative. Not sure if I understood everything, but the pictures really enhance your words well.

One Q tho: The port seems to have a really nasty narrowing just after the intake valve/guide. Why isn't the port opened there ?
Same goes for the exhaust port. There seems to be enough material there to do so.

As with respect to the 80% guideline you mention: The stock 2,0 liter intake port has a diameter of 36mm. 80% of this is a 45mm valve, so a 44mm valve would indeed not need that much (if at all) trimming of the intake port.
I have seen many 914 heads tho with the intake ports opened up to the max, which would be about 38/39mm without welding.
Now with such a port, applying the 80% would indicate that a 48mm valve could be supported well by that port (if the meat would be there...).
Make no mistake that I fully subscribe your believe that a larger valve/seat than 44/45mm will weaken the head!
Just an observation.

Thanks Len!
Best regards,
Walter

Walter the narrowing you're referring to in the intake port is the guide deflector. The port doesn' t really narrow as it appears in the photo. As for the exhaust that is a stock exhaust. The narrowing there is the guide boss. The port actually flares out around the boss. On high rev engines I knock this boss down and round the guide.
The issue, as far as intake valve sizing, as limited by the intake casting (without even getting into the limitations of the exhaust and how they effect intake sizing. If you can't get it out, you won't get it in!) are that even though tricks can be performed to size the port up to 80% of the valve for 46mm-48mm, the bowl area directly beneath the valve just doesn't have the material to form a proven, effecient shape. Look closely at the modified port. You notice how the port transitions nicely into the machined bowl area? Now notice that the cross section thickness of the bowl wall is uniform and pretty darn thin. Take out any more material (which you have to do for a larger valve) and you run the risk of busting through the wall. Throw in a little core shift and you are taking a hell of a risk to shove a bigger, heavier valve that the exhaust can't handle anyway. I've said this before, and I'll probably say it a thousand times more. The bigger is better crowd is the easiest to beat on the track and easier still to beat from red light to red light on the street.
Len

HAM,

This is great information. Thanks for posting it!

Listen to HAM... Nothing so far has touched his work on my dyno- amazing.

So let me make sure I understand---You suggest NOT blending and leaving sharp transitions, in order to _reduce_ turbulence?

On the intake I can see it causing an *increase* in turbulence, possibly improving the mixture, thus the low required timing, at least partially.
(Don't know what your doing in the chamber)

No arguments on the valve sizes...

The exhaust port critical flow is sometimes supersonic, so a "sharp" edge on the short radius should not be an issue, and might be a feature.

I will never argue with "works" on principle... But I may be a wee bit confused.

The chamber shape was stock 2.0 914. Turbulence in the port is bad for mixture quality, velocity is good. Turbulence is caused by obstructions, sudden changes in cross sections, big differences between the length of the short side and long side,and some other details I won't go into here. Turbulenceand pressure gradients in the port go hand in hand. If a gas charge, or slug, is not capable of following a sudden kink in a port elongating the short side will not help it follow the bend. Think of a swift moving stream that comes to a sudden bend. Erossion will cause the long side, or outside, of the turn to open up. Now the creek is suddenly wider in the turn. The water on the inside moves much slower and the current is pushed away from these edies. In a port the same thing happens. The slow moving area to the inside creates a pressure gradient. This draws the vaporized fuel, which is heavier than air, out of suspension.
Engines are full of compromises. Just the fact that there has to be any curves or bends in a port is less than ideal. Consider that a fast moving exhaust charge is capable of moving down a conduit through a maze of manderel bends without losing much velocity in the bends. The trick in a port is to work all bends, short or otherwise,around the same tangent. Sometime this means making a sharp turn. And if you think the turn in a typeIV head is harsh, just look at what the smallblock chevy guys have been up against all these years, intake and exhaust. Also consider that a 2.0l twincam Alfa with 40x44 valves makes an even harsher turn. Check out the short side of one of those ports if you get the chance.
Hope that clears up some things.
Len

sweeeeeet .

ive been missing some crazy stuff again i think this is winning imformation and i have total respect for the person willing here to share it with the world respect to you some top tips there magnificant.

Thought I'd bring this to the top...



Don

Belongs in a "best of the STF" stickie.

I think we should add link to this thread with cutaway pictures of other T4 head's
viewtopic.php?f=1&t=71523&hilit=cutaway