Mike, all,

Edited:


The port goes flat at 0.25", but at vastly higher flow rate. I think this shows that we are opening the port up to support what the valves are capable of.
And perhaps we are limited by the flow capacity of the valves themselves, at 0.25 inch opening.

I seem to recall that valve flow reaches it's max at 0.25 X valve diameter. After that you can open it up all you want, it still don't increase flow (but it reaches full flow earlier).


With 28mm valves, this means 7mm opening.
At 7mm opening, the curtain area amounts to 0.95 square inches.
One square inch can flow 110 CFM max.
This means that each valve can flow max 105 CFM @ 100% flow efficiency.
Set the flow efficiency to 80%,( I hear that is a reasonable and respectable number), then expect the valve to flow 84CFM.
2 valves flowing 168CFM expected, at 7mm opening.

This corresponds well with my latest measurement, 168 @ 6.35mm.

If this holds water, then the focus should now shift to working the valves (back cut) and the seats (multible angle throat and deshrouding) to try and approach the 100%.

Are there any obvious errors in this assessment?

Cheers, Finn Hammer

And this is why math is power , Stay in school kiddies ! I so enjoy reading their work. Another blue skies , 60+ day here. Heading to the foothills to find a victim for my 93 mph wonder , I’m gone

Here is a look at the exhaust ports from the header flange side. Discussion 1 post down

Wow that's a big improvement in flow.
Tanks for sharing your photos and findings so extensively.

I'm wondering if you could round the valve guides or even profile them to get more flow.

And is it possible to remove them before you start milling and put them back afterwards?

Finn, I'm lovin your results.  I think I'm gonna try your configuration on the Hammer Head.  Try to do a comparison of my current flat-floor D port to your geometry.  

My most recent exhaust port flowed 130 cfm at 15".  If I apply the correction factor (.732) to your 28" data I get 138 cfm.  Your corrected data indicates that your configuration flows about 8 cfm more with a smaller outlet.  That's worth lookin at.  By any chance did you test it at 15"?  

I should also note that your clean up around the guides and guide bosses looks a lot better than mine.  That's a very difficult area to access.  Nice job.  You noted that you enlarged the seat throats a bit.  Did you happen to measure the seat throats.  I've been contemplating taking the throats to 88% (.970").  How big are your seat throats?

Are you planning to measure the port volumes when you are all finished up?  The stock int/exh ports are about 112cc/66cc respectively.  I usually end up around 103cc/71cc when I'm finished.  I'm thinkin your exhaust port might have a bit more volume.

Thanks for posting the pics and graphs.  Very informative.  Very helpful.

You need to be concerned about the special head stud that runs up through the floor of the exhaust port.  The head loses a lot of rigidity when you enlarge the port.  The loss of rigidity combined with the elevated temperature causes the internal threads to yield.  The stud slowly pulls out over time.  You can eliminate the problem by replacing the special stud with a threaded insert that installs from the top, through the exhaust port.  It's fairly simple and completely eliminates the problem.  This old post provides details.

http://suzukisavage.com/cgi-bin/YaBB.pl?num=1665791582

Really nice flow numbers Finn.  I find it gratifying to see how the two benches compare.  If you ignore the actual flow numbers and simply look at the percentage of flow increase, we seem to be very close.  

On the head I recently modified, the exhaust flowed 27% better.  I didn’t enlarge throats by any appreciable amount, just what was necessary to blend the seats in.  Your second exhaust attempt was a similar configuration and you improved it by 27%.  

Then you opened up the exhaust seat throats to .940” (about 1mm) and your numbers improved to +40%.  My Stage III pumped flow up to +39%.  Very close.  I think you have found my lost flow.  I must have overlooked the seat throats.

When I encountered the mysterious loss of flow, I did a cross check using the HammerHead.  I assumed I had configured that head identical to my stage III.  The Stage III is currently on my engine, and I wasn’t about to rip it off just to see what the exhaust was flowing.  Since I’m sure that the former flow data came directly off the Stage III, I bet I opened up the throats on that head.  Time to test that on the HammerHead.  Totally cool, so glad you are doing this project, and very appreciative of you sharing the data.

Those seat throats look like they are critical.  I also have pondered the concept of choking the inlet tract just before the seat.  I have read and viewed supposed experts claiming that it creates a venturi or nozzle that accelerates the flow.  I’m not sure.  Eric Weingartner discusses throat size a lot, but he generally relates it to leaving sufficient material to cut the various angles below the 45.  I can confirm that if you get carried away with the throat, there’s nothin left for the 75 and the 60-degree cuts.  But that’s a bit off topic for now.  The exhaust port is the current topic.

I recently made a cutter to allow very accurate enlargement of the valve seat throat, but that cutter is specifically for the intake.  I think I need to come up with a similar cutter for the exhaust seats.  Then I want to try and take those seats right out to 88% (.970”).  And as you stated, pay particular attention to the back cuts on the valves.

I’m in the middle of a project to salvage a head with wiped cam bearings.  It is proving to be quite a challenge.  After this project is complete, I will be breaking out the flow bench.  Can’t wait to see how you do with the intake port.

BTW, I recently purchased a set of Honda 34mm intakes off eBay from an outfit called "Getor" (located in Germany).  The valves were reasonably priced, and the shipping was cheap and fast.  They appear to be high quality Japanese replacement parts.  The stems are a little longer than stock LS valves, but I've been running these Honda valves for a long time with no adverse effects.  The eBay item number is 394193921223.

Finn, I assure you that the stud that runs into the intake port is strong enough.  No need for any special treatment on that stud.  I do use a slightly longer stud so that it runs up through the epoxy to provide an anchor, but the M8 x 1.25 threads are more than adequate.

The exhaust side is subjected to extreme heat.  Aluminum loses strength at relatively low temperature.  That's why the factory used a special stepped stud.  The M12 threads afford significantly more shear stress area so they can get away with the reduced material strength at higher operating temps.  But when we open up the port, we reduce the rigidity of the head which results in increased stress at the center fastener (the 8mm stud).  Bump up the CR and the situation gets worse.  It's not the cyclic heat, it's the cyclic stress combined with reduced strength.  The aluminum threads give up.

I believe you will find the High Heat putty much easier to form than a paste.  Standard JB Weld paste settles and runs.  It's great stuff and even has a higher temperature threshold, but it would prove difficult to form into the basic shape you are shooting for.  Mineral spirits (used sparingly) helps to keep the stuff from sticking to your finger as you shape it.

Win, lose, or draw, I intend to write a report on the cam bearing fix.  It's just too interesting to pass up.  Gotta share.

As you have found out, it is not all that impossible to fill that void, from a welding tech. Point of view.
But the question, as I see it, is: would it help?
Personally, I would prefer to weld on the other side, fit shorter springs etc. and then remove the ridge in the port instead, producing a better port top surface.
Welding on a head seems risky to me, I could imagine all sorts of problems arising from misalignment due to warpage etc. I am not going to do it.