Wannabriden's D Port Re-Port

Been a minute since the last update.
Been busy though and getting a few things done.
Wife's eye is improving on track and should soon be healed.
Unfortunately she tripped and fell while working in her garden.
Twisted her knee. Had the MRI on Friday. They will read it on Tuesday.

Still been getting plenty done in the shop.
I've opened up the junk head.
I fully tested it just Raw with the initial grinding.

Then I did stage one modification on the junk head.
It was fully tested again.

Then I have done a stage 2 modification to the junk head.
It was fully tested again.

I have just completed the 3rd stage modification to the junk jead.
It has not been tested yet. Waiting for the clay to dry.
A video coming shortly covering the transformation.
Takes a minute to assimilate all the data.

The stock head has had a pocket port done as well as valve work.
It has been tested in all iterations...
Completely stock w/lapped valves.
Casting flaws removed (pocket port) w/ new cut 45* seat/valve.
Casting flaws removed (pocket port) w/ 3 angle valve job.
I'll be reviewing those results in a posting in the next couple of days.
You might be surprised... or not.

Testing welding {brazing) the aluminum this week.
Have a few different options to compare...

Lastly, I have been working to reduce the margin of error when testing the pieces.
After some great conversation and guidance from an absolute guru of flowbenches;
I was able to bring the bench to nearly absolute.

As a result of trueing the bench to actual cfm flow it lowered my numbers.
As I stated from the beginning it is about getting the best, most accurate possible.
It is also extremely important to note that the only real value in the numbers is the ability to compare them percentage wise.
For that reason previously posted values still have value despite being inflated.
However they should not be compared directly (without correction) to the numbers used going forward.
Most important should be the understanding that the cfm flow numbers are perhaps some of the least important pieces of information regarding a heads performance. Unfortunately it is the number most relate to. If you have come this far down the rabbit hole you understand.
Anyway I have the Bench Recalibration Video attached if you are interested.
Reports on the fun stuff coming soon.



  • Orifice Plate Specs.jpg
    Orifice Plate Specs.jpg
    83.2 KB · Views: 45
Welcome Back...
Welcome Back.
Been more than a minute.
Unfortuanately I caught a bug and have been out of action for a bit over 10 days.
5 of those were spent bedridden. The last 5 spent regaining my strength.
I'm now about 90%. Anyway, just an update of sorts for those that are folowing.

Basically I discuss some of the standards created by the group at SAE.
We have already discussed many of "formulas" that have resulted from their studies.
If you are not familiar with the organization and what all they do I would encourage you to google.

I explain some of the ways we use the data and information gathered as well as how I go about setting "targeted" values.
It's not too deep, but it can perhaps get a bit confusing. You can always message me for any questions.

I also give an overview as to some of the modifications that I have performed utilyzing the junk head.
It has gone through several morphs at this point. Complete testing has been completed on each iteration.
Still very much data gathering although I am now formulating the basic port design from the data gathered.
Testing has shown a lack of potential being reached once getting above .20 lift.
I believe that is a result of the guide and guide boss not yet being addressed.
That is the next step in the testing process.

I will be shortening the guide per Jacks specifications in his D Port thread.
Obviously the guide boss will be addressed, as well including optimizing the shape once it has been sized.
It is this obstruction that is creating the bulk of the issues with the medium to high lift flow.
This modification will be performed on the junk head as well as both of the stock ports I am testing simultaneously.

I will then move to tuning the CSA (cross sectional area) and readdress biasing the SST (short side turn).
Lastly I will also be looking into unshrouding the valve.

It is my hope to have a heavy mod version and compare it to minor mod versions which stay nearly stock.
Obviously all of the above will be compared with straight stock versions as well.

Enough jibberish.
If you followed all that, the video is probably not necessary, but it is included nonetheless.

Last edited:
That must have been a nasty bug, good that you are up and about again.
I find you videos very useful and always learn something, thank you.

With regard to shortening the valve guides. I have been struggling a bit to work out if Jack shortened both the exhaust and inlet valves the same.
From the photos in his thread the exhaust is shortened as he described but the inlets look like they are shaped like the Lillie replica guides.
What is your understanding of the modification he did to the inlet valve guide.
Hey Signal,
Yea it got me pretty good.
Much worse than the one that shut down the world did.
I am happy you are able to get something from the postings.

As for shortening the valve guides....
VERY good question.
Jack was very detailed on his treatment for the exhaust guides.
Not so much on the intake guides.
In his thread posting #49 he talks about it,
This bit of guide information was given to me many years ago and have held off given it out untill Kibblewhites cloning of Lillies guide dimensions.
For those using stock guides,you might find it useful but I don't cut my exhuast guides like whats given in this document. Looking at Lillies E/G mods,I'm lost as to why he machined them
that way ,can't image he raised the port roof that much,

That seems to indicate to me that he had the information that Lilli used in profiling his guides.
The way he stated it made me think he followed Lilli's intake profile as he provided the information but made no other mention regarding the intakes.
He makes specific mention that he profiles his EXHAUST guides differently than those of Lilli.
He then provides an amazing how to on his method of modifying the exhaust guides.
As you state it seems to appear from those pictures that the guide does indeed have the Lilli profile.
However I can't say for certain. Perhaps Jack will clarify if sees this.

I have the D port that Hugh modified (supposedly following Jack's guidelines) which is MUCH different altogether.
That head has the intake guide shortened as shown in the attached picture. It is the guide labeled "modified" at the bottom.
As a reference the guide was shortened 18% from stock compared to shortening the exhaust guide 13% following Jack's guideline.
Most definitely reduces the obstruction area inside the port at a high flow area. The question becomes how critical that extra 5% is structurally.
With the indications that this head was on an engine that was ran hard I can say there does not appear to be any wear or play between the valve and the guide. Of course I have no idea of the mileage involved. The Lilli profile extends into the port a bit farther and reduces obstuction by reducing the diameter of the guide portion that extends into the port. There were no exhaust guide modifications performed on the head in my possesion.

As I have these shortened guides already I am hoping to test them as well as a Lilli profile which I have yet to cut.
I have attached some pics of the D Port modified guide/guide boss to illustrate the approach used by Hugh.
I also included a picture of a stock guide boss for comparison.

Hope that helped explain my interpretation (although I could be completely wrong) of what was done.


  • 20230217_203540.jpg
    135 KB · Views: 56
  • 20230217_203740.jpg
    168.3 KB · Views: 52
  • 20230609_151555.jpg
    174 KB · Views: 48
  • 20230609_151744.jpg
    126.1 KB · Views: 56
  • 20230609_153028.jpg
    138.6 KB · Views: 53
Last edited:
I've often found that sometimes charts are easier to grasp that simply posting numerical values.
I've picked up on some basic spreadsheet understanding to help present some of the findings as I go through this project.
More importantly it allows me to simply enter the data once and it performs several different calculations for me.
Easier and less chances for errors. Win, Win.

One of the key tests I perform would be what I call the "Valve DP" test.
I use a valve that allows me to take a DP (differential pressure) reading at any position.
I originally only tested at the SST (short side turn at intake center), Plug (90* from SST on plug side), LST (long side turn 180* opposite the intake center), and lastly the Inside ( 90* from SST on inside wall opposite the plug).
I then started testing at 8 points adding those that are centered between the original 4.
Think N-NE-E-SE-S-SW-W-NW. I find the extra data useful although it adds to the workload.
I am not going to bore you with all the numerous tests I have performed but have decided to give a glimpse.
This is a randomly selected modification to illustrate the data and I think you can get a picture of how it can be used.
This selection is from the Mod 2 which I gave a brief overview of in the previous video "Let's Catch Up"
Obviously the purpose of using the Valve DP testing is to be able to see how the flow is distributed around the valve.
The attached pictures are all of the same modification at different lifts. The lift values are notated on the picture.
Besides the visual representation I included the percentages for each slice. Keep in mind that the percentages are rounded.
It is also important to understand that the DP reading noted can NOT be corrected to CFM.
It is meant to provide a visual representaion and the values are for comparison only.
As a bonus I included a graph of the same Mod 2showing all 8 positions at .30 lift to give you an idea how that works.
I did not place any values on that graph but did label the Big 4 (SST-Plug-LST-Inside) to help orient you to what you are looking at.
As always I am happy to answer any questions.


  • Mod 2 .1.jpg
    Mod 2 .1.jpg
    24 KB · Views: 38
  • Mod 2 .2.jpg
    Mod 2 .2.jpg
    23.8 KB · Views: 36
  • Mod 2 .4.jpg
    Mod 2 .4.jpg
    25 KB · Views: 36
  • Mod 2 .3.jpg
    Mod 2 .3.jpg
    25 KB · Views: 36
  • Mod 2 All 8 .3.jpg
    Mod 2 All 8 .3.jpg
    19.3 KB · Views: 41
Last edited:
I realize that many are simply more comfortable when looking at CFM numbers.
I created a graph for comparing the various tests I have been performing with one another.
It is important to understand how I arrived at these values.
We all understand that through one entire event (ie: intake valve opening to closing} the valve only sees the max lift once while it sees all other lifts twice.
For the data used in the attached graph I add all data points from min to max and back to min again. All lift excepting the max are seen twice.
This is a far more accurate representation. I call this the "Round Trip"
Since I was only testing at every .10 of lift originally (I started testing at every .05 after doing the original baselines) I only used those same data points for use in this graph so we are comparing apples to apples.

I gave a basic overview covering the Raw, Mod 1, Mod 2, and Mod 3 in the previous video "Let's Catch Up"
You will also notice a Mod 4 and three others:
PP3-This is a stock head with a basic Pocket Port combined with a 3 angle valve job
PP45-Same as above excepting it has a basic fresh cut 45* stock style valve job
PP3ins- The same port as above except I have added an insert within the port. It is very much informational and testing for data.

The graph is fairly self explanatory.
NONE of these exampes include any guide or guide boss modification. YET.
So for those that may be curious... see the attached.


  • Round Trip CFM Comp.jpg
    Round Trip CFM Comp.jpg
    42.6 KB · Views: 42
The head that kick started this is Garrett's (@wannabridin ) D Port.
I took baseline testing of the stock heads and Garrett's D Port at the beginning of this.
Those values were created using the incorrect 62.5 orifice coefficient (Explained and shown in the "Recalibration" video previously).
Unfortunately after gathering all the data I removed all the epoxy and the head is not able to be retested.
It doesn't matter as the raw data used (The DP) hasn't and would not change. Only the coefficient used to convert the data.
I am able to recalculate the data using the correct 45.3 calibration for my bench as well as perform the corrections to 28 inches.
That has been done and the results are shown in the attachment.
I also converted the stock data that was posted with the D Port originally to help with reference.
The D Port still has the same 10% gain (total CFM) over stock. As always, it is the percentage of change that we are primarily interested in.
I am also including a graph of the Mod 3 data just for the sake of comparison to show where we are with some of the testing comparatively.
Again NONE of this data includes any Guide/Guide Boss modifications.


  • Orig Stock.jpg
    Orig Stock.jpg
    27 KB · Views: 41
  • G D Port.jpg
    G D Port.jpg
    28.3 KB · Views: 45
  • Junk Mod3.jpg
    Junk Mod3.jpg
    26.2 KB · Views: 44
  • Stock D Combo.jpg
    Stock D Combo.jpg
    59.8 KB · Views: 47
  • D Port Mod 3 Combo.jpg
    D Port Mod 3 Combo.jpg
    66.7 KB · Views: 46
  • Original Combo.jpg
    Original Combo.jpg
    294.7 KB · Views: 42
Time for another status update.
Despite not posting the work has continued, excepting this past week which I spent in Arkansas visiting family.
All is wekk with them and I have been anxious to get back at it.

I have performed well beyond 2000 individual tests so far.
I know this because I just cleared my data storage on the DP meter.
There have been a variety of port designs created and modified.
Each tested, some thoroughly and others limited.
I am nearing the end of the design testing.
I only have 3 more major changes to consider.
I didn't feel it would be productive to post each of the iterations.
The final design will be shared in detail for those interested.

I am going to share some of the designs and testing that I have been doing.
Hopefully I will be able to do so without creating much confusion.

In the post above you can see that the MOD 3 outperformed the D Port.
It would be easy to just rebuild Garrett's head with that design.
His preference however has been to await all the upcoming tests.
I firmly believe they will result in more gains.

I am attaching pics of the port opening from some of the MOD designs as well as a comparison graph showing CFM flow.
I am aslo including the average CFM flow per square inch using the "All 8 Formula Round Trip" formula.
I have touched on this before in a video but to refresh, here is the breakdown:

Raw- This head had all casting flaws removed as well as having the bowl area opening up on both sides.
130.97 CFM PSI


MOD 1- Added constriction at the opening, emphasizing the "Vane" on the left side of the intake opening inspired by the stock head's vane which is shown in the yellow circle of the "Stock Opening" pic. I also raised Floor.
132.52 CFM PSI

MOD 1.jpg

Stock Opening.jpg

MOD 2- Added more constriction with more emphasis on a larger vane as well as adding constriction to right inside wall.
Added some filling and shaping it into the bowl area. Then Blended contricted area into bowl area. Raised the floor 2mm more.
135.27 CFM PSI

MOD 2.jpg

MOD 3- Reshaped opening reducing the Vane and the right side constriction as well as lowering the Floor and added the "Speed Bump"
137.74 CFM PSI

MOD 3.jpg

MOD 4- Removed Speed Bump
131.97 CFM PSI

Round Trip CFM Comp.jpg

Note the gains made in the first 3 Mods achieved by adding constriction (Venturi) to the opening including the raised floor as well as filling the bowl
and then blending the constriction into the bowl and adding the speed bump.

Also note the CFM loss by simply removing the speed bump.
Last edited:
Something I have yet to share has been the pitot tube testing.
A pitot tube has a main opening which detects pressure as well as openings perpendicular which detect static pressure.


It is commonly used on aircraft for airspeed as well as many other situations.
It can measure air in Feet Per Minute (FPM) as well as Cubic Feet Per Minute (CFM).
Once calibrated they are quite accurate. My Pitot tube has yet to be calibrated.
As I use the measurements for comparison purposes only, absolute accuracy is not critical.
Only consistant measurement. Note that the Pitot has the same markings as my Flowballs every 20 mm.
This aids in consistency when testing various points withing the port.
You can also see that this Pitot is small and designed for use within the tight confines of a port.

I performed this set of tests at .40 lift and max vacuum depression.
The purpose is to map the velocities of the different areas within the port.
I began with the tip of the pitot at the very edge of the intake opening (where the red "intake" mates to the head).
I call this 0 mm as the measurement is taken at the very mouth of the port and not within it.
The pink areas are the slower sections while the blue areas have more velocity.
The slowest measurement is highlighted with a red circle and the fastest with a gold circle.
With this port design we see that the center zone top to bottom excepting the high middle is the slowest zone.
I will tell you the this port design is similar to MOD 3 with a thicker "Vane". It also has the bowl filling removed.

MOD 12.jpg

MOD 12 Pitot 0mm .40.jpg

Going deeper into the port at 20 mm (which is about the depth of the constriction) we can see that the air slows.
Since this is the location of the restriction the air is expected to slow. What is interesting is seeing the higher velocity air
shifted toward the floor and the slower air middle up toward the roof. Notably that area has the least restriction as seen in the pic.

MOD 12 Pitot 20mm .40.jpg

Going deeper yet to 40 mm, the testing location is just past the restriction and as expected the venturi effect increases the velocities.
The faster velocities rise up through the center zone top to bottom. The 2 slowest zones are apparent in the left and right upper quadrants.

MOD 12 Pitot 40mm .40.jpg

Lastly I tested at 60 mm which is located at the mouth of the bowl which is just past the swell at the valve stem in the stock port.
I have shown this previously in the thread. This widening of the port here while entering the bowl slows the air creating pressure behind it
that helps load the chamber (discussed in the example of the Ideal Port) as well as making it easier to negotiate the Short Side Turn.
The deficiency of this port becomes obvious. The fastest air is all located to one side while the slowest air occupies the other.
It isn't coincidental that the faster air flows along the side that the Vaning deflects toward.

MOD 12 Pitot 60mm .40.jpg

In conclusion of this example...
I hope it is obvious to those following how taking the extra steps to obtain this information is useful.
As I said in the beginning, CFM numbers alone can be misleading.
This particular MOD 12 flowed an average of 132.43 CFM PSI using the ALL 8 Round Trip Formula.
For Comparison...
Using the data from Garrett's D Port.
Unfortunately I only tested his head at each .1 interval of lift.
Therefore I can only use the BIG 4 (0.1-0.2-0.3-0.4) Round Trip Formula.
Obviously I can simply use those same points from this MOD for a direct comparison.
D PORT - 86.73 CFM PSI using BIG 4 Round Trip.
MOD 12- 100.91 CFM PSI using BIG 4 Round Trip.

Even using the STRAIGHT 4 Formula (CFM @ 0.1 + CFM @ 0.2 + CFM @ 0.3 + CFM @ 0.4) it is clear.
D Port - 429.25 CFM Total
MOD 12 - 487.58 CFM Total

Even with the obvious deficiencies observed in this port design it flows more.
Most likely not as efficiently although I didn't have the Pitot for testing when I baselined the D Port.
So I won't really ever know,

Anyway I hope this gave some ideas to those with an interest.
Utilizing all the various tests help pinpoint areas to address as opposed to guessing and settling for "Good Numbers".

CFM Flow
Valve Differential Pressure
Smoke Testing
Pitot Velocity Testing
Swirl Tests

Each of the above tests provide different specific data that can be used to analyze deficiencies and aid in optimization.
More to come....
Last edited:
Props for a boatload of hard work and dedication to a task. Super interested in how this pans out when your flow bench tested design hits the streets!
Thanks for sharing!
Cough, I kinda think this isn't appreciated as much as it should be here.
Last edited:
Appreciate the kind words gggGary.
I have a great interest in learning the details and nuances of porting, particularly the xs650 head.
Much of what I learn here can be applied to other applications. When I ported in the past (many years ago) using my shop vac to pull strings and tug flowballs I was able to achieve some gains.
Having the bench allows me to delve much deeper. It's a learning/discovery process for me.
When doing things you enjoy it is not so much like work. Once I have the final port design, the actual modifications won't be as time consuming. I am extremely interested in the feedback once the heads hit the streets.
My intent is to target at least one member (besides Garrett) with a great deal of seat time on xs650s to give me feedback on their seat of the pants experience and other feedback. I would expect gains of 10% or more if the bike is capable of using the additional flow. More cam, scavenging exhaust, more cc's as a few examples. A better flowing head accentuates the power gains from each of those components increasing each of their gains. Of course a head can be designed for stock motors as well, and indeed I am also working on that. I will be sharing that info in it's own thread.
I don't think the thread gets a large following simply because it is a re-port of Jack's radical design (radical meaning intensive work) and most members have stock motors or basic bolt on modifications so they have no real interest.
Ultimately I made this thread for myself and to document the work done to Garrett's D Port.
It is the easiest way to keep him up to date.
It is a bonus that a few other members show interest and hopefully not only enjoy but gain something usefull by following.
Thought I would show an example of how the volume of air (dependent on valve lift) affects the area velocities within the same port.
The example above was taken from MOD 12 at 0.40 lift.
Let's compare those results with the same measurements taken at 0.20 lift from the same Modification.
When I talk about comparing the velocities being faster or slower; I am talking about within the same chart,

At 0mm
Note the similarity of the slower air being located in the center top to bottom with a slight shift.
The fastest air point remained in the same location, while the slowest air shifting downward in the center column as the lift increased.

0mm Comp.jpg

At 20mm
Similar pattern with the upper left quandrant showing a slight shift to a faster velocity at the lower lift.
In both comparisons so far the slower air occupies the same zones in general with only slight variation.

20mm Comp.jpg

At 40mm
Here we see nearly the entire port seeing similar air speeds (13,000+ fpm) excepting the very top center which is the slowest; as well as the 2 fastest points (bottom center and bottom right) mapping at 15,000+ and 14,000+ respectively at 0.20 lift.

40mm Comp.jpg

At 60mm
Here at the beginning of the bowl area we can see a major difference between the 2 lifts being tested.
Keep in mind that all these tests were performed at max vacuum depression.
At the 0.20 lift we see that virtually the entire port measures a similar velocity.
Even more than we saw at the 40mm test. 6 of the 7 zones measured 12,000+ velocity with a high of 12,965 to a low of 12,652.
That is a max difference of only 313 fpm. Extremely equal. The slowest air is located in the upper left quandrant.
This is a similar chart to that seen at 40mm excepting it has shifted left, no doubt due to the guide boss.
This pattern is much different than what we see at 0.40 lift where the volume of air moving through the port is far greater.

60mm Comp.jpg

In Conclusion..
Remembering earlier discussions where we determined the lift point which dictated what part of the intake tract controlled flow.
At lower valve lifts it is the valve, valve seat, and throat that control flow.
The higher lifts are controlled by port design.
In general, the volume of air moving through the port at lower lifts is not sufficient to create enough turbulance
The deficiencies within the port design are revealed with the higher volume/velocities created by increasing the flow with greater lift.
That is also one of the reasons for using a flowbench that can create a deep depression and having the ability to actually test at 25+ inches of water.
In the above comparison, once the air volume became great enough the Vane's deflection of air caused an obvious disruption.
Changes are a coming...
Last edited:
I'm a little curious as to why you didn't take time to properly build the D port for a flow comparison as you would of most likely found the numbers
would have collaborated with what gains I achieved. Could the corners with a bigger radius yield more flow, it's possible but when you look back
at what #s were achieved over a stock head would it really matter that much, I mean there are no perfect port shapes designed to do it all. Just my opinion
Hey Jack,
Great question.
The D Port design is a known.
I have zero question about the gains you achieved. I don't believe you need any validation of your work. I am absolutely certain of your gains. Your opinion matters a great deal when it comes to this subject.

Buliding a proper D Port is something that I could still do. Garrett's head is opened for it already. My objective with Garrett's blessing has been to test other designs looking for gains. Those were unknown, at least to me. So that has been the focus. So far I have been on a mission of discovery The final decision on port design has yet to be made. I am in regular contact with Garrett and he is always informed of ideas, tests, results, etc.

I am a HUGE fan of your design, however there are a couple of issues concerning the D Port for me. The dead areas in the square of the "D" which I have mentioned and you referenced is one. Most likely just filling that area and maintaining the rest of the design would be sufficient as long as the floor was widened to the turn. I believe that would increase the efficiency.

The biggest issue for myself when contemplating the D Port is the additional work required to mill those corners which are dead zones anyway.
There is also the additional need to modify the intakes for port matching. I see the potential to create the same basic design, just deeper within the port eliminating the need for intake modifications, while reducing the CSA, gaining port velocity. The floor can be widened at tge bowl without milling away the port opening.

These are just thoughts and ideas among others that I am fortunate enough to be able to test.
Of course many, many ideas go backwards.
Others work out quite well. The goal for me would be to design a port that that is active, efficient, and shows significant flow gains.
I would like that design to be something that the ambitious members can duplicate with the least amount of difficulty. There are a few VERY skilled members here, with the equipment to duplicate your proven design. Although it may seem I have departed from the original intent (and I have to some degree for the reasons above), ultimately I still lean heavily to the basics of your design and study intently how you addressed some design issues. I have been combining them with other ideas, sometimes successfully, sometimes not.

Copying your port would have been the easiest route for me to take. I could have been finished long ago. Considering that Garrett's head already has been milled open for the D Port design there exists a greater flexibility for me as I go forward.
I am impressed by what you created.
I am glad that you still follow this as you are perhaps the most knowledgeable and experienced person around when it comes to these heads. Your feedback and questions can only help myself in this endeavor as well as those others that are interested.

It is my hope that you can understand why I didn't just rebuild it as a proper D Port right away.
You are absolutely correct that there are no perfect ports. It is definitely about give and take.
The numbers you achieved are absolutely outstanding and I am beyond amazed that you were able to achieve them without a flowbench.
It is a testament to your knowledge and experience. Ultimately it is my objective to achieve similar results without having to take the additional steps discussed above.
I believe I am very close now.

I always appreciate your feedback Jack.
I want to be very clear that my efforts are NOT a dismissal of your D Port design.
I am simply attempting to utilize the best parts of it to attain similar gains, without having to use the more extensive aspects mentioned.
Part of that give and take...

Your design and the gains you achieved are the measuring stick. It is the standard.
If you are ever anywhere near Mississippi you are more than welcome to come spend some time.I would love to have some discussions and see what you can do with a
bench to test ideas with. The invitation is open. I've got a place for you.
Last edited:
Rectangular and oval ports both have their pros and cons with oval for more torque and rectangular yielding more airflow for top-end power. So I chose
to incorporate the two port shapes to form my D port which allowed me to widen the entire floor to the short turn for a straight shot to the valve with little widening of the floor at the short turn. On the modern Mopar hemi head, it is a straight shot with straight walls to the short turn, and I was advised to build
the port that way to limit resistance for increased velocity and flow through the entire flow curve. If you want to work the bowl area by widening it, search for the late Larry Cavanaugh who was famous for making big power on Kawasakis, but you need to proceed with caution as you are limited in how wide you can go,my bowl work consisted of working the long side to straighten it out to the backside of the valve. Larry made his power in the bowl, etc. Raise the floor 3.5 to 4mms to reduce the erupt turn the flow must make at the S/T, despite what port shape you decide to go with, it's a must in my book, you want flow gains and to adjust flow percentages if needed. Later
Last edited:
Yes indeed.
Excellent nuggets of information there Jack.
I am in agreement with all those points.
In fact I opened the bowl on the junk test head supplied by gggGary and encountered reduced velocities. I filled that back and shaped the long side for far better results.
I absolutely could not agree more about raising the floor and have found my best results when duplicating the original slope of the floor.
I've had success with the "Vaning" to create additional swirl, however it still costs me some flow. Especially at the highest lifts where the maximum air volumes and velocities exist.
Of course that is exactly where the flow needs to be maximized. I haven't yet found a compromise that I am satisfied with and believe me, I have been looking. The Vaning idea (which was inspired by the stock head) has been set aside although a very small vane may yet be incorporated, it would be as a final addition to a performing port only if testing proved it to be an enhancement. I did learn a considerable amount about deflecting/directing air from those studies.

One advantage of your design creating the restriction at the mouth of the port is the extra 15-20mm of runner length gained between the restriction and the bowl compared to setting the restriction inside the opening. That is nearly a third of what is available, which is considerable.
The guidelines would say you need 40-60mm of parallel runner before the turn. It's a tight fit.
It is also notable that your design with the straightened side walls, elimintaing the stock head's "bulge" at the valve stem creates a far more parallel runner. I have done multitudes of tests modifying that area alone.

Most of my testing and path of "Discovery " is complete. The upcoming updates will be more about bringing together all the information as I meld a final design together. It will most likely be a hybrid of sorts much as you meshed that oval and rectangular port together. I feel certain that your influences will be apparent to those paying attention. I'd like to thank you again for sharing your knowledge and inspiration.
Bluzplayer, The one area of the port I didn't explore much was what I call unshrouding Or eyebrowing the area around the guide boss and transitioning
it to the bowl and long side. You don't hear a lot of discussions doing this mod on motorcycle ports with the exception of Nortons( the idea came from the XR750 oval ports and on a BSA single) where it all originated by calling the mod Narley porting by JS Motorsports, just google Narley porting at Access Nortons. I had every attention of trying this mod but life priorities put this idea on the back burner. I believe this concept could be adapted to the XS for bragging rights but you have to start cutting a bad head open to find your porting limitations in the bowl and roof. Later and have fun with your flow bench.👍