Busa Stage 3 port testing....

[Competition CNC]

We just finished the development and static testing of the Hayabusa Stage 3 cnc port and valve job. Dyno testing will be in about 3weeks and real world testing will be at maxton at the end of June.  The test engine will be a 1607cc engine in June and a 1650cc engine shortly after that.  We are VERY optimistic on its performance potential at this point.

Just a few things to note, It has the smallest radial velocity bias of any cylinder head we have ever tested.  The average velocity at the MCSA is now solidly sub sonic so on a running engine it should translate into an extremely efficient port.

This head isnt for the average street guy, its an all out race head designed for max power above 9000 rpm.  We are planning on offering a slightly modified version for big displacement drag racing at some point.

I'll be at Maxton in June,  hope to see you guys there.

[TRNorBRN6001]

Cool, I just might need one of them!!!!! I'll check your site for pricing and what all is involved with the head.

Hope Karl does not come poking around the motor section, it would be funny to surpise him at Maxton. If you have not told him your headed to Maxton, Don't, he would get a kick out of the surprise. Goes for you too Ryan.

[DarkFalcon]

We just finished the development and static testing of the Hayabusa Stage 3 cnc port and valve job. We are VERY optimistic on its performance potential at this point.

Just a few things to note, It has the smallest radial velocity bias of any cylinder head we have ever tested.  The average velocity at the MCSA is now solidly sub sonic so on a running engine it should translate into an extremely efficient port.

This head isnt for the average street guy, its an all out race head designed for max power above 9000 rpm. 

Y2K, if I understand your use of the term radial velocity bias it refers to in-cylinder charge motion. How would or did you you measure this? All intake ports should be solidly subsonic.

[Competition CNC]

I think Karl knows were going to be there allready...... I hope he can make it, I want to see that beast 1427 zx12r of his run.

Give me a call and I'll go over whats involved in a stage 3 head.  Its quite a bit more expensive than a stage 1 or 2 head, we have to do O/S bronz seats, guides, more chamber work, o/s valves etc. so the heat can disipate properly when held at WOT for several minutes. Longevity is key.

Good question DF.

You measure the velocity with a pitot tube and a monometer and you start in the center and make a map of the port velocity in clock positions working outward in a spiral.  You do this at the same place depth wise in the port.
Then you step in a given amount, typically .5" and do it again and again.  When you look at all the radial velocity maps in the order you created them in, you can see things you cant see while the air is traversing the port in real time.
In our banner add on this site is a radial map of a port, thou its not the stage 3 busa port.

These maps are marked out before you start so you can keep the pitot tube at the same slice of the CSA in the port for each map.  You can then measure in Mastercam X2 what the exact CSA is at each map and determine the average velocity for the port at that exact slice of the port.

The colored rings show the flow bias which is typically highest at the shortest line in the port. Like if you took a string and held it taught threw the port and tried to keep it as straight as possable.

[osti33]

Good stuff!!

Can't wait to hear about the results.


Gary,the cat is out of the bag. Karl already knows.

[entropy]

yep, karl knows!

good stuff Jim.

I am real interested in the kinda stuff Jim is doing, but it makes my head hurt when I try to understand it.

[Competition CNC]

I hear ya Karl. 
If you read the SAE papers 2004-01-0998  and 2006-01-3653 it makes it easier to understand.  They explain port dynamics WAY better than I ever could.

If I went into the nitty gritty details of our designs but it would bore the hell out of everyone and no one would care anyway, so I don't.

Even better, get a copy of "Design and simulation of 4 stroke engines" by Gordon Blair.   Its always within arms reach for me ... 

When Jaime asked what I was reading one time I handed it to her.... she read a page or two and shook her head and said her brain hurts from it.       LOL

I kid her now and threaten to make her read it once in a while.   

[dakinebusa]

What's a few differential equations between friends.
How many porters have heard of Navier-Stokes?
Great reading.....

[Competition CNC]

I bet your local weather man has...   

[DarkFalcon]

Interesting Y2K. I am familiar with port mapping using a pitot tube and measuring velocity pressures but I had not heard the term radial bias as it would suggest movement away from the center.

[Competition CNC]

Well, basically the air in the port doesn't travel straight down the port.  Nor is the velocity equal in all areas of any given cross section.  So Its a way of mapping the different areas of velocity. I call it radial mapping because thats how I map it.

Once you have created the slices or  "radial maps" you can view all the slices in mastercam and see the radial movement of the airflow as it travels down the port and see where the high velocity areas are and or the flow bias is in a given port design.
This allows you to keep the port velocities at the airspeed you want while increasing the total cfm.  It also allows you to see where velocities are too high so you can see the areas where sonic choke will start and build from.  Sonic choke is typically why you will see an engine "table top" on a HP graph instead of peaking and declining.

Another benefit of these maps is the ability to predict fuel saturation streams and or fallout areas and dead spots.

We don't have a wet-flow adaptor yet on either of our flow benches so we have to use these maps.  I want to get the wet-flow setup soon.  Not so much for the bike stuff but for oval track intake manifold testing.

[dakinebusa]

In fluid mechanics we call these velocity profiles.
In a straight tube they vary from fully parabolic in laminar flow to nearly flat in turbulent flow.
When the pipe is not straight a lot happens to velocity profiles.
Essentially, a vortex develops that interacts with the velocity profile to make it assymetrical.

This is in well developed flow many diameters downstream from the entrance.
In a short port with a varying cross section and curvature it gets even better.
Then if the flow is pulsating and the entrance and exit is assymetrical....well, bring a large supercomputer.

The high buck guys use a robot positioner to step a hot wire anamometer all throughout the port to completely map the velocity profile.
This experimental data is fed to both engine and model developers so they can tweak thing.

[gazza414]

The hot wire is less intrusive to the port flow ..rather than a pitot tube ...it sets up its own special local conditions

good stuff Jim and good to see some good investigative research you're conducting.....gotta see if it works in practice in the dynamic mode.

 

[Competition CNC]

In fluid mechanics we call these velocity profiles.
In a straight tube they vary from fully parabolic in laminar flow to nearly flat in turbulent flow.
When the pipe is not straight a lot happens to velocity profiles.
Essentially, a vortex develops that interacts with the velocity profile to make it assymetrical.

This is in well developed flow many diameters downstream from the entrance.
In a short port with a varying cross section and curvature it gets even better.
Then if the flow is pulsating and the entrance and exit is assymetrical....well, bring a large supercomputer.

The high buck guys use a robot positioner to step a hot wire anamometer all throughout the port to completely map the velocity profile.
This experimental data is fed to both engine and model developers so they can tweak thing.

The busa port develops an intense vortex at the short turn between the two valves.   Typical of 4 valve split ports.  and the pitot tube does have an effect on the vortex when measuring velocity.
Thats where I could use a hotwire setup like that, or some kind of micro transducer so it wouldn't change local behavior.  But we are only actually measuring velocity there and not swirl so its not too much of a problem.

I'd really like a micro swirl transducer of some type to directly measure rotational movement.  I've looked high and low for something affordable and no luck.

Garry, maybe you can plan a business trip to the USA at the same time as maxton...? 

[dakinebusa]

The real tool here is the Laser Doppler Velocimiter.
Make a mockup of the port from cast plexiglass so the laser can probe everywhere without upsetting the flow.
Put the mockup in a square tank of refractive index matched oil to avoid refraction problems.
I used to run an LDV setup lashed to the table of an old CNC Bridgeport to completely survey flow fields.


http://images.google.com/imgres?imgurl=http://www.tut.fi/units/me/ener/laitteistot/EFD/gfx/LDA_principle.jpg&imgrefurl=http://www.tut.fi/units/me/ener/laitteistot/EFD/LDA.html&h=270&w=500&sz=84&hl=en&start=3&tbnid=OMs2o783ET2hYM:&tbnh=70&tbnw=130&prev=/images%3Fq%3Dlaser%2Bdoppler%2Bvelocimeter%26gbv%3D2%26hl%3Den%26sa%3DX

Here is a pic of the test setup in the NASA 80' by 120' wind tunnel that can be entirely surveyed by LDV.

[Competition CNC]

Very cool.   I wish I had a budget that would allow me to use such tools.

[DarkFalcon]

Y2K, in light of discussion of three dimensional flow, port mapping and port velocity profiles, I still am having a hard time constructing a metric that would measure "radial velocity bias". I understand the tools, procedure and methods but I am wresting with specific metric or calculation that would permit comparisons between ports. I can see how you might calculate the bias for one slice of the port but I do not understand how you combine the slices of one port and come up with a unifying measurement. An example would be most appreciated.Thanks.

[Competition CNC]

Look at the map in our banner.  You can see the colors and pattern.  Lets say that slice was 2" down in the port.  The next map at say 2.25" down the port will be slightly different looking.  Then the next and the next.  So as you look at them you can see trends, like the highest velocity area moving to another part of the port, etc.

[gazza414]

Jim , in reality both pressure and velocity can be transposed wrt to these graphs ?---just another way of representing the data?

and yes , would love to come along to Maxton... another year I'd say unfortunately

[Competition CNC]

Well yea, the pitot tube is actually measuring pressure when its hooked to a manometer.  So yea.....

[DarkFalcon]

Y2K, do you guys start getting nervous and start looking for die grinders when measured velocity pressure equals the test pressure?

[Steve S]

Well yea, the pitot tube is actually measuring pressure when its hooked to a manometer.  So yea.....

Jim, isn't velocity simply the square root of measured pressure times a constant?

[Billy D]

Any performance results yet instead of this Dr Emmett Brown (from Back to the Future who can't figure that name out) talk? 

[dakinebusa]

Well yea, the pitot tube is actually measuring pressure when its hooked to a manometer.  So yea.....

Jim, isn't velocity simply the square root of measured pressure times a constant?

That'sa Bernoulli's equation.
You are correct.
More information than velocity or pressure is needed to quantify a turbulent flow field.
Knowing the vorticity of an unsteady flow is important too.

[Competition CNC]

Well yea, the pitot tube is actually measuring pressure when its hooked to a manometer.  So yea.....

Jim, isn't velocity simply the square root of measured pressure times a constant?

Yea, what you are actually measuring is the pressure differential.

stagnation pressure = static pressure + dynamic pressure.

That'sa Bernoulli's equation.

Yep.