Honda / Acura K20a K24a Engine Forum banner

1 - 16 of 16 Posts

·
Registered
Joined
·
81 Posts
Discussion Starter #1
Because I want to build a K24 normally aspirated for my Exocet project, doing valve seat and port development will be extremely useful. I came across an early RBB head on eBay at a good deal. It's got 35mm intake and 30mm exhaust valves.

I stripped the head down and cleaned it up a bit, but I don't have any sort of soda blaster so there's a bit of carbon still around. The intake seats are in amazing condition all things considered.



I pulled out one of the flow bench adapters. It allows inserts for various bore sizes, the last of which was for a BMW 2002 with an 89mm bore. It wasn't as nice as I would like so I bought another PVC fitting and some epoxy to make a new one. It's the cheapest way to get a representative 87mm bore, but its' only 2 or so inches deep. Seemed a waste to bore a 4.5" Aluminum bar, and I'd be on the lathe all day. The adapter is using locator pins and I'll use clamps to secure the head.




Not bad on alignment, the cylinder looks to be very well centered between the 109mm x 94mm head bolt pattern.

I still need to make a gasket for the head and machine up a device to allow opening the valves and measurement of the valve location.



Because I have a few cylinders to play around with I'm going to split development into two areas. Development at the seat, and then development in the port. Here's what I'm thinking so far:
  • Stock valves and ports
  • Back cut stock valves and port
  • Ferrea 6000 series 35mm valve and stock port
The Ferrea valves are of their "Super-Flo" design, meaning they come with a 30 degree back cut and necked down valve stem below the guide. After I get the baseline and a few valve mods, I'll start with removing casting marks and blending the bowl in a bit.

I need to get a 5.5mm pilot to play around with changing the seat angles, but I'd like to try a 75 degree back cut to blend the seat into the throat a bit nicer and see if that yields any improvements.

Lastly I'll start looking at changing the ports themselves. I don't want to make too many changes to the port just yet. I'm still working through some flow calculations to approximate some averaged air flow velocities. Anyone can get big CFM numbers hogging a port open, the challenge is to improve flow while not reducing velocity. Velocity helps with mixture distribution, and helps with getting that last little bit of air-fuel into the cylinder at high engine speeds, when the intake valve is closing after the piston is trying to push the intake air back up the port.

I also made a silicon mold of the intake port to try to see where changes could be made. Incredibly difficult to visualize so far, but stay tuned for results.


 

Attachments

·
Registered
Joined
·
81 Posts
Discussion Starter #2
I did a couple measurements on the K24A2 inlet port mold last night. They're pretty rough approximations of cross sectional port area in general.



Section A is the cross sectional area at what I'm calling the port entrance and I'm using that as the baseline to judge the areas further down stream.

Section B is about an inch into the inlet port, slightly smaller here than up stream as a good velocity stack should be.

Section C is just behind the valve guide but after the septum has split the port. It's approximately circular with a 27.5mm width and height. It represents 77% of section A.

Section D is the diameter of the inlet seat. It's machined to about 30.9mm.

Obviously the smallest area is around section C. This isn't necessarily a bad thing, it means that there is a venturi designed into the inlet port. Intake velocities will be highest through this area, which will help by generating turbulence that mixes air and fuel.

It's probably sized quite well for the stock design, but as you start adding things like camshafts, you're trying to squeeze more air through this area in shorter amounts of time, so it will choke flow.

I'm trying to find more information on the topic, but I've found anecdotal evidence of keeping average port velocity in the 100 m/s range, and I believe peak velocity in the 0.55 to 0.6 mach range for peak power.

That range of mach numbers is a balance between two things: First, obtaining enough velocity that the column of air in the runner and port has inertia to force itself into the cylinder when the piston is coming up the bore after BDC. Second, the velocity is low enough to not present excessive losses from drag as it moves through the port and runner.

There also seems to be a slight mismatch in the machining for the seat and the casting for the throat of the intake port, maybe 1mm or less. Seeing as the 30.9mm seat ID represents a throat-to-valve ratio of 88%, I'm thinking I can open it up a bit to better match the throat to the seat.
 

Attachments

·
Registered
Joined
·
81 Posts
Discussion Starter #4
Out of interest, what silicone did you use for the port mould? Looks like it turned out very well!
I found this stuff on Amazon:
https://www.amazon.com/gp/product/B004BNA6VA/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

Two part RTV silicone. I didn't have any decent release agent so I sprayed in some silicone spray lubricant into the port. let it cure for 2 or 3 days and then separated what I could with a pick and small screw driver. In the end I managed to get a hold of it and gave it a good pull and it just came right out. It's way more resilient than I was expecting LOL
 

·
Premium Member
Joined
·
956 Posts
Make sure you bolt up a manifold to the head when you cfm test. Bench flow testing is deceptive without one.
 

·
Arouse the DAMPFHAMMER!
Joined
·
6,054 Posts
Because I want to build a K24 normally aspirated for my Exocet project, doing valve seat and port development will be extremely useful. I came across an early RBB head on eBay at a good deal. It's got 35mm intake and 30mm exhaust valves...
Very very nice informations you share with us Scider. Many thanks for that! I really the approach you took, old school and best way to analyse the port itself, as to make that head flowing for a specific application isn't that easy.

My suggetion for finding a first order assumption of the switch point where valve and where port dictates how much flows would be to calculate the cross sectional area of both, throat and valve flow cross section over valve lift. It is just a rough assumption as the flow efficiency is also a function of the valve lift. Surprisingly less porters keep attention to low lift flow when doing NA heads, but those do keep attention there now why. NA engines need a huge force to accelerate the air column, if it is started befor TDC by underpressure exhaust wave, arriving shortly before EVC (exhaust valve closing), then the air column get accelerated very good, which helps for the entire revving band. On dedicated engines 6 mm intake valve lift is already reached at TDC, so it is critical to know what are your supporters.

What application is it going for and what idea do you have of the torque curve?
 

·
Arouse the DAMPFHAMMER!
Joined
·
6,054 Posts
I saw that, I've been side tracked with building my Exocet so K-series stuff has been on the back burner unfortunately.
Wow, that's an amazing welded metal mass. I guess around 50 m of welds :D. Do you like welding now?

I may have the chance to 3D scan one of the heads, probably the K20Z4 head. I will let you know if I go this route or do the old school stuff like you did. Anyway, I really like what you do here for your knowledge and what you share of it. I know this takes massive time amounts, as I am in a similar situation. Two jobs and a lot of DIY to create tools for knowhow and knowledge building. Would be easier when having some million bucks at the bank account, but it would be maybe less fun and intense.
 

·
Registered
Joined
·
81 Posts
Discussion Starter #11

·
Arouse the DAMPFHAMMER!
Joined
·
6,054 Posts
:D...understood. My first stainless steel weld with the WIG weld machine on two 1 mm thick plate was nice and smooth. Welding need a lot of experience to mix material, weld material, heat or current right of course, but a keep cool hand is a basic thing you have or have not. I didn't like this eye shelter thing, when you see your weld flowing only when you spark it up. I would need a automatic fast response accommodating light filter if something like that is possible. Finally I stopped my welding excursion...

They have a chassis development thread on the grassroots motorsport forums, they use CNC laser cutters and CNC tube benders to make the raw materials, and a pretty extensive jig setup to weld it all together. It's not perfect by any means, but it'll be a fun and never ending tinker toy.
Sounds like a bunch of fun car :D.

I hope I can soon report about the K20Z4 head investigations.
 

·
Registered
Joined
·
39 Posts
If you haven't read "Power Secrets" by Smokey Yunick, I strongly suggest reading his chapters on cylinder head flow before beginning. I am a long-time engine engineer and consider what he has written to be extremely useful. A lot of information on all of the details of port and seat flow in this book.

Dave
 

·
Arouse the DAMPFHAMMER!
Joined
·
6,054 Posts
If you haven't read "Power Secrets" by Smokey Yunick, I strongly suggest reading his chapters on cylinder head flow before beginning. I am a long-time engine engineer and consider what he has written to be extremely useful. A lot of information on all of the details of port and seat flow in this book.
For a non-scientific 2-valve head book 190 Euro (~220 USD)? What makes it worth?
 

·
Registered
Joined
·
81 Posts
Discussion Starter #15
My most recent purchase was David Vizard's How to Port & Flow Test Cylinder Heads. It seemed quite a reasonable bit of literature, but you can find a majority of it online for free:

as well as articles from the book on this website:
 

·
Arouse the DAMPFHAMMER!
Joined
·
6,054 Posts
My most recent purchase was David Vizard's How to Port & Flow Test Cylinder Heads...
Yes, thanks Scider! I read this many times, I like his approaches, and smiled often when he talks about himself and his achievements. Unfortunately many of his approaches and answers are related to the 2-valve concept, where huge valves and valve lifts, swirl, small chamber volumes, huge complex ports and shrouding freeness are effective. It's a good place to learn, but to start with 4-valve heads I miss many answers regarding flow regime design, de-shrouding approaches, valve seat designs and port wall divider techniques to name a few. Harold Bettes wrote a nice book about flow bench work, which is more interesting regarding the way to get somewhere. A real 4-valve head approach book I still miss.
 
1 - 16 of 16 Posts
Top