!!!!Sorry for being all over the place and maybe being a bit repetitive with my other posts!!!!


Hi I didn’t get any Email updates so just saw the questions... anyway. I am rebuilding after overheating twice once the coolant cap didn’t get screwed on rite and blew off then when another driver had it and the fans didn’t go on in stopped traffic, the first time the temp gauge only showed minor fluctuations as the sensor is on the plastic thermostat housing. I could have kept going but there was more blow-by than I wanted because the rings lost just a bit of tension, other than that the build was solid and I could just replace rings, But I also hit over 1-1.2G's,,,, just cornering around town all the time, and my logs have been showing that the oil pickup is getting uncovered at times and although the bearings are smooth there is more wear than I like. I am going to dry sump and upgrade the cooling system. So that means a whole knew approach to the valve cover stuff.


Goals are important to decide at the start I feel that with a cooling upgrade and better rods you can push enough power to need extra injectors water-meth etc. Going over 400hp and some crazy torque are quite doable, but of coarse the trans drive line will take a beating.


Head work is important of coarse, pay attention to the area around valves so that there is a sort of air reserve. In the past there was a limit to this as it could stall the air flow and allow fuel a chance to start separating out but we have direct injection and a relative deep draw small bore. Cams also help. Imagine the air flow as the valves are closing air crashes into them and swirls around think of the pressure waves as well. You will see that unlike some other engines I’ve worked on that the runners make a sharp turn into the cylinders I try to get a better outer radius, the inner small bump wedge is something you will need to decide about. I rework this until its almost gone, at idle it probably does cause some issues, but at just 1000rpms it seems better to me from my unscientific vacuum cleaner tests. But others feel it turns the air. Anyway even a small bit of work helps greatly just don’t polish, an 80 or even 60 grit finish will help keep any air from sticking remember air slows towards the outer walls it can even stop, small turbulence's will actually help flow to a point.


I haven’t decided about a new turbo or not but if I do it will be quite large I think because then you get nice large pressure waves and hopefully can run less boost, which I am thinking will maintain and compliment the high flow intake valve setup----- yes too much thinking will get you to the Stokes equations-----. A small bit of lag is fun to me I like the feeling of the turbo kicking in and when at the track I keep my rpms up so its not a huge issue of throttle feel in the turns. Standard efficiency charts are good but just part of the picture.


Boost pressure is not directly related to flow, flow think flow then think boost, cooling the charge air is not just about detonation but more flow better pressure wave profile. Work out the overall flow of air for the power you are working towards and try and set up a vacuum cleaner test,, its not truly related there is a whole lot of subtle quirkiness going on but is helpful, try a bunch of 90deg turns then make some nozzles from aluminum flashing, for a flow meter I used a small fan and volt meter and a water tube for vacuum, super crude but fun and to the point, now try different textures also just listen you start to get a feel for how changes work then try it on your cylinder head. Yes a real flow bench and practice and different engines is the way to go but you have to start somewhere.


Found this somewhere hope it helps






OLD SCHOOL FOR COMPARISON NOT WHAT YOU END UP WITH ON THESE HEADS

 

That cut a way would be much better if they cut it at the center of the port with the seat rings so one could see how thick the aluminium is.
I started my porting; just contouring and lowering the bowl a smidge.
Does going too smooth cause "stiction" on the surface? With DI, I thought just air would benefit from a really smooth surface.
Fluid/gas likes a rougher surfaces to reduce cling and enhance fuel atomizing turbulence.
Not sure with DI though
Wonder if this applies to both exhaust and intake
Any thoughts out there on this

 

Hi Indimanic flow bench tests show more air flow with a bit of a roughed up surface, I know it seems a bit funny at first but my friend ran an old polished intake on his flow bench for me and you can even hear a sort of sucking sound when you get the surface too smooth. On DI intakes we dont have to worry about fuel mixing so its a bit easier. The exhaust side is just as important as the intake and I dont know why it doesn’t get as much attention, and the same applies about a rough surface, but as exhaust deposits form they make there own surface so I don’t polish but go a bit smoother. The factory surface is not exactly a mistake they use the cast surface on purpose its just we have different goals.


With flow remember as you move towards the outer wall the velocity goes towards zero. So by leaving a rougher surface you make turbulence swirls so that flow closer to the wall becomes faster. But of course the turbulence can slow things down as well so its a balancing act and more of an art even with computers because we still have some things to figure out about flow and chaos stuff.


Also remember the valves are opening and closing interrupting the flow and the goal is to fill the cylinders with as much air as possible so the whole swirl thing gets even more fun. Two valves intakes tend to have a natural swirl which helps. Oh there is the term tumble which is more like a wave crashing than a sink draining. Tumble is why the N14 intakes look like they do. A good cam will help its just going really wild on the cam will require rethinking valve springs and then the valve train.


Velocity--- these heads are set for velocity so we cant change that up too much but filling faster and smaller runners or larger and slower runners is another whole story.


There were N14-18 WRC rally cars and they made there own cylinder heads so ive asked around to see if I can get one or get some picks of how they were setup but I think there was no Vanos allowed, also want to get some other DI engine head pics to see how things have changed.


Anyway I am still messing about with porting these or I would just sketch out the profile. When I get a burnt up head Ill cut it into a few more sections and post them up.


Sorry I got a bit off my Mini for a while but after thinking about it Ive decided to stick with them. Ill post up the dry sump once its going.

 

Thanks for info;
Hard to wrap my head around 0 flow at the surface but it's published everywhere.
What I need to assuage is the conflict now with DI motors, how much turbulence does one need from air traveling down the intake???
The air (air only) is flowing through the entire intake. Everything is there is super smooth. The air gets to the port and.... should we need tumults of air.......b/c???...in the past it had liquid(fuel)......but now not..... so just air......the DI injectors creates enough turbulence and atomization with a ???? zillion psi of fuel blowing .....sorry
So, everything I do read about air flow over a surface indicates smooth for max speed is best!!!!
I have my head ported and mostly finished just cannt decide on the final finish.

 

80 grit would be a good middle ground Ive used 60 and 80. even using a grit you can use an old bit of 80 and go light to give a good surface that will avoid the sticking air. With port injection you might even dimple the surface to get extra mixing but that is at the cost of the fastest air flow. Practice the final finish steps on a bit of scrap when it looks good you will know.

 

You're talking about the boundary layer. Do a search, there's some good articles out there that describe what's going on. Basically if the surface is too smooth, there's no boundary layer, just turbulent flow at the surface so drag goes up. You need a little bit of surface roughness to keep the air (or other fluid) attached to create the boundary layer and reduce drag. For example the hull of a boat that is polished to a mirror finish will have more drag (and therefore be slower) than one that is sanded with 600 grit. It is counter-intuitive but there is a lot of science to back it up. How much surface roughness depends on the density and viscosity if the fluid and the speed, i.e. the Reynolds number. A college level course in Fluid Dynamics will cover all this.

 

Yes the boundary layer in fluid flow has some good science but it works towards a general understanding. The Reynolds number is a good point to start or check but only the beginning especially when you are designing. Air changes as humidity pressure and temperature change and as you increase the volume and or velocity it can make optimized solutions fun. Luckily testing is not to hard but can take a few design cycles to figure out. The part that remains art is the subtle changes that will make a worthwhile difference. If this stuff is interesting to anyone look into the Stokes equations they are a nice hobby. Remember small increases in efficiency can make a good improvement as more flow is considered and more complex flow systems so its worth thinking about.

 

Thanks for the input and feedback
SQB & E-S, your reference resonates with me as I have a super keen interest in Americas Cup boats for some 43 years; from climbing on Intrepid I & II to seeing us loose the cup in Newport while working for the Italian team and 35 years later to winning it back in my own front yard. BUT.............,,,,,,,,,,
all the written stuff above speaks to air with fluid or, in out beloved other discipline, its 99% water rushing past a surface we speak of.
Thats very different that just air w/ condensation and humidity!!!!!
Bottom line as I have deduced save all the verbage...........It appears just air; save hummidity, temperature, bla bla bla LIKES a smooth surface for MAX flow i.e. velocity of air.
I feel a fuel injector at 1200 psi or ??? will suffice for proper atomization and turbulence.
Can one imagine what happen in a cylinder when air rushing at 700 mph gets wacked with 1200 psi of fuel blasting away??????
Seems like a the perfect catalyst for what I desire.
The ports will be done 320 grit with cross buff from 3M and the utilization 180 grit on short side radius and just prior to the bowls.
The dyno will tell us what happened.
NOW...has anyone found what sealing washers are available so one can index spark plugs?
Just checked them all and they all point in different position; none towards the intake or exhaust. Why would one them pointed toward the incoming path of the airl????
I would think towards the exhaust direction would offer a modicum of needed protection to the flame kernel!!

 

Seems we share a passion for boats as well as MINIs... I have been following the America's Cup since the 70s, and in 95 during the AC in San Diego I got to know the guy who was doing all the computational fluid dynamics work for Team New Zealand. Then seeing the AC72 foiling catamarans going 50 knots on San Francisco Bay was epic. But I digress.

The key thing is air is a fluid too, just like water, only not as dense, and all the same principles and math apply. It has been a long time since I studied Fluid Dynamics in college, and I know just enough to be dangerous so I won't go any farther. I like this quote from the Wikipedia entry on the Reynolds number: "Fluid flow is generally chaotic, and very small changes to shape and surface roughness can result in very different flows."

As I see it: Laminar flow in the intake manifold and head will have less drag vs. turbulent flow and therefore deliver more air to the intake port. Laminar flow means there is a boundary layer which means you need appropriate surface roughness for the air velocity and density. Thinking about how the flow changes dynamically with varying turbocharger boost and the intake valves opening and closing makes my head hurt. I think you are on the right track to have things a little bit rough instead of mirror polished, but that is only a guess and it would take a flow bench or a lot of CFD simulation to know for sure.

Interesting discussion. Post pictures when you're done!

 

I don’t have a lab and this stuff is a hobby for me so yes it is kind of fun but crazy to follow the flow all the way through an engine. The interesting part is that even though fluid flow is so important there are some fundamental things that still need to be learned. Mostly because the tools and equations that are on hand do a good job but not a great job especially when you try a new design. I was surprised to learn that it wasn’t until 2008 or so that someone did an experiment in relatively high resolution to see that fluid goes from laminar to turbulent and back, it of course seems obvious, but is it? I would like to see some work at the electron microscope level and see what the outliers in the experiment might show.


Anyway air as a fluid is one way to look at what happens but I think its a bit old at this point and gas and liquid do have some differences. I now try to really think about what terms mean and even look up the word roots to try and find the core ideas. Science and Math have to many buzz words and conventions that can cloud some nice basic ideas, or overly simplify something complex.


So from my rough experiments and reading into the polished porting and why it causes that terrible sound. I think what happens is that it creates very still air at the outer wall but nothing is perfect and engine flow has disruption in the valves opening and closing so when the air finally gets a disruption (regression back wave) its kind of catastrophic and builds into huge turbulence. What we want is to control the turbulence and get it to work for us so yes surface is important as is shape. Oh don’t forget the shape of air and walls. Air as a shape is what? Add time or try and take an instantaneous slice what are you really seeing. Is laminar flow hiding something interesting at super high resolution?


Off the super deep end this stuff has led me to creating fractals and then trying to decay them into something that is not utter mush. ((Utter Mush) is my new science word)......

 

I went down this same rabbit hole a while back. I'd ported and polished small block heads years ago but these turbo animals were new to me so I studied several books, wrapped my brain around the math, etc. What I learned was that you need approximately eight feet of straight pipe to achieve true laminar flow. Any bump or disturbance and it's gone. So our choices are somewhat turbulent air or very turbulent air. I was also able to spend a couple of days with two local shops that port heads picking their brains for all I could get. One old timer was kind enough to direct me to an study that pretty closely matched his experience with a flow bench. (He won't touch valvetronic heads BTW) In the study the total flow reduction due to wall friction from a sand cast surface on an intake runner accounts for approximately 4% of the total flow loss through the port. Polish the runner and the loss drops to 3% so we gain or loose a total of 1% flow through the port based on the surface in boosted applications.

Since we can closely correlate flow in lbs/min to HP we can make a pretty good guess of loss based on surface. If we want to make 400HP at the crank we need approximately 40 lbs/min of air flowing into the engine. If we leave it rough with 60 to 80 grit we leave somewhere between 2 and 4 HP on the table. Your biggest gains will come from having four to five angles cut on the valves as opposed to the three from the factory, reworking the bowl and blending the short side radius. (Expect the last one to affect idle quality)

I think a more important question in regards to the final surface is whether or not you intend to ever run meth injection? If so leave it rough, if not polish away.

 

Found an interesting book here: https://books.google.com/books?id=KAdDhXNozSsC

The author got started with a DIY flow bench working on the head for the original Mini. A bit wordy but very easy to follow and very good explanations. I'm going to have to get a hardcopy.

 

From what I’ve found Bosch uses thermal real time scans of engines and you can see the large and some-what smaller turbulence but not super fine turbulence. Yes a longer run helps get laminar flow but is not the only way, bends, tapers, and other features can be used to both create turbulence and laminar flow at least for the short run (I like nozzle design stuff). Even with turbulence you have a boundary layer at the wall Im not sure at what scale the effects change to something else. What surface quality is, is kind of hard to get into. Im not a fan of our current systems they tend to be about averaging. I also am not a fan of the N14 or 18 intake tubes or manifolds Im still thinking about the actual runners. Testing a polished vs say the surface I go for, (a fresh 80 followed by an old 80 lightly with a mixed pattern then a clean up to be sure there is no loose material so you get a good final finish with some depth) is probably splitting hairs on a turbo, also when we say polished I think 10,000 grit or so mirror finish lapped with a machined surface.


If you haven’t done a lot of polishing it can also make the surface more uneven it takes me forever to get a good polish on a new project and the process is always new. If you want to maintain control in your cutting but want to go smoother maybe 320ish would be better it doesn’t clog to easily cuts evenly and leaves a nice finish use only new materials old will clog up too fast.


Oh remember on street or even track day set ups your turbo is not always boosting, and its important to get the engine reving as quickly as you can to get flow going so you can boost harder. Also the goal of getting a set up that uses less psi in the intake is a good goal so pressure effects on surface may depend on high psi.


If you want to go for polished what about a coating then polishing it could help prevent carbon build up which makes all this a moot point anyway.


Ok Ive tried taking some pictures of test pieces and its not helpful. I should also say when I use 80 the brand (3M) and amount of pressure are way important. I think I default to a coarser gritt because of its ability to cut and last a while. I tried some fresh 200 wet-dry its closer to my final finish but doesnt have the depth........ arghhhhhh. A good finish is a process that takes time and practice please really try some scrap its not something you think about once you get it, but it involves a certain touch. I dont think you can get to far off if you go slow and there is a range that is useful it is tuning after all.

 

Does anyone know if the N14 and N18 pistons are interchangeable?
I hear the dome is slightly different but would that cause problem?

 

The RealOEM web site illustrations show a different dome for the N18. Descriptions also specify mfgr --- my '07MCS is Mahle.

 

I went CP pistons so not sure. How much did the head get planed I found an 18 thou number but have used 21 thou off on one, also had 2 thou off the deck, there is a good bit of room in the chamber. In the end there is always that note about checking for interference using the clay bits (one good reason for APR studs) and an old gasket but as far as function some builds seem to have used a standard dished dome and I haven’t heard any complaints. The OEM dome is for the DI super lean mode so as long as there is no crashing bits it should be OK. There are more pistons if you look at Peugeot.

What compression are you going for? How smooth did you decide to go on the porting, sorry if there are different ideas, but Ive always at least cleaned up my castings and every good build Ive ever seen running has been ported so Im sure you will be OK. PS Ive always liked a loppy big cam idle on my na engines kind of miss that I could try harder I guess.