PLEASE do the numbers with full radius bells raised off of the plenum floor
This thread I will talk about how we are building the intake manifold for a 1.8T VW cylinder head. One of the many custom pieces going into the Jetta race car: http://shift518.com/showthread.php?t=20487
Starting with what pieces I know I wanted to run; AEB large port runners, 75MM throttle body I got to crunching some sketches in SolidWorks. Solid works is nice because it allows you to quickly design the manifold, see your ideas in 3D and catch any physical design flaws quickly, prior to eating up materials and time. Also included in SolidWorks is the Flow Simulation tool. This allows you to flow test the parts with many different variables, to get a good idea of how the manifold will act in the real world. You can animate it also, and with inputting the cam profiles, timing, RPM, etc. you can actually open and close the "lids" on each runner just as an intake valve would on a motor!
Now to start this, I am by no means an expert at Solid Works, mechanical engineering, or intake design. This is the first manifold I will produce from scratch, so i am way behind the 8ball here, but I have been researching this for a few months now, and reading ALOT of information about this entire process. If anyone has even gotten the chance to use Solid Works, it will take you a week to make a simple bolt! Lots of seat time is required to fully understand it. BUT with some friendly tips from friends (BoxerSix and Mafdark) I am getting the hang of it.
So lets get to the good stuff.
Started off by going shopping. Ross Machine Racing sells all kinds of stuff to build an intake manifold from scratch. I picked up some 1/2" flat stock for the plenum floor, a 2 foot section of their 5" D shaped plenum, some 1/2" stock for the TB plate and a D shaped end cap.
I knew I would need an AEB intake manifold to start with. Idea is to simply use the German engineering they already put into it and hack the top off it! One pass across the band saw and i had just the runners left. After that we machined a slight recess into the 1/2" flat stock to position the runners in and help weld in position.
At this point I had to design the plenum. Now allot of people out there just weld on the D shape plenum cap the end and slap a TB flange on it and call it a day. Sure this will increase the volume of the intake but how equal will the runners flow? Some good reading material with just this information is here to give you a good idea: http://forums.vwvortex.com/showthrea...est-Results***
Notice the Homebrew AEB lower + RMR D plenum. Total flow is much higher than stock but look at runner #1, its seeing far less flow than the other 3, and even they aren’t very close together. That would create a slightly richer combustion in cylinder 1.
So this is where SolidWorks comes into play. I designed the manifold with the AEB lower and the simple D plenum on top and added a 75MM TB to it for an inlet. This is what it looks like. The inlet boundary condition is set to 10lb/s and the outlets are set to ambient pressure.
Notice the turbulence and the uneven flow into the runners.
Now look at what just the tapered plenum after runner #2 and a .5 deg angle on the throttle body does for things.
Same taper on the plenum for the rest, just changing the throttle body angle +2 deg or so each time:
You can see how there is a “happy medium” with the throttle body angle. I am thinking between the 85 Deg and the 83 Deg ones would be the best.
Here are the numbers for the surface goals I set too help prove the changes.
I will pick this up in the next few days. I am finishing up the intake floor tonight by smoothing the transition into the runners and adding some material where its needed. Once that’s finished I am ready to make the final cuts and weld the rest of the plenum up.
Please feel free to chime in if you see some flaws or think I am on the right path.
Build a bunch of manifolds to test changing on variable at a time and I will. I was going to but ended up not having the time at work to do it. I've got a list of about 12 different design variables floating around when you're ready to start cranking them out.
In the mean time, we can at least use what's posted here to get an idea of what works better than others.
What's your opinion on accounting for the starting and stopping of air? BTW, isn't the only time air isn't moving through the manifold is when it's off?
The most efficient manifold would have a TB on one end of a plenum thats the same size as the volume you intend to have for your displacement. On the other end of the plenum there is a funnel that goes to the back side of the valve with a length that resonates to accent the power in the range of RPM you intend to make power at.
Now go build it (hell build me one too ).
As mentioned I have been using SolidWorks for 3 months. Not NEARLY long enough to understand, gather the needed data, compute and assemble a transient flow simulation. Thats what you are all talking about when it comes to fully simulating the actual flow of an intake manifold with the valves opening and closing, in firing order across the runners.
Since your calling me out, I will call you out too. I will pay someone $20 paypal, for them to give me the excel sheet for the transient flow analysis data needed to run the simulation. Firing order, RPM, cam specs and what ever else you need i have.
Anyways, the correct way to flow test a manifold on a computer, you need the cylinder head also, as well as some length of charge pipe ahead of the throttle body... here is the throttle body.
If anyone has an AEB head designed in SW, maybe kindly pass it on?!?!?
read through this thread, i'm sure you'll find some good info:
no offense, but that design is pretty awful, you didn't even radius the runner mouths at the plenum floor.
here is the one i am building for my car. the reason for the turndown is hood clearance issues, but the transition cone from the flange to the plenum flows smoothly with no sharp corners.
No offense, but having a radius isn't enough to KILL an intake design. Yes air will hit a sharp edge and create a low pressure circle of air just under the drop but its not horrid.
Also your design is using a velocity stack, a different deal than just a radius in. 2 different principals.
Man there is so much hate out there when it comes to homebrew intake manifolds!
here is a video of that homebrew manifold in question, the one being hated on. here it made 522 WHP @ 7800 rpm and 28psi with a 35R. otti modified it to suit his needs...... does it work? sure it does. could someone make a better one? sure they can. but can you beat it FOR FREE???? i dont think so.
i happen to know otti very well, and the guy who gave that manifold to him for free sounded like a pretty cool guy too.
sure, maybe it could have done better, maybe not. but that manifold was made in the earlier 2000's or so. it first was in a Corrado with a 1.8T swap. it made ~325 whp with an all stock motor and 57 trim, and we all know that turbo sucks balls by todays standards.
so hate on. when you DO something, put it up. but until then, SHUT IT UP.
If you ask an engineer it will take the latest very expensive software and alot of very expensive hours to build a intake manifold. If you ask your welding supply people what it will take they will tell you that you need a very expensive welder. If you ask your welder it will take hundreds of hours of labor. yea you get the picture
In the end are your results going to be worth the extra cost and complexity? Thats up to you to decide.
I have an intake manifold thread in another part of the forum and a few people came in bashing what I was doing and I have shown proven results with the way I build my intake manifolds. I was displeased.
The main issue that I have with this type of testing (as well as flow bench testing in general) is that the air does not flow in a continuous manner.The starting and stopping and the backing up into the runner and plenum isnt accounted for. There is more going on here then flow.
As for the transient flow stuff. You took a good leap from your seat and keep bringing that up, yet I still don't believe that you can back up your statements with data. My offer stands, I will pay you for the spread sheet for the data to properly accomplish this. If you always compute this information to test your designs, if I were in your shoes, I would have a nice simple spreadsheet where all I need to do is input the RPMS, pick a firing order, change the cam profiles, etc and spit out the information. Seems like easy money if that were the case.
Dont get me wrong I think you have a good start but I dont believe you have made any case at all to justify the extra cost.
What spreadsheets might you be willing to share? And by the way, please don't take any of my responses are coarse. I am just a simple, say it how it is kind of guy. I get to the point quickly to progress forward.
Annnd boom goes the dynamite
Welding inside was a bitch. I need a stubby gas lenses bad. with the stubby back cap I had just enough room in there to get it done. I am jealous of the watercooled Pyrex torch setups that are all of about an 1.5" long total!
Either way, this thing is a beast wont blow apart under pressure and will work well. I also might get it flow benched for the heck of it from the shop putting the head together for me.
Now I can get the fuel rail, and all the plumbing, IC pipes going and inter cooler placement rolling in the car.
BTW its for sale. $650