Rattling around 2.9 intake idea

[RangerAx]

Okay...let me start by asking to spare the "you can help the 2.9 by swapping in a v8" comments.
That being said...please tag along my little ride and detours to get to my point/question.
The other idea, amongst my frustration of the 89 sittin in the garage for a year and a half and at times, seemingly no hope of getting anywhere anytime soon, I decided that while I have no idea where money will come from to do things, maybe I could start tearing her down and prepping for some of the work I HOPE to do in the future.
While thinkin on it, I'm starin at the engine and taking mental notes of what I need to do and what I'd like to do and I'm staring somewhat blankly at the upper intake manifold when the thought crosses my mind about the throttle body from an 86 I had to crap because it was cracked and warped. I've actually spent time searching EBay for throttle bodies from just about anything that look like they might be somewhat compatible. In the past, I've been frustrated by the limitations on functioning cowl/ram air hoods and the like.
So finally a little closer to my point...it dawns on me that the upper intake is kind of ackbasswards...that is, air comes in the TB, down the upper and has to turn back on itself in the upper...when spoing, I realize .."hey, those down tubes on the intake are about 2" x 3"... "..so you can guess where I'm going I'd bet. So here it is: obviously I'm not real short on imagination..but I lack some of the tech savvy. I'm looking at this intake thinking about clearances and how could I build a new intake manifold short of casting (or should I say..build a redesigned upper). And it would seem to me that 2 x 3 rectangular tube with 1/4 or 3/8 plate cut and shaped to sit down on the lower, with pretty much whatever sitting on top of it, would replicate the outside appearance, with a more straight in approach, could be made shorter and customized. Then I start asking myself, what about all the other crap...the idle air control the vacuum lines...so here's the question for you guys and gals. It can't really be that simple can it..what would have to be done to make all those things work internally?
My thinking here is...if it's doable, KISS it, and make it easy enough to reproduce it, if not to sell it, then at least to be able to explain it to the next guy to go it (say on here..go figure of all places).
I guess what I'm sayin here is..edumicate me a little on the doability of this thought, yay, nay, maybe, you need to do this but not that, etc..and if it's doable, brainstorm on the best way to accomplish it? Cuz my brain is still stuck on ..2 x 3 uprights welded to 3/8 plates with a tube on top made anyway I want it....

 

[MAKG]

Designing a good driveable intake manifold is not at all that simple. Air can turn corners fairly well; it's not mixture (THAT'S a problem unless the fuel is completely evaporated, and it never is).

Even in the 80s, it was conventional to design these using a rather involved technique called "Computational Fluid Dynamics" (CFD). When you're spending upwards of a billion dollars rolling out a vehicle, you can afford a few million to run the supercomputer models to make the thing drive decently. These focus in routinely on the manifolds, as temperature flow can affect them significantly (and that's a standard CFD output).

Sharp corners tend to make drag, but more importantly, a wide open channel flows like molasses. It's counterintuitive to a lot of people who haven't done any CFD, particularly unsteady CFD (unlike the flow bench model, a real engine DOES qualify, and that difference is truly huge).

An intake manifold that is wide open often works just at really high flow, and can't do a damn thing at more standard street flow. An eyeballed intake manifold may not be able to ever see the flow it's best at because the engine just can't ingest (and doesn't need) that much.

People do NOT guess at these. There is a reason they are complex castings. If you guess at it, you will most likely make an undriveable POS. And given your R&D budget, it's not very feasible. Now, your desktop computer may be able to outcompute what Ford had in 1985, but you don't have the models. Complex boundaries are a HUGE amount of work to generate, and if you get it wrong, it will just be nonsense (in particular, you can generate fake viscosity in the flow from badly sampled grids).

And by FAR the most important engineering concept is, "if it ain't broke, don't fix it." KISS is very important, but it's hard to be simpler than doing nothing.

Now, you may be able to streamline the intake manifold (that's an aggressive porting job), but you still don't have the budget for that, nor the motivation that it's necessary.

A properly used 2.9L is a fairly well designed -- and powerful -- engine for a small, relatively simple multiport V6. It gets just about 1 HP/CID, which is not bad at all for a naturally aspirated gasoline engine. You aren't going to do anything but screw it up by guessing. It's not a 454 big block that has been put together by a couple of rabid monkeys.

 

[RangerAx]

Alright..input appreciated, but allow me to say this...I do know what CFD is...not to say I know how to use it. I also understand how much design goes into them..further I understand the efficiency of the 2.9.
Here's the one counterpoint I do have...if they do such a bang up job at the factory...why are there so many aftermarket manifolds for other engines including ones made by Ford...not to mention headers.
Have you read the story about how Edelbrock got started? Now today they surely do use CFD, but back then...nope. Am I comparing myself to the like of that start? NO...let me make that clear....NO. Only point I try to make with that is...a better design CAN be had without CFD, which yes, you did allude to.
All that being said, I am not suggesting that my guesswork could not be a total failure..in fact, I'd be willing to wager I'd have a better shot at failure than success at the very least on the first go around.
My question was seeking input and ideas, on ways to expand the options a little on an admittedly already good engine....and while I can not argue the logic of CFD or the lack of logic in guesswork...I can argue that I could have done without what (at least to me) sounded like a parental lecture pointing out the futility of any effort whatsoever that went IMHO WAY over the top.
I'm not looking to start an argument, and rest assured I will not waste any more of my time on this one...but next time, maybe you could simply answer the question before you, and actually try and teach me something useful like how the IAC and vacuum ports would have to be handled, even if ultimately my goal is unattainable particularly without any budget (which I fully admitted to).
Bottom line: Spare me the holier than thou attitude...I'll fully recognize from this and other posts your knowledge and ability, but I refuse to accept being belittled, which sir, again IMHO is exactly how your reply reads. Tech advisor for TRS or not...you are better than me, and while more knowledgable on some subjects than I, certainly not more intelligent.

 

[MAKG]

The IAC and vacuum ports would need to be handled exactly like they are now. A bypass across the throttle plate and a port into the plenum.

CFD is not the only tool out there, and it does have its limitations. But whether or not you want to hear it, the alternative is NOT just guessing. There are several techniques, some of them more effective than others. Use of a flow bench alone, for instance, can fool people (it's steady flow -- engines are not).

Streamlining means something very specific, and you missed that. It means making the boundary follow the flow -- i.e., the streamlines. It does NOT mean making an entirely new path, nor does it necessarily mean rounding all the corners. An effective technique is to keep the flow velocity constant throughout the port.

I'm certain you don't want to hear it, but you won't have a significant chance of success anytime soon.

And how the heck are you even going to know if you don't have a dynamometer at hand?

The myth that a lone tinkerer with no budget can produce anything significant in a mature technology is just that -- a myth. It's not done that way.

 

[RangerAx]

Okay...truthfully I didn't even intend on looking back at this thread, but you know what killed the cat...
I very much did not and still do not wish for this to turn into a thread filled with nothing but arguments, but please allow me just a little leeway to demonstrate a point and then I will cease and desist as it were , because I for one and more interested in conversations and debates as opposed to heated arguments (even if you or anyone else insists they aren't "heated").
A myth about a lone tinkerer with no budget, fair enough and I can't argue it word for word, but I can raise some stories of success that aren't far from it, and let me just preface that by saying that in now way am I attempting to compare myself to these people on a direct level.
I've already mentioned Edlebrock (and yes he started that way long ago and now they use far more advanced technology to do what they do).
How about James Duff, or Magna (an auto parts manufacturer in Canada )?
And then there's the incomparable Jack Roush, as well as a still mom and pop operation named Trim Tool And Machine Inc in Ohio. Oh..one more, OCC.
Now obviously these did have budgets and in some cases such as Mr Roush experience and far more backing then I could probably dream of. That being said, I'll also happily admit the applications of said examples aren't necessarily related to my particular idea.

Edit: lone tinkerer in mature technology....I thought that's why I was on TRS posing this question, to seek the ideas and thoughts of others that might CONTRIBUTE to the forward progress and not just seek to find all the reasons it will fail. I'm perfectly capable of finding fault in my ideas all on my own...thus being so straight foward about a budget or lack thereof among other things. (end edit)
So then, this brings me to this: I've never entirely believed the notion that "you won't know if you don't try", because as two grown men, I think we can at least agree that sometimes you do know. My projection for success? Meager at best I am sure, but please don't pander to me about being sure I won't want to hear it. I am certainly open minded enough to accept criticism, that, for example, could be better put "it would be extremely difficult for you to design better flow characteristics without CFD or a flow bench".

All this being said, truthfully, I'm not even sure how well or accurately I've expressed why I took offense to your reply. I will go so far as to say, I may have taken to much offense to it, as TRS is not the only thing in my life (can you say "teenager"..mine's been a pain). As such I do feel an apology on my part is necessary for some level of over reacting.

So then...on the IAC and vacuum, simple enough and I'm glad that at least is as simple as I might have thought. As for guessing at a better flow design? Well if I can't do it myself, makes sense, but I be able to find some help in that department if I look.
I think ultimately the bigger goal would not even be to improve the flow, so much as to keep it reasonably close to stock but allow for a wider array of options on modifying the intake system as a whole. Guys have modified MAF systems to a 2.9, and looked at mustang throttle bodies and so on. The throttle body itself for a 2.9 is efficient , but as we know, earlier ones have a 58mm, and as I'm sure most of also know, that 58mm is hard to come by...wouldn't it be nice to be able to grab a 4.0's or something and throw it on there? Or maybe for the most daring , design an intake that would make a turbo or supercharger 10 times easier to set up? Fantasy? sure it is, but sometimes great ideas start from dreams (and surely more often nightmares of failure to). And rest assured that if some form of idea came about that was more than just remotely realistic, I'd find a budget somehow, not to mention a dyno.
So.....maybe we can start this over? (I hope) and try a simpler form of the idea, keeping in mind at least some of what's been mentioned about how it would or would not work.
Let's suppose, you wanted to build a new upper intake that turned instead to the driver's side (no, offhand I have no idea why you'd want to), you could duplicate the outside appearance with 2 x 3, but as you mentioned, a more channeled approach would be better suited to maintain more or less factory flow, as opposed to a great big open 2 x 3 tube. Okay...rattle that around, I think that could be achieved with some work, maybe by using round tube inside the rectangle? Not sure there. Just hack off the existing manifold and work from there? Im guessing (yes guessing) that would not go well in trying to weld new portions back onto it.
You know...it just dawned on me that Eaton superchargers at one point made one for the 2.9 that I think actually fed air into the back of the manifold to allow for the pulley up front...wonder if I can find a picture of that at least...
okay..enough fantasy for one post...

 

[RangerAx]

ooo..ooo..quick thought I had..more a question simply on the basis of the theoretical construction: If you could and did use 2 x 3 tube, how well would that in fact work for heat issues and what not since it's not cast...would you have to try and find 2x3 aluminum tubing as opposed to steel? a certain grade? Is there a reason anything other than a manifold that is cast would causes SERIOUS issues?

 

[MAKG]

A cooler intake charge will make for (a little) more power, but you can probe your computer and find out what the intake charge temperature is right now. I strongly suspect you'll find it's ambient under heavy-throttle conditions. Which means cooling doesn't matter. Modern intake manifolds are made out of plastic anyway. The reason for using aluminum was not thermal conduction (though aluminum is just about the best thing for that). It was weight. Aluminum is quite a lot lighter than cast iron. Plastic even more so, but you have to design one that won't leak and will last (which is why it's recent).

The avenues you're going down have already been tried. That's why I argue against trying to improve a mature technology. Your examples are mostly improvements in marketing and distribution, not technology. Nothing James Duff did (past tense -- he's not doing much anymore) was an attempt to beat the factory at their own game. His experiments in suspension were to adapt a vehicle to where the factory didn't intend -- and he met with mixed results. This means it wasn't a mature technology.

Same thing with Offenhauser. He did his work in the 30s. CFD was done using 2D patches on idealized surfaces, using slide rules and big piles of paper. Experimental results were VERY new, and centered around airfoils. CFD as a volumetric technique wasn't tried until the Manhattan Project, and the impracticality of their approach even with spherical symmetry is one of the strongest reasons why we have computers (the other big one -- more widely known -- is cryptology, though that could have been done with special purpose computers -- CFD requires general purpose ones).

Most of the "improvements" people try aren't. The change in peak power from 1987 to 1988 (when that big throttle body went away) was zero. You can't evaluate them without objective measurements somehow, and scientific method in their investigation. This last part is far more important than it may seem -- there is a well known effect in science called "confirmation bias," which means you're too likely to get a positive result if you're aware of the incoming results during the test. This is the reason for double-blind tests.

You've chosen a VERY difficult location to try stuff out. You can get some feel for it perhaps with a desktop dyno (it's a lot cheaper than casting a lot of aluminum manifolds). Free-flowing intake manifolds trade off driveability. And making a driveable, efficient, powerful engine was the factory's goal.

But one place you might do better is forced air, as the factory clearly didn't try that. Though be VERY careful to get it right the FIRST time or your engine won't last at all. And research the issues for getting this on the road. And speed-density EFI isn't a good candidate for that (they were never designed for an intake manifold under pressure, so they go out of range), so it's going to be tough.

Most importantly, prototyping is always expensive and time consuming. It simply must be modeled first or your duty cycle will be so low that you'll just make a big pile of aluminum scrap. Investing in a desktop dyno would be a very good starting point here. That would make the prototyping directed, at least. Undirected prototyping won't get you anywhere.

 

[Psychopete]

Interesting topic. I didn't read through all of it because my computer is hooked up to a 32" CRT TV right now, and it's a bitch to read anything even at 640x480. Long story.

2.9 MAF computer is available from 90(+?) california emission Rangers. It's a strait forward install, especially in 88+ Rangers. I bought the PCM from a place in California called Subway Truck Parts for about $80. No way in hell I'd find one here .

But I did you see mentioned the ability to put a 4.0L TB on a 2.9L. I was going to S/C my 2.9L with a 5.0L MAF & TB, I have a ~90% complete adapter kit for the 5.0L TB that I need to get rid of that's just setting in a box if you are interested in it for experimentation. I've since decided to build a 5.0L due to me wanting something that was fast and would last a long while.

Pete

 

[BlackBII]

That was a very interesting read, I love how Gross's responses are always so informative, yet in a way that you wouldn't expect.

It seems that modifying/creating an Intake manifold without proper tools would be tiresome and irritating. Without a desktop dyno, how will know that your design will flow better than Ford's? You would have to go through all this work and if it's a success, then great, but if not, and the truck is no longer streetable because the design was flawed, then you wasted your time/money.

But if that's what you enjoy doing, trying new things and seeing things for yourself then why not?

You referenced James Duff, he experimented with the suspension etc because it did not work well for his application, and there was room for improvement, Ext. Rad arms, Steering geometry etc. But if something works, and works well, why try and change it?

The 2.9 is an awesome street motor in my opinion, works well for rbv's. And like Gross mentioned, it would be tough to get a manifold to perform at a streetable level better than Ford did it in '85 without all the Technology and Funding. Sure, you could make it flow real nice at 5,000 rpm, but will you be drag racing this motor?

Why not look at what you can do for the 4.0? Which is an already improved 2.9.

Now, Im not trying to argue or anything of that nature, just stating my opinion. I am also interested in the topic, as there is always a better way to do everything, history has proven that. The 2.9's intake could be improved i'm sure, but, I would leave that up to the new technology they have nowadays, not some innovative thinking in a garage.

I'm no engineer, just my opinion man.

 

[BlackBII]

But using some 2 x 3, some plate and a welder does sound like the best way to go about it in a garage, can you Tig weld?

I have made intake manifolds before(for Forced Induction, custom applications where it was required) and I tried to make it smooth/straight as possible, slightly larger runners than stock(had a slightly larger turbo/supercharger) and a close design to what it was stock, as the manufacturer usually does a pretty good job, but, without having any background in Flow Dynamics and such, I have no idea if they performed better or worse than stock, but a 10.5 sec 1/4 mile made me happy.