Backyard aerospace engineers and car forum fluid dynamics professors, sink your brains into this one. Everyone else can skip the boring Bill Nye crap in this first post and have a look at the pictures and read my conclusions.

I had a spare 12a manifold laying around and started looking into ways to modify it. The general consensus that I found:

1) The ACV is the Buzz Killington of horsepower and should be removed, and associated coolant and exhaust passages should be blocked off.

2) Thou shalt not port-match the runners to the block, as this increases reversion.

3) Polishing the runners will adversely effect fuel vaporization.

I saw a few pioneers scattered across the net that were cutting horizontal channels into their intake runners. The concept was that the channels would tumble the air around the surface of the runners enhancing the boundary layer and increasing air velocity. Another effect of the turbulent boundary layer is that it churns up fuel vapor that would otherwise stick to insides of the runner, increasing efficiency. I toyed with trying the same, but came up with a slightly different approach:



Dimples! Yes, lots of shallow dimples in the surface, like a golf ball. In theory this should accomplish four things:

1) Slightly increased runner volume. All those little added spaces do add up.

2) Creates a shallow boundary layer that closely hugs the runner wall, allowing more room for laminar flow in the center.

3) Decreases the amount of flow separation at the insides of the bends by way of "the golfball effect." Basically, the dimples give the fuel/air mixture a little traction as it rounds the curves of the runner, so it doesn't crash into the opposing wall and create a low pressure wake just past the apex of the turn. I know, this really needs a drawing but I haven't made one. Sorry, I can't find my crayons.

4) Look really cool if I ever have to remove my carb in the presence of other car nerds.

So that's my concept. In the next post - implementation!

 

Awesome!! I love it! Good job

 

Looks like a great idea. Stock manifolds are a dime.. well, cheap.
How did you make the dimples ?

 

The first thing I did to this manifold was remove the ACV butterfly. This thing is a huge flow impedance. And as if it just being there isn't enough, I found the one on my stock manifold didn't even open all the way. The nerve of that thing, right? Good riddance. The entire ACV system was removed.



Now we're left with a less restrictive, but still inefficient manifold. You'll notice Exhibit A here has one primary port that is significantly smaller than the other. Also, two pairs of ports have a large channel, while the other two just have a smaller hole between them And what's with this big wall in the channel? I didn't send my fuel and air shooting through venturis just to have it slam into that. Time for some porting:



I used a combination of a dremel and a pneumatic porting tool to do all the cutting and grinding. The result is a much more open, streamlined design. Unfortunately, it's still a piece of swiss cheese with all those holes left by the ACV system. Enter the magic putty, Quiksteel. It's like JBWeld only much easier to work with. I used it to plug up every hole in the manifold save one - the vacuum source for the brake booster. I then used the dremel and a very small spherical grinding bit to grind out the shallow dimples all along the walls. Obviously, this took a great deal of time. I also smoothed out some casting imperfections. The result:





Lastly, I cut a block-off plate out of a sheet of aluminum and blocked off the massive cavity left by the ACV. I didn't take a picture of that part so you'll have to take my word for it. Before bolting up the manifold, I installed freeze plugs into the block to seal up the coolant passages.

In the next post, how does it run?

 

I very seriously contemplated cutting the channels into my runners, but haven't, yet. I am very interested in seeing the results of this. Document it precisely and effectively please

 

I'm currently in the final stages of building a special modified Nikki to go along with this manifold. I chose to try the manifold first with my stock Nikki so I could see the difference in just the manifold mods and to more easily narrow down any problems that arose. My stock Nikki is just that - stock. The rats nest has been removed and the secondaries are mechanical, but everything else is original. It's been rebuilt and has been running beautifully for a long time so it made for a good test platform. My car, an '85 GSL, is a stock port 12a with 127k miles. Exhaust is a Racing Beat header collected into a Magnaflow muffler, which dumps out of a passenger side pipe. Loud. Fuel is (was) delivered by a cheap Carter pump with a Holley regulator.

My biggest fear when starting the car after the installation was a vacuum leak. There were a lot of holes in that thing, and some of them were not easy to find. Happily, it fired right up and idled smooth. The only clue that something had been done was the exhaust note, which was even louder than before. Go figure. On the road, the car ran great. There was a noticeable increase in power, particularly in the low and mid range. The car didn't bog at all putting around the neighborhood at below 2k RPM, and at WOT it pulled hard from 4k up until about 6k. The top end seemed a little stronger, but was less of an increase than I expected.

The next weekend I got a chance to really test it at one of the regional autocross events. The last event I'd run the car in was with the stock exhaust manifold, so the addition of the header and modified intake made for a much better experience. On my last couple of laps, I was able to top out 2nd gear at WOT. I noticed a bit of a sputter and power loss above 6k... it felt like the car was running lean. The drive home was further evidence - the car cut out at only half throttle. I'd burned up another one of those damn Carter pumps! I got my wideband hooked up and was able to see that the car was running very lean through most of the powerband. AFR's would hover in the low 14's until the carb ran dry, at which point the gauge would max out.

So tonight I installed a Holley Red pump and a new filter. It screams all over the powerband. AFR's are now in the low 11's up past 5k RPM, leaving me some more room to tune. I think the manifold is a huge success. The improvement is noticeable from 3k all the way to redline. I can't wait to put a heavily modified "SuperNikki" on top of it. Look for that thread in the near future.

 

I can only do so much documentation without a dyno, unfortunately. I would love to get this car on one, and if I had an entire day I wouldn't mind bolting up my factory manifold to see the difference on paper. But for now, "it feels faster" is gonna have to do

Thanks for the early compliments, everyone.

 

Yeah, I wasn't really referring to a dyno, more economy. That was my main focus when looking to do this to my old DD 85 GSL with a Nikki. I have also thought about doing it to my dellorto on my SA, and I may in the near future. Just want to get it running spot on beforehand, so I can see exactly how much of a difference it makes.

I was also rather buzzed when I made that initial post

 

I don't really have any benchmark for comparison when it comes to fuel economy and this car. It's an autocross and weekend car, and I drive it like the world is ending. It also never sees a full tank since I like to keep the weight down for events.

I think you should give it a try though. Part of the theory is that it should improve atomization of the fuel, which means less is wasted per volume of air. So it might boost fuel efficiency in that regard. However, another thing to consider is that it should also increase intake velocity, which would create a stronger signal and pull more fuel through the carb. I think this mod is definitely geared toward performance and not economy, but it's worth a shot.

"Proper testing procedure":

 

My car always feels faster after I detail it. Seriously.

I don't place a lot of faith in the calibration of anyone's 'rump dyno,' including mine. It's pretty much impossible to be objective about the perceived results of hours of loving labor implementing one's own brilliant ideas, absent some sort of hard numbers.

Doesn't have to be an actual dyno, though.

Taking the averages of several before and after timed runs up a specific hill, for example, is a good measure of relative engine power gained or lost.

Relative improvement in throttle response can be measured using a video of the linkage and a test tach; at 30FPS, counting the number of frames it takes RPM to rise to the same number after throttle input is a pretty precise yardstick.

Not saying your theory isn't sound, but data trumps theory every time. I'm a fan of data.

 

I thought golf ***** were dimples so they could get through the air faster and straighter. Someone tried this with a car and it didn't help any. I don't think this would do much besides tumble the air, I believe for dimples to work correctly, the object has to be a sphere. I think mythbusters did something like this?

 

that someone was Mythbusters but I don't recall the results

 

this is true. its actually documented fact, you will perceive your car to run better after the windshield is clean.

 

of course it clean make it slicker cutting through the air lol

 

In that case, my car should've felt slower because it hasn't been washed in ages!

The Mythbusters did test golf ball-like dimples on a car, and found that it gave them a significant increase in gas mileage. Check it out: http://www.yourdiscovery.com/video/m...s-dimpled-car/

I don't think timed runs are a good benchmark, unless done at a track with proper timing and identical weather conditions. Where I live, there is no such thing as "identical weather conditions." For instance, the temperature is the about the same today as it was yesterday, but the humidity is about 40% lower and the winds are gusty. There's just too many variables, and the change from just this one part is going to be smaller than the margin of error. Not to mention the effect of removing the ACV could be greater than any effect of the dimpling. Nothing short of a set of dyno runs is going to give us hard data on it.

That being said, I think the car is faster The effect of the dimples themselves may indeed be negligible, but that's the way most power modifications are - improvement on these cars typically comes from the sum of a lot of small modifications, not one big one (unless you throw on a turbo.)

 

As Lizard FC stated in his origional post, the concept behind the dimples is to create a smaller wake. In terms of the golf ball, the wake takes more energy from the golf ball than friction caused by air, thus dimpling it will allow it to fly farther. I'm still taking my fluid dynamics classes, but to my understanding the theory behind this mod is acceptable. You could set up an experiment with teledeltos paper to see how much of a wake is created in the runner...

 

Very true... and ultimately, if you think your car is faster, and it makes you happier, then the mod did as intended, right?

Just saying that any kind of controlled, repeatable test using fairly precise measurements (like time over distance, or RPM over time) works better than "butt-dyno" for relative measurement of improvement. It doesn't give you precision or absolute numbers, but it serves for comparing before/after.

Hey... maybe the dirt on your car is providing a 'dolphin-skin' dimpling effect!

 

The dimpling seems to work when air surrounds the surface, not sure about the other way round. You'd think that if it really worked, racing bodies would have been doing it for years....

 

So, since Mythbusters proved it to work, the real question is, who are we going to get to start creating dimpled body panels for SA/FB?

 

This will work....^
Next we need to find out at what speed it will actually help.

Need to take it out to the salt flats........The only place duct tape and a hammer will make you faster.

 

Divin, you're right... I think I'm happy, therefore I am

There are companies out there started to test stuff like this. I found it intriguing that a cable TV show, and not NASA engineers or a car company, were the first to scientifically test it! And just because no one is doing it yet doesn't mean it couldn't be practical. Case in point, google Singh Grooves.

I doubt you'll ever see grooves or dimples in production manifolds because manifolds are cast cheaply. Small details require better, more expensive manufacturing. Car companies are more focused on cutting production costs than producing better cars.

 

An update on this:

I burned up yet another cheap Carter OEM replacement pump right after installing this manifold. So I switched to a Holley Red. This caused the car to run far too rich up top, to the point where it would bog - even with the FPR turned all the way down below 1psi! No matter what I tried, the stock Nikki, that manifold, and that Holley setup did not like each other. It seems like the unmodified Nikki had become a bottleneck.

So I removed the stock Nikki and installed the experimental Nikki that I'd been working on all summer. After some fiddling and breaking in, I must say that the result is very pleasing. The car now pulls hard well beyond 7k. I have to lift around 8k because the car just wants to keep pulling. This is on a stock port, mind you. I'm going to do some accel pump tuning, advance the timing, and possibly rejet the primaries. I hope to get this on a dyno sometime in the future.

 

thats cool hope to see what you get out of it. i may have to do something similar sometime along the way.

 

so thats what those dimples are for. When i took off my lower intake manifold i found the same dimples in the pipes illl take some pics when i get back from vaycay. i thought some one tryed to port them and failed.

 

The results of this are intriguing to say the least. It's a shame the ITA Rulebook would probably never let us get away with this.

Not only that but we don't have a spare intake manifold to fart around with right now.

+1 for ingenuity, that's for sure.