Help balance air flow in lower intake

BKB

Registered User
I have been cutting up my lower intake and trying to improve the flow. I dont have anything fancy for testing just a gas powered leaf blower and smoke. I first tried a stock set up and it defiantly favors the passenger side of the intake and also the front ports. Also you can feel the air only comes out of the top 1/4 of the ports and is directed straight down towards the ground. My lower intake has a raised roof and has now for about 10 years. I duplicated the triangle inside like a stock piece. With the raised roof you can feel the air is exiting threw the whole port and it comes out in a angle the matches the intake ports of the head. I spoke with Coy years latter when he dyno tested a raised upper and he said it was useless up until about 6k rpm.
The problems im having with my lower is that the front ports flow like crazy and the rear ports are a little lazy. The rear port on the passenger side is by far the worst. The side to side flow has improved greatly over stock, overall airflow is no comparison the modified intake just flowes many times more then the stock. My question is, the way im testing this does it prove anything. Beings this is a roots blower and it creates boost by pressurizing the whole intake tract does blowing air threw it prove anything? Anyone have any ideas how to improve flow to rear cylinders???

Thanks
Brian
 

Attachments

  • INTAKE 001.JPG
    INTAKE 001.JPG
    237 KB · Views: 296
  • intake 1.JPG
    intake 1.JPG
    169.6 KB · Views: 317
  • intake 2.JPG
    intake 2.JPG
    69.3 KB · Views: 262
How are you so sure that the rear cylinders are not getting enough air? I have found that my back cylinders always run lean compared to the fronts. Even with a raised top manifold like yours.

I have surmised that it may have to do with a pressure difference in the manifold. Perhaps there is more pressure at the rear ports than the fronts so that even though they flow less, they are flowing it at a higher pressure than the fronts.

How exactly your manifold is ported has a direct effect on this too. When Randy made mine he did not angle the entry towards the front because he observed a similar problem that the smoke path was showing a favoritism towards the front cylinders. He purposely made the manifold to not favor the fronts and like I said, my front cylinders ran a little richer than the rears.
 
The only way to have an effective flow split is to do so within the vicinity of each port. Splitting flow in the return plenum with an open intake manifold will just regroup the air when it enters the manifold itself.

Air collisions create more turbulence than physical contact to some edge or wall, so it's something that needs not to be ignored.
 
Randy Baker had a thread on the subject. Hes done more testing than anyone I have heard of. He installed a divider in the lower part of the plenum and showed smoke blown through it with and with out the divider. The smoke with out the divider tended to be heavier to the outside curve and with the divider it was more even. I beleve the idea was the passenger side rear port got a better air charge than the drivers side. I was impressed with his mod and did the exact same thing to my lower intake. If I am wrong than Randy can jump in and correct me. I have done quite a few mods, just to my own SC after reading Randys posts.

Ken
 
I have two new different manifold design.What David D. has.
I need his back to update.Its outdated. I have updated the
stock casting design.I don't think it can be improved anymore
by me. And i have a sheet metal manifold I hope to be testing
before spring.

With staying in the relms of the stock manifold / intercooler routing
design.And raising the top of manifold. I don't think it is possible to
get a perfect balance.

But there are steps that can be takin to improve on the stock
manifold/plenum casting. On the air splitter in the plenum.
your looks like the splitter both ends need to be longer,
and centered. Could not see the bolt flange end of splitter..
But it may work better if splitter protrudes into the manifold past
the plenum bolt flange. And take the top off your manifold
and weld in 10 directional vains. And raise the floor. And the top
of your manifold needs to be higher.

Turbulance in the manifold is the biggest problem that I see
that hinders linear air flow to the heads intake ports. I
don't see balance as the biggest culpret in robbing horsepower.
but its there.
 
Well I'm not about to cut the top off and do any major reworking of the intake. This is a show car more then anything, it only needs to produce my personal goal of 400rwhp and that is not a problem. I see what you are saying about the divider protruding down into the intake but i am pretty happy with the left to right balance as it is now. Unless you are saying it is producing turbulence. I am going to play with some modeling clay to see if i can come up with a simple solution that will help direct air into the rear ports. If you have any other ideas you would like to share please let me know.
 
Are you guys considering how air stacks up behind the valves that are closed when looking at the way air flows through a port? I would think that, as well as the sonic effects of the valve closing play roles in this as well that go beyond simply how the air flows through the full open manifold.

I'd say we need to find a engineering student or someone that's currently working somewhere with access to modeling software. They can model valve events, cylinder filling, and air flow though the manifold with the proper measurements. i've seen a few things posted on other forums for different engines, wish we could do it for ours. Maybe when they work on the new head design for our car they'll do some modeling.
 
I don't know if I can explain this or not I will try.

Build weldup filling in both sides of the manifold outside of
plenum boss and the rear port on each side of manifold.
This will give you enough aluminum to shape the inside of
plenum boss & inside of manifold.you can move these walls
back after the weld build up.To expose the 3 and 6 cylinder
port.

I have one of the first raised manifolds that Charles W. of MP
built.I am told this is where COY M's came from. I bought mine
from Jim Demmit. They did not flow to well because of the stock
shape and size of plenums. Plus I seen zero attempt to do any
type mods to the insides of the manifold, except port matching
and removing the triangle. When I first made a raised manifold.
I did not know that it had been done before,until I bought Jim's.
Then about a year later I was talking to Charles Warner of MP.
And he told me he had made and designed the first raised manifolds.

Charles Is the only person that knows what design changes I have done
to the first raised manifold he built.This will be my last time I modifi
a manifold, And It was done on the first raised manifold built by
Charles W. I have done a little motor testing. and I don't think
I can improve the stock manifold/plenum casting's any further.
Than my last manifold # 14. After modifing this many castings
I have learned to switch to a bolt on top, instead of starting over
on building another manifold to make changes..And being build
with removable top. it will be easy to manufacture a casting.

Randy

Thanks Randy
 
Last edited:
Are you guys considering how air stacks up behind the valves that are closed when looking at the way air flows through a port? I would think that, as well as the sonic effects of the valve closing play roles in this as well that go beyond simply how the air flows through the full open manifold.

This was what I was thinking. But this would only be chaning while checking on one CYL since it will have constant pressure once the rest of the valves are in an event right?

Chris
 
I don't know enough to say for sure. But pressure waves created by the air bumping into the back of the intake valves, other pressure from the closing of an exhaust valve and the fact that they all are moving in some sort of pattern can lead to flow differentials that are hard to see without computer modeling. Now folks were building very good intake manifolds before computer modeling based off of trial and error and skill. Computer modeling just makes it easier to figure out for those that don't want to do the trial and error or those that don't have the skill.

Lucky for us there are a few folks that have skill. I just was wondering out loud if the impact of the valvetrain and pistons was being considered when looking at supported air flow patterns.
 
Functions of air inside of the cylinder head itself is quite basic. All of this can be determined on a flow bench, which is part of the reason for working a head while operating a flow bench. Turbulence throughout the whole intake system can be plotted, amount of reversion, choke points, and of course flow.

Reversion is something that I found to be a huge thing with all forced induction. Whenever that valve slams shut, air bounces off of the back of the intake valve and collides with forced air entering the port. Reversion is the most detrimental cause of turbulence. Ways to rid of reversion is port volume, cam specs (mostly ramp speed/angles) valve size/seat angle, and mostly intake manifold runner shape/volume. Since the SC is an open plenum design, then reversion will never quit.

Pistons don't necessarily effect the flow, but of course bore size does. In small bore motors, the cylinder becomes the shrouding aside of the combustion chamber. Narrow bore spacing, like Pontiac, actually put valve reliefs in the block to compensate for shrouding. SC heads are canted around 12*, so bore will hurt low lift flow more than high.
 
I don't know if I can explain this or not I will try.

Build weldup filling in both sides of the manifold outside of
plenum boss and the rear port on each side of manifold.
This will give you enough aluminum to shape the inside of
plenum boss & inside of manifold.you can move these walls
back after the weld build up.To expose the 3 and 6 cylinder
port.

I understand what you are saying and was thinking of doing that. Thats alot of welding and i would think it would really warp the intake. I guess its such a flimsy thing that bolting it down will straighten it.
 
Casey brought up a great point (he usually does)

One thing we were always stuck with when regrinding cams is teh ramp angles are what they are as is lobe seperation and a slew of other specs that just cannot be economically altered to ones specs when regriding a cam.

Now that we have the option of new cam grinds there are limitless possibilities as far as what we can get. (or close to it) Smaller cams with better streetability characteristics making the same power as the larger regrids of yesteryear is just one OBVIOUS advantage. However cam companies put hours and hours of research to see whats best for an application....How much they have done for a positive displaced blown 3.8 I dont know...Usually though the cam companies as well as many engine builders have software that can at least test the changes in certain parameters have..And some know from having 15,000 worth of junk parts lying in the corner that didnt work.

Much of our knowledge of what has worked is based on regrinds for the most part..And all we could play with was lift, duration and cam timing...

Anyway I have 103 fever and am feleing a bit loopy...But still..A good topic to discuss;O)
 
Any benefit to splitting the intake into two halves? I mean to totally take the devider from the rear plenum and run it down into the lower all the way to the front of the intake. I would think that you would be able to take a plexiglass cover and watch as the smoke changed its path from a large path with what seems to be a promotion of swirl, to two smaller more stright shots. I am not sure how much low speed would gain, but no doubt there would be some...with the top end suffering some as well.

Chris
 
One thing we were always stuck with when regrinding cams is teh ramp angles are what they are as is lobe seperation and a slew of other specs that just cannot be economically altered to ones specs when regriding a cam.

Now that we have the option of new cam grinds there are limitless possibilities as far as what we can get. (or close to it) Smaller cams with better streetability characteristics making the same power as the larger regrids of yesteryear is just one OBVIOUS advantage. However cam companies put hours and hours of research to see whats best for an application....How much they have done for a positive displaced blown 3.8 I dont know...

Actually that is not quite accurate. The same profiles can be ground on any cam core. "Billet" cams as we know them are not blank in the first place.

Any cam goes through several steps from a true piece of billet to a finished cam profile.

1) The raw billet is machined into the general layout of a cam application. This means the "valleys" in between the lobes and the general shape including the snout, any gear drive, and lobe/journal spacing is determined.

2) Then a basic pattern is ground into the cam. This step puts the firing order into the cam and generally locates the valve timing events. After this step the cam actually looks like a cam.

3) Then the cam is hardened. The depth of the hardness is partially determined by the process used and the alloy the billet was made from.

4) Then the cam has the final bearing journal sizes ground and polished.

5) And finally the end profile is ground into the cam and it is finished.

Now what this means to us is that ALL cams for the 3.8 are regrinds. That's right folks, there is no such thing as a true billet for us. The closest I've ever seen is a step 1 cam. In fact I have one such cam and if someone ever wants to change the firing order on their 3.8 I can do it. The final cost would be about $750 though.

See when people like Comp Cams grind a cam, they always use a pre-finished cam. For the price that we pay for a cam there is no way Comp (or anyone else) will go through all the steps to grind a true billet. What you are getting instead is a cam that was originally intended for maybe a marine application or something like that but is still new and is then re-ground into your profile of choice. This is why even a "billet" cam needs longer pushrods.

The difference is that the cores used by the cam grinders is closer to a performance grind than our stock cams ever were and the heat treat is much deeper so there is less fear of actually cutting through it.

So technically yes, there are some profiles that can't be ground on a stock core that can be ground on an aftermarket core but those are few and far between and don't fall into the category of a street cam such as people are likely to use in their SC. The biggest plus of a new cam is a more accurate ICL for installation purposes. Due to the lobe placement on a stock cam the OE regrinds end up greatly advanced in some cases and LSA is limited to 110-112 deg depending on core and chosen lobe profile.

;)
 
I would think that you would be able to take a plexiglass cover and watch as the smoke changed its path from a large path with what seems to be a promotion of swirl, to two smaller more stright shots. I am not sure how much low speed would gain, but no doubt there would be some...with the top end suffering some as well.

Chris

I think this test would be better to use food-die as as marker
to map air path, Wet flow test. Then strong and weak ,turbulant
air signals can be veiwed after air test in finshed.. Using smoke
is a good visual test, I do it myself. But wet testing with dies
will give lasting results that can be studied long after completion
of the air testing.

BKB you may be able to use your smoke tester to wet test.
make a spay bar across the plenum. To inject food
die. Plastic tube would work . Then do a test. Its
best the have all parts being tested with die to be
sanded to a slight polished surface.Surface reads easier then.
And die clean up it easier, if you want to retest ..

Randy
 
I understand what you are saying and was thinking of doing that. Thats alot of welding and i would think it would really warp the intake. I guess its such a flimsy thing that bolting it down will straighten it.

It would take around 2 hour,including preheat,weld build-up,cool down.
depends how your set up.

Yes you would need a jig to reduce warpage to do any welding.

I use an SC3.8 block with heads. Bolted on enginestand. Then
have a heated under block .with no oil pan. I have the heater
enclosed up to the oilpan flange. Its just a tin insulated box
with no top or bottom,I made. This lets me pre-heat and
control cool down , and to stress releave. And I weld using
purge chamber or and purge blockoff walls for back of welds.

I perfur a manifold to be total weld build. The epoxy route
is a great time saver and works great.

What you have done with you manifold will help out nicely
for top RPM power.

Good luck on your project Randy
 
Last edited:
Dave's right. Points out something that most don't realize.

A 100% true custom Billet cam has to be done from a blank itself. Yes it cost $750 just to have the cam ground within a certain spec, but you have to buy the blanks first. Most companies supplying blanks will only sell you atleast two at a time. Cheapest I have ever bought the blanks for were $1200 for both of them. They use one to set the program with and another to do the final copy. Add about $700 to the blanks that you bought to have them ground and you have your final price. This is unless you run a Chevy or most Ford belly button combo's that are available through wide use.

In a cam you have how far a valve opens, how long it is open, and how fast that valve opens/closes. As far as ramp speeds/angles go it is always best to open that intake valve as fast as you can. The fastest ramp angle that you can possibly go with is 90*, but that is only easily attainable with 45mm cams. The SC cam is a 60mm cam; therefore, it is much harder to reach that high ramp speed until you get to around .850"-1.000" lift with high durations.

When the valve closes it is a whole different story. Most cam profiles match the ramp speed/angles to that of the intake and that's not a good thing to do without revisions. A way of helping get rid of reversion is to grind the closing side of the intake lobe to a less steep angle to slow the valve down while closing. This will control the speed, in which the air is reverted back through the intake whenever the intake valve is closed. The faster the valve closes the faster the air moves in the opposite direction, which means a stronger impact onto forced air coming in causing more turbulence.

Another reason for running a slow closing ramp speed is to control that unnoticeable valve bounce. If you think your valves don't float because it's unnoticeable then you've been deceived (maybe). A way to tell is get finger nail polish and paint a straight line down each side of the valve spring. If the valve spring isn't in the same spot as whenever you painted them, then that means the were spinning on their seats from valve bounce. You can control that mostly by loads of spring pressure and/or a soft valve seat such as Copper Beryllium or Ductile Iron.
 
Casey,

Not untill the past year or so have we been able to have our own cams ground or had we even any aftermarket new cam choices. So there were never any bellybutton cams so to speak except typical regrinds we would use which would never be exactly the same unless the same exact core was used.

Currently the problem really isnt getting custom grinds made. Knowing what to have ground is. There may however be a limit to what we can specify ramp angle/speed wise however if I remember correctly. I've only ordered about 4 to date and they were able to come up with what I wanted without complaint. From a durability/streetability standpoint. So we are not exactly starting out with a BLANK peice of billet steel but a basic core configuration.

Perhaps Dave or someone will better know how many of these basic core configs are available and what the ramp angles may be
 
A great cam profile is one of the easy parts to design & come up with
for the 3.8 SC engine..

We need a better air path to flow enough air first. IF compairing it to
which came first chicken or the egg . On this existing 3.8 motor
the manifold needs upgraded before the cam. Then the heads before
the cam. Then the numbers will be there to lay a cam.

And I think trying to reshape stock casting's is a cool thing to try and do.
expecialy when parts turn out to make more power and have nice looks.

Thanks Randy
 
Back
Top