Port behavior at different depressions - Do the flow patterns really change??No.

[Nick]

Thanks Larry,

That's great info.

Perhaps this should have it's own topic, but here seems like a good place. What is the best shape for the short turn? Should the turn be as gradual as possible? Should there be a U shape just before the seat or should it look more flat?

It seems as the short turn on the 289 heads I'm playing with is way to tight, compared to aftermarket heads I have seen.


Anyone want to share? I am finally getting to use this thing I built, now the real learning curve starts!!



Nick

[Tony]

Thanks Larry, priceless stuff.

[gofaster]

Nick,
You asked a very good question. This is where the porter earns his keep, and the fluff and buff man goes away.

Generally speaking, you want a smooth radius from the floor, around the shortside, and into the seat. Some heads do not have enough material in this area to allow you to make the shape you want. A spare or junk head, or a plaster mold of the head is a great thing to have when you are working out your design. That way, you can grind, weld, add fillers, measure and test, measure and test, and measure some more, raise the floor, raise the roof, teardrop the guide, widen the turn area to try to slow down the ss, go back and fill it in or grind it back out.....

You get what I mean? If you have other things to do in the shop, go do them, and leave the head sit for a day or two. If you're like most of the people I know who do this work, your mind will keep working on the head while you are doing other things. When you come back to it you may see something that escaped you before.

If you take an approach where you make one change at a time, then measure and flowtest, you'll get results that will tell you whether you are moving in the right direction or not. Once you establish your own "sacred porting ritual", a systematic approach to working from the seat to the entry, and from the seat into the combustion chamber, things should fall into place for you.

One more thing, always try to make your flow test procedures into another kind of ritual that insures that the head is presented to the bench in exactly the same way. i.e., lightly greased guides and stems, tight sparkplug (or dummy plug), head in proper relationship to the bore, no airleaks at the gasket or ex. valv....etc

This may not be the answer you wanted, but it is how you'll find it.

[Thomas Vaught]

Agree with gofaster's comments.

The "short turn" typically is the key area on most
non full race heads. You can take a production pontiac head that would flow for example 185 cfm
at 25" delta p and do some good short turn work and make that head a 240 head at the same test pressure. Trying to get that head to now flow 275
cfm at the same test pressure is a LOT of additional work.

You trade the cast iron head away and get an Edelbrock head that flows out of the box 275 cfm.
The same tricks and maybe a small abount of material removal and you get a 300 cfm head.
Getting to 340 cfm becomes the challenge. You pay for what you get basically.

You have a better casting and you get more "easy" airflow.

The new Tiger Pontiac head will flow over 400 cfm
at the same test pressures but is over 5K to buy.
A lot of normal guys don't necessarily nead a 400 cfm head but you can run a lot less camshaft on a
large engine with one.

I guess it depends on what you are trying to do.

The man who used to upgrade the 185 heads to the 240 heads made a LOT of money over the years. LOL!

Tom V.

[maxracesoftware]

At what point in the Valve Lift Flow Curve
does it give you the most problems ?

its usually around what Darin Morgan (Reher-Morrison)
+ Harold Bettes (SuperFlow) call the
"Convergence Point"

Convergence_Point = Valve_Diameter * .25

Convergence_Point = the Valve Lift when the Curtain Area
equals the valve area

Examples=>
1.940 OD SBC int valve *.25 = .485" Lift
2.055 OD SBC int valve *.25 = .514"
2.125= .531
2.300= .575

around this convergence lift and above,
as the AirFlow and its Speed (MPH) "transitions" from
the Valve Area to the Curtain Area, is where
separation may begin to occur .... especially as the
Intake Valve's infulencing effect upon the Pressure on
the Short Turn Curve,..is moving further down out of the
Bowl "AWAY" from the Short Turn....thats why a
BackCut on an Intake Valve will not always
show you 100.0% gain in Flow with certain
combinations of Short Turn Apex Height + Short Turn Curve Shape !

go study your Flow Test Sheets, look at whats happening
around and above .25 L/D Ratio, (Convergence Point),
and you'll see that before .25 L/D Ratio,
the "Rate" of Flow CFM increases is healthy,
but "after" .25 L/D Ratio...its starting to rapidly taper off

you'll also notice that when you do have a
situation that as you increase Valve Lift and Flow your Heads,
that Flow CFM may decrease...thats usually around
the .25 L/D Ratio.

Example=> typical SBC Head being Flow Tested around
.500" Lift or so may have Flow CFM Numbers actually decreasing as you go from .450" towards .500" or .550" Lift
as you get into the "transition flow " or .25 L/D Ratio,
or "Convergence Point" area

you'll just about NEVER observe any type of Cyl-Head
having a Flow Transition Problem "BELOW" .25 L/D

its always around .25 or higher

[Mousehouse1]

This getting good guys keep it coming.

Thanks Larry and others for this post. I know it will help me when I do some cylinder heads.

[Nick]

Udated 1:28pm The valve measures 1.67in not 1.78in


That makes perfect sense. These small block ford heads drop off farrelly rapidly at the convergence point. There is another thing I would like to ask along the same lines. Valve size and potential flow. The two heads I am comparing have 1.67 and 1.94 valve sizes. The 1.67 valve size lags way behind at mid lift numbers. I'm not sure what the calculation is for potential air flow for a known valve size. From the flow tests I have seen, it appears to me the low to mid lift numbers are mostly the same for heads with the same size valve. I do see how the more expensive heads exceed past the convergence point.

The 1.67 valve produces a nice looking curve, but the 1.94 has a lot more air flow in the mid lifts.

Here are the numbers. There heads are very similar. I ported both.

302 1.94in 289 1.67in

.050 28 27
.100 56 52
.150 88 78
.200 119 101
.250 148 121
.300 171 137
.350 180 151
.400 185 164
.450 189 174
.500 192 180
.550 192 184
.600 192 188

So is the mid lift difference because of the valve size?

My reason for working with this 289 heads is to learn. It was free and is very similar to the 302 head I am using now. Both are small chamber, and both have tiny intake runners about 140cc.


Any comments would be appreciated.


Nick

[maxracesoftware]

this "Convergence-Concept" might still be unclear to a few,
so to help clarify =>

Below .25 L/D Ratio
the Curtain Area is LESS than the Valve Area

at .25 L/D Ratio
the Curtain Area EQUALS = the Valve Area
(a Poppet Valve Characteristic)

above .25 L/D Ratio
the Curtain Area > is GREATER than the Valve Area

i usually use .37 times the Intake Valve Diameter
as a guide for the "minimum" Camshaft Intake Lift
i need to use ....at .37 L/D Ratio...its "just enough"
to get the Intake Valve out of the way,
to let the Port do its "thing" .

on NHRA SuperStock Heads that are of the older
designs, trying to use more than .39 or so L/D Ratio
doesn't usually gain HP at high RPMs...as these type
Cyl-Heads have Short Turn Apex Height + Short Turn Curve
shapes that can't handle very high Air Speeds
at too much more than .39 L/D Ratios ....so that seems to
always limit your "Trapped Ve %"
no matter how you Port or Dyno Tune these type Cyl-Heads/Engines..you can NEVER achieve actual
"Trapped Ve%'s" above 110.0 ..most Times only
104.0 to 109.0 Ve..that in turn is an indication it can't
handle the AirSpeed MPH...because past .39 L/D Ratio,
the Short Turn Apex + Short Turn Curve "needs" the
Valve closer in promimty to help "Turn" the mixture
because its too much of an abrupt Short Turn Curve,
sometimes it needs NO BackCuts also....as it makes the
Air Turn further around the valve , effectively fooling
the Cyl-Head into thinking it has more Short Turn Apex Height + Curve.

you are never going to get an old 23.0 degree Cyl-Head
to actually measure a "Trapped Ve%" of 120.0 or greater
..its just too hard to TURN that Curve (NASCAR example)
and maintain Air Speed + attachment

in a modern Head with a better Valve-Axis -to- Port Axis, plus better and Higher Short Floor/Apex-Height/greater Short Turn curve..it now can handle much higher Air Speeds and get you
120+ Ve actual Trapped Ve Numbers.
on these type Cyl-Heads i use at least .41 L/D Ratios ,
and mostly .42 L/D Ratios as baseline starting point
on Intake Side.....sometimes even higher to .43+
...this INCREASES "Effective Flow Dwell-Time" above
your .25 L/D Ratio drastically.

[DaveMcLain]

Another place where the cylinder head can get strange at least from what I've seen is at or around .300 lift. I don't see the flow drop off of nearly being a linear increase in flow for an increase in lift but what I see is a drop in the stability of the port, what causes this effect? Because I don't think it's the same phenomenon..

Also, when you say "backcut" do you mean a valve with the stem diameter cut down or one with a 30 degree angle cut before the seat angle?

Great stuff too by the way!

[larrycavan]

Larry M.

What's your feeling on spreading lobe centers when port velocities are high?

I've seen 4V Suzuki ports that have been epoxied to a D shape. The claim is two fold:

1 - get the short side closer to the length of the long side
2 - the higher velocity of the filled port increases Ve and is best taken advantave of by pushing lobe centers up to 113 - 115.

Larry C

[maxracesoftware]

[larrycavan]

Thank you Larry!