Fixing bad port work...

[Rick360]

I ported a set of 034 bowties for a friend about 7yrs ago and recently got them back to try to apply what I have learned here over the past few months. I bought pitot tubes and hooked them to a digital readout and made a few mods to my bench to read higher pressures/depressions.

The engine is 355 13:1, changing to roller cam 268/276 108lsa, Victor Jr, 750 carb., 1.875 heades, 5500 conv, 3200lb car, 5.13 gear, 9" tires, ran 6.89 @100 in 1/8 in ~ 1000ft DA weather with these heads and solid cam 264/270 .600".
The heads have 2.055/1.6 valves. Ports were ~198cc's. Avg CSA was ~2.29sq.in.

Problem #1:
The int. valve throat is 1.86" (90.5%)or 2.62sq.in., I know its too big. The pushrod area was 2.20 sq.in. (too small) and the area before the short turn was about 2.25sq.in.

Problem #2:
The intakes flowed 270cfm(28") @ .600" and gained slightly on up. When I flow the port @ 40" the flow goes down above .500" lift.

Problem #3:
The intake seats were radiused/blended not sharp on all angles.

To fix #3 The seats were redone but couldn't get edges under 60 in all places, but 45-30 in chamber and 60-45 in bowl are now sharp and some have 75-60 sharp. Are these the most critical edges? Will this help? Back cuts will be put on the intake valves since the bench showed with the new seats that back cuts were better. (they weren't before)

To attempt to fix #1 I opened the pushrod up as much as I can (~2.45sq.in now) and enlarged the area above the turn (~2.45 sq.in.). I can't get them as big as the valve throat without welding etc. (beyond the scope of this project). I also layed short turn back and bigger radius to fix #2. The port now flows 270+cfm when flowed @40" and converted back to 28" and doesn't "back up" at high lifts. I tested velocity (28" depression) and got a high of 322fps and low 264fps (avg 294, std dev 18) checking 9 places at the pushrod. The short turn is a high of 400fps and low of 242 (avg 309, std dev 55) checking 9 places also.

It appears that the short turn will have a choke problem according to Larry Meaux's previous posts on port velocity. The pushrod looks OK? Is there anything else I can do to correct this or at least minimize the effect of it?

Putting a bigger valve in would seem to be adding to problem #3 but would help problem #2. If a bigger valve went in the seat angles would be better and the throat % would be less. The disch. coef. would get worse since the flow can't pick up much with this CSA. Is staying with the 2.055" valve the best plan?

Lessons learned for next time, but what about these? Should we scrap them?

Sorry for the long post but I wanted anybody willing to help to have as much info as possible. Thanks for any help or advise. I haven't even started on the exhaust yet.

Thanks,
Rick

[dbusch]

After reading Larrys thoughts on cross sections throughout the runner I now notice there are lots of heads that are screwed up like you describe. Huge cross section at the throat compared to very small over the SS and at the pushrod (an unported SBF twisted wedge head comes to mind). Having (at least) the same cross section at the pushrod as the throat and slightly more at the SS seems to produce a better flowing port, especially when coupled with a decent SS design...

[maxracesoftware]

To attempt to fix #1 I opened the pushrod up as much as I can (~2.45sq.in now) and enlarged the area above the turn (~2.45 sq.in.). I can't get them as big as the valve throat without welding etc. (beyond the scope of this project). I also layed short turn back and bigger radius to fix #2. The port now flows 270+cfm when flowed @40" and converted back to 28" and doesn't "back up" at high lifts. I tested velocity (28" depression) and got a high of 322fps and low 264fps (avg 294, std dev 1 checking 9 places at the pushrod. The short turn is a high of 400fps and low of 242 (avg 309, std dev 55) checking 9 places also.

It appears that the short turn will have a choke problem according to Larry Meaux's previous posts on port velocity. The pushrod looks OK? Is there anything else I can do to correct this or at least minimize the effect of it?

Putting a bigger valve in would seem to be adding to problem #3 but would help problem #2. If a bigger valve went in the seat angles would be better and the throat % would be less. The disch. coef. would get worse since the flow can't pick up much with this CSA. Is staying with the 2.055" valve the best plan?

At what Valve Lift did you measure Velocities ?

if you have a .700" Lift cam
you would need to measure those Velocities minus -.100" Less
like at .600" Lift if you had a .700" Lift Cam

that would account for things like Lash Loss and valvetrain deflection loss,
and would be closer to measuring velocities at the point in the Lift curve where the Piston is moving very fast and the Heads are flowing with their highest velocities that the engine will effectively see.

in other words, an OEM Factory #034 Bowtie head's "as cast" pushrod area would not have too much velocity at the pushrod area if a Cam like an OEM .390" int / .410" exh stock hyd Cam was used.
so if you use a .400 to .550" lift cam , that pushrod area is probably good enough just like its cast.

if you're using a small Hyd or Solid Cam, you probably never need or want to open up the pushrod area...you will need the velocity there
at lower RPM ranges and with smaller Cams.

if you had a .550" Lift cam, you would go back and check all those velocities at .450" lift (-.100") ....and you might find out the speed thru the pushrod area is acceptable just like its cast.....as you get the Heads to Flow more and more, and use a Cam with more Lift you have to go back and checkout your port velocities to make sure they are not too fast.

[dbusch]

Larry is there a way to mathematically calculate what the approximate velocity would be at the throat at certain lifts and then make a determination as to what the SS and pushrod area needs to be for that lift? It would be nice to be able to do this without a flowbench or velocity probe, but maybe its not possible....

[Rick360]

At what Valve Lift did you measure Velocities ?

Velocities were checked at .600" and 28" water depression.

the new roller cam has .670" gross lift with 1.6 rockers.

This is a race motor w/5500 conv., not a street motor. The pushrod area was already opened up some to get the 2.2sq.in CSA. At 28" the flow is 275cfm @ .600"lift.
The engine would fall off if shift was raised above 6600. Seemed like peak power should have been higher than that unless some induction problem existed beyond that rpm.

Rick

[dbusch]

seems to me the critical velocity at the pushrod would occur at a certain rpm regardless of what lift the valve is at. In other words, what if you had good (not too high) velocity at .500 lift but had enough cam (a .600 lift roller with 270 duration @.050) to go to an rpm that that cross section would reach critical velocity at. So, do you size that cross section for good velocity at .500 lift or for the ability to reach a certain rpm???

[Guest]

Velocities were checked at .600" and 28" water depression.

the new roller cam has .670" gross lift with 1.6 rockers.

This is a race motor w/5500 conv., not a street motor. The pushrod area was already opened up some to get the 2.2sq.in CSA. At 28" the flow is 275cfm @ .600"lift.
The engine would fall off if shift was raised above 6600. Seemed like peak power should have been higher than that unless some induction problem existed beyond that rpm. ----Rick

Here's a few Equations to help

FPS = ( CFM / CA ) * 2.4

CFM = FPS * CA * .41666667

CA = ( CFM / FPS ) * 2.4


RPM = ( FPS * CA ) / ( Bore * Bore * Stroke * .00353 )

FPS = ( Bore * Bore * Stroke * RPM * .00353 ) / CA

CA = ( Bore * Bore * Stroke * RPM * .00353 ) / FPS

where;
RPM = point of desired Peak HP
FPS = Feet per Second
CA = Cross-Sectional Area in Square Inches (smallest measured)
============================================

use
RPM = ( FPS * CA ) / ( Bore * Bore * Stroke * .00353 )

and your Engine
6771 rpm = ( 614 * 2.2 ) / ( 4.030 * 4.030 * 3.480 * .00353 )

pretty close to the 6600 RPM "Choke ?" you are experiencing


then solve the other way
CA = ( Bore * Bore * Stroke * RPM * .00353 ) / FPS

2.145 = ( 4.030 * 4.030 * 3.480 * 6600 * .00353 ) / 614

the 2.145 rounds off to 2.2 ..pretty close to your 2.2

This is a race motor w/5500 conv., not a street motor

yes i know..i was just showing an example of what you would need to do at the pushrod area if you went towards a stock OEM cam or smaller lift cam than you run. Just an example to show that a Porter should not just hog-out the pushrod area if the speed there is below Choke velocity with the Cam lift that will be used, and the amount of Flow CFM the Heads flow

To attempt to fix #1 I opened the pushrod up as much as I can (~2.45sq.in now) and enlarged the area above the turn (~2.45 sq.in.). I can't get them as big as the valve throat without welding etc. (beyond the scope of this project). I also layed short turn back and bigger radius to fix #2. The port now flows 270+cfm when flowed @40" and converted back to 28" and doesn't "back up" at high lifts. I tested velocity (28" depression) and got a high of 322fps and low 264fps (avg 294, std dev 1 checking 9 places at the pushrod. The short turn is a high of 400fps and low of 242 (avg 309, std dev 55) checking 9 places also.

FPS = ( CFM / CA ) * 2.4

264.5 fps = ( 270 / 2.450 ) * 2.4


and 2 more equations

Velocity_FPS = ( Pitot_Press ^ .5 ) * 66.2

Pitot_Press = Velocity_FPS * Velocity_FPS * .000228183

the 264.5 fps converted back = 15.96 " of Pitot_Press
15.96" Pitot_Press = 264.5 * 264.5 * .000228183

on your Short Turn = 400 fps
36.5" Pitot_Press = 400 * 400 * .000228183

i guess you measured this 400 fps by flippping the Pitot Probe upside down with the probe resting on the Short Turn's Apex ?
and the 242 fps was with the Probe flipped the other way measuring Roof velocity

and the 322 fps was measured with the Probe resting against the pushrod's curved wall's apex on the exhaust wall side
23.7" Pitot_Press = 322 * 322 * .000228183


i wouldn't worry too much about your peak Short Turn velocity of
400 fps (36.5" )...just don't build up that Short Turn area higher than it is with epoxy even though it might show you a FlowBench CFM gain..it will be false Flow, the engine will loose HP/TQ .

the pushrod choke point is "before and further away" from the Short Turn
the pushrod choke point will dominate before the Short Turn separation problem.

Pontiac engines usually have same Choke problems you are seeing
relatively large CID size and valve sizes when compared to available pushrod area sq.inch

GM solved most of the Pushrod area problem with the 15 deg, 18 deg, SB2.2, Canted-Valve heads

[maxracesoftware]

Forgot to LogIn previously

Rick, you might be interested in these FlowBench numbers

from my SF-600 FlowBench which reads on the conservative side

this was the 1st design GM cast-iron #034 Bowtie head
"without" the waterjacket kickout around the exhaust side of the head
and also with the thinner thickness chamber walls

#034 OEM out of box shape with GM out of box 2.020/1.600 valves
Flow tested on 4.000" Bore Fixture at 28"
had a radius entry on Intake ports
and a short exh pipe [with a radius lip] attached to exhaust ports

Lift---Int-----Exh
.200-118.8--107.9
.300-170.8--139.8
.400-206.5--146.2
.450-218.2--------
.500-230.0--146.6
.550-223.7--------
.600-217.4--148.8
.650-218.4--------
.700-219.4--148.8
.750-219.4--------


this is another 1st design #034 cast-iron Bowtie head
Ported all-out max effort but no pushrod welding 223 CC Port
just maxed-out at the pushrod area 52.5 cc chambers
Titanium 2.085/1.585 valves with 30 backcut on intakes only

4.000 inch Bore Fixture, No-Pipe bare exhaust port

Lift---Int------Exh
.200-134.4---106.7
.300-195.1---150.3
.400-240.9---200.6
.450-259.8---------
.500-275.0---216.0
.550-286.2---------
.600-293.3---222.0
.650-297.6---------
.700-302.8---224.1
.750-308.3--------
.800-311.0---224.5
.850-314.2---------
.900-317.0---225.2
.950-317.7---------

same head but on a 4.125 bore flow fixture
Lift---Int
.200-138.0
.300-199.1
.400-245.6
.450-263.1
.500-278.7
.550-289.7
.600-296.1
.650-302.0
.700-307.5
.750-311.4
.800-315.8
.850-319.3
.900-322.5
.950-324.0

the head was flow tested above .750" Lift just to gather extra data


for another comparison,
here's flow numbers on ProTopLine alum heads CNC ported by Ron's Porting with 2.100/1.600 valves on 4.125 Flow Fixture (tested 12-7-2004)
Lift---Int------Exh
.200-130.9---115.2
.300-184.5---152.0
.400-229.0---180.1
.450-246.0
.500-263.1---198.5
.550-278.7
.600-291.3---209.2
.650-298.8
.700-302.8---212.4
.750-304.3
.800-305.5---214.3

Putting a bigger valve in would seem to be adding to problem #3 but would help problem #2. If a bigger valve went in the seat angles would be better and the throat % would be less. The disch. coef. would get worse since the flow can't pick up much with this CSA. Is staying with the 2.055" valve the best plan?

Lessons learned for next time, but what about these? Should we scrap them?

about the largest valve sizes that will work in a #034
is something like 2.085/1.585 combo with stock valve stem centerline distances

and a Picture showing about the largest you can get the pushrod area without welding

http://www.maxracesoftware.com/Bowtie_i ... ir_640.jpg

a cast-iron #034 Bowtie can do pretty well against the ProTopLine alum heads with pretty close to same port volumes and valve sizes.
the only problems with cast-iron is they can crack and are very hard to properly repair if they are damaged.

[Rick360]

Larry, I am off all the way thru the lift curve compared to your numbers even after figuring for the smaller size valve. Those are some impressive numbers for that head. I wish I could have a look at the chamber and SSR on those. lol.

What is the difference between the first #034 head and the later ones they called bowtie II's? The ones I am working on are the later ones I think.

Any tricks to getting that mid lift flow you can divulge?

When I flow at 28" and flow @ 40" now and convert back to 28" the numbers match nicely until the valve is >=.600". From .600" on up the higher depression will show less (converted back to 28") than when flowing at 28". Does this indicate the SSR is still a little turbulent at higher velocities or will this always happen at the higher depressions/velocities?

With a 2.085" valve and 90% throat the pushrod and SSR CSA's can't be as big as the throat CSA. On Darin Morgans interview (I highly recommend this interview) he says this is bad. Does that only apply to a very high rpm engine. On a lower speed engine (7k) does it hurt HP to have the pushrod smaller than the throat?

Thanks for all the help.

Rick

[maxracesoftware]

Larry, I am off all the way thru the lift curve compared to your numbers even after figuring for the smaller size valve. Those are some impressive numbers for that head. I wish I could have a look at the chamber and SSR on those. lol.

What is the difference between the first #034 head and the later ones they called bowtie II's? The ones I am working on are the later ones I think.

i have a few Pics i can post later
also have a brand new #034 head band-sawed into different sections
i'll take a few closeup Pics of cut aways, and aluminum templates

the 1st time i saw the #034 cast-iron Bowtie was at Brodix in Mena, Ark

Mr. J.V. Brotherton was giving me a tour of Brodix
and there was a cut-away section of #034 head
i asked what a GM cast-iron head was doing at Brodix ???
He said not to tell anyone, it was top secret, and the Head would be released soon to the Public

i guess its been long enough time now to tell about it .
even have a Pic i can post of that head sitting at Brodix , i'll post it later.

here's a Link to Head casting at Brodix
http://www.maxracesoftware.com/casting_ ... r_head.htm


there are atleast 4 different versions of #034 castings
the 1st one i saw and ran, had thinner chamber wall thickness and wasn't kicked out on the exhaust side

2nd revision= thicker chamber walls + kicked-out waterjacket area on
exhaust side

3rd revision= basically same thing but now had a cast in indention at where the heat-riser location would be on the Intake flange

4th revision = the current head now..same as 2 & 3 , and i'll have to say more tomorrow about

When I flow at 28" and flow @ 40" now and convert back to 28" the numbers match nicely until the valve is >=.600". From .600" on up the higher depression will show less (converted back to 28") than when flowing at 28". Does this indicate the SSR is still a little turbulent at higher velocities or will this always happen at the higher depressions/velocities?

you've probably checked for leaks or leakage, especially when flowing at higher test pressures at lower lifts the exhaust valve can partially be sucked opened at high test pressures

whenever i flow the Intake Port i use a metal device to increase exh valve spring pressure, then just pop it off when i flow test the Exh port
as in this Link
http://www.maxracesoftware.com/Valve_Open_Fixture2.jpg


1 more thing, i always Flow Test with a HeadGasket and the Head is always bolted down to the Flow fixture.

on my SF-600 the Intake port will usually follow the standard pressure conversion equation all the way up to 48 inches
but i occasionally run into Heads that look great at 28" Inches, but fall off pretty much as test pressure is increased to 36 or 48 inches on the Intake side. Lately half the LS-6 or LS-1 heads people send me look OK at 28 inches , but when flow tested at 36" or higher fall off 5+ percent of flow sometimes at .500" lift and above...i usually redo the valve job, bowl, short turn and fix the problem, the result is they make more HP/TQ and continue to make good HP after peak HP, they don't fall off as quick and use sometimes a lot less fuel.

http://www.maxracesoftware.com/VinceFMSdyno.jpg

some of the LS-6 heads take a dive at .500" Lift and above, as early as 30 inches of test pressure , but look OK at 28" inches
if you flow only at 28" inches you'll never know ?
you should spot check Int ports as high in test pressure as you can

100 PerCent of the time in every Head i fixed this problem, the RaceCar has went faster down the DragStrip !!

on the exhaust side, i've seen where all 4 exh ports might Flow the very same Flow Curve and numbers at 28" inches ...but when Flow tested at 48" inches, 1 port might flow a bunch better, 2 might increase steadily, and 1 might be a Turkey, hardly no increase ....but thats not common, but should be spot checked when you Flow Heads.

so if you believe in the .55 Mach Theory ?
(as its been pretty accurate in your case)
614 fps = 86.042 inches of water test pressure
around 3.107 psi
enough velocity fps to give you 121.14 Ve % if things are right, less Ve% if not

130.0 Ve%= 4.4088 psi or 122.08 inches of water or 731.44 fps or .6554 Mach

if you've looked at some of the data in Philip H Smith's work with the Morrison Pressure Indicator, way back then he was seeing depressions as high 3 to 5 psi

i think we should be Flow Testing at least a minimum of 36"
and more like 60" inches or 105" inches would be better

28" inches = 350.3 fps approx half the speed of 700 fps or .627 Mach that a modern Engine can attain....so usually 28" inches is a good representation of what Flow -vs- HP the engine should make,
but i think the better method would be to Flow test at pressure differential the Engine sees. Besides flow testing at 28" on a SF-600 Bench is a pain sometimes, as you have to look downward behind the cyl head, especially if you bolted an Intake Manifold and Carb on the Head.
36" inches is more Eye Level, and where 28" was on the older SF-300 benches.

i'm planning to increase CFM capacity of my FlowBench to 1500 CFM and going all digital to flow at 60" inches or higher.

Any tricks to getting that mid lift flow you can divulge?

the #034 Head had HeadBolt .500" tubes pressed into between the Intake Port pair, and that divider wall was perfectly straight to the tube, sort of like a Brodix Supr Spec casting, roof was raised and maxed out as high as possible and retained .100" thickness all the way to the spring seat ..the Bowtie head likes this if done correctly., the width of the short turn was as wide as the valve od or at least 2.050 i'd have to look at my notes

[airflowdevelop]

Larry,
I am assuming by your email that their is no other reliable way to messure the differential other than digital? Or making your own inclined? I just finished tying 2 600's together, and adding 6 more motors...but have no accurate way to measure differential other than changing the angle of the manometer. I have tried some of dwyers digital handheld stuff, with nothing more than frustration.

We have seen data that indicates depressions exists eclipsing 120".


Dennis

[cboggs]

I have a question about flowing at higher depressions, ..

If you have say a SF-600, .. can it be as simple as making a sepperate flow box with say another 10 vacume motors to tie into the 600.
So, .. when you need the extra "suck" you just turn on the extra flow box??

Now about measureing the depression. A manometer is easy to make right, .. can't we just make a "U" manometer that goes up to the 60" or so we want to measure???

I do agree with Larry tho', .. if there's electronics to do it, .. and software to do the math, .. I'm all for it.

Heck I still can't grasp the math to figgure the curtain area for a certian valve & lift, .. then figgure the potential flow. I have to use my software.

Curtis

[Guest]

I have a question about flowing at higher depressions, ..

If you have say a SF-600, .. can it be as simple as making a sepperate flow box with say another 10 vacume motors to tie into the 600.
So, .. when you need the extra "suck" you just turn on the extra flow box??

Now about measureing the depression. A manometer is easy to make right, .. can't we just make a "U" manometer that goes up to the 60" or so we want to measure???

I do agree with Larry tho', .. if there's electronics to do it, .. and software to do the math, .. I'm all for it.

Heck I still can't grasp the math to figgure the curtain area for a certian valve & lift, .. then figgure the potential flow. I have to use my software.

Curtis

An additional "sidecar" chamber is probably the easiest way...

I have played with the dwyer electronics, and they are nothing short of terrible for keeping accuracy.

I have also played with audie thomas's deal....And I think that is great! Although allot of people look down on it. If the accuracy is as close as he claims, their should not be a problem.

The vertical manometer is the easy part. The differential is the trouble.

I have heard of people using 2 u-tubes, and doing some funky math to accurantly come up with % of flow (This is already calculated for in the audie deal)


Thanks
Dennis

[airflowdevelop]

Don!

You need to fix this dagblasted thing so it keeps you logged in!


Dennis

[maxracesoftware]

If you have say a SF-600, .. can it be as simple as making a sepperate flow box with say another 10 vacume motors to tie into the 600.
So, .. when you need the extra "suck" you just turn on the extra flow box??

Now about measureing the depression. A manometer is easy to make right, .. can't we just make a "U" manometer that goes up to the 60" or so we want to measure???

Larry,
I am assuming by your email that their is no other reliable way to messure the differential other than digital? Or making your own inclined? I just finished tying 2 600's together, and adding 6 more motors...but have no accurate way to measure differential other than changing the angle of the manometer. I have tried some of dwyers digital handheld stuff, with nothing more than frustration.

We have seen data that indicates depressions exists eclipsing 120".
Dennis

making another "Cabinet-Box" with the same Blue Formica and style side by side with the original SF-600 and just using a switch to power on the extra motors/box when its needed..was what i'm planning to do .

the extra cabinet space would add additional counter top to place a flat screen monitor and computer, printer

also planned to make cable controlled ducting valves to duct hot air to outside of Shop to get rid of the extra increased heat out of the FlowBench room
Need to make outside duct with a valve and screen to prevent small varmits or such from entering

what about using Mercury as manometer fluid..that should let you scale down 150" inches down to current SF manometer heights

at those test pressures, what about pressure transducers or some sort of strain gauge to convert pressure useable to computer program ?