Port velocity vs cfm

[travis]

After reading some excellent real world info in this post, I’ve got some questions related to port velocity vs cfm...

viewtopic.php?f=1&t=840&start=30


From what I got out of that post, 2 ported OEM heads with nearly identical flow numbers, but one being much faster than the other, the slower port velocity head made a bunch more power and was noticeably quicker at the strip. I think the takeaway from that was that pumping losses with the faster port velocity head hurt power much more than the raw flow numbers would lead you to believe. I think this confirms the old saying that flow isn’t everything.

So...I’m trying to look at this from a different perspective relative to something I’m working on now. I’m building a low $$$ 351w, trying to make the most out of the OEM 1.78/1.45 valves heads. I would like to see 350-ish HP out of it, peaking around 5000-5200 rpms, with as much power in the 2500-5000 rpm range as I can get. I don’t care what it does after that point. I’m not limited by any rules as it is just a street engine, but am budget limited.

My porting efforts so far have gotten me 176/134 I/e respectively, both peaking very shortly after .500” lift, but also both hitting 90% or better of peak flow by .400” lift. Right now my intake port velocity is pretty consistent at 260-271 fps everywhere except for 1 spot of 375 fps on the common wall side of the floor of the short turn apex.

Using the info found here http://www.diyporting.com/E7p3.html it looks like the best place to easily get a bunch more cfm is by lowering the port floor right before the short turn, and putting a better radius in there. My concern is that it isn’t going to take much more flow to get the velocity numbers up to the point of being too fast everywhere. As small as these ports are it doesn’t take much flow increase to really pick up the velocity.

Basically, at what point do I stop before I start shooting my self in the foot?? At this point I am at stock flow numbers below .200”, with increases from .200” and up, with slightly slower velocity’s than stock. I could probably lay the short turn back and pick up some flow there, but on my other iron heads doing that killed a bunch of low lift flow (under .300-.350” or so) for big gains above .400”. For relatively short street cam timing, this seems counter productive...looks to me like it would kill AUC on a .500”-ish lift street HFT cam.

Short version...would I be better off with lower flow and lower velocity, or higher flow with potentially a too fast port?

[jarmoyp]

"I think this confirms the old saying that flow isn’t everything."

"I could probably lay the short turn back and pick up some flow there, but on my other iron heads doing that killed a bunch of low lift flow (under .300-.350” or so) for big gains above .400”. "

This move kill the velocity which is good. You forget at the end of story that flow isn’t everything.

[travis]

"I think this confirms the old saying that flow isn’t everything."

"I could probably lay the short turn back and pick up some flow there, but on my other iron heads doing that killed a bunch of low lift flow (under .300-.350” or so) for big gains above .400”. "

This move kill the velocity which is good. You forget at the end of story that flow isn’t everything.

The problem with that big gain above .400” is the port went way fast...415 fps iirc, and went turbulent at .650” or so. Not that I would run that much cam but still that port was getting crazy fast

[steve316]

With your intake on you will have less air speed & flow. I have tried to lower air speed by making port larger; then the flow picked up and the air speed was right back were it was. the plus was the car went faster. Do the best you can, your combo may want more air speed than you think.

[Walter R. Malik]

If this is a race engine and not something driven at part throttle a lot of the time, anytime the air flow requires less duration to make the peak torque number, it will probably go slower if you are forced to use less camshaft than optimum for peak horsepower.

[rebelrouser]

If you can gain access to one, a smoke machine will pressurize your crankcase with smoke that will glow with a UV light, it will find any small leak spots in your engine. Lot of general automotive repair shops use them for evap system and leak diagnosis. I had some issues with my engine not building low idle vacuum, and while the vacuum pump worked well enough to stop any visible oil leaks, the smoke machine showed me leakage from the rear main seal and the corner of the oil pan, fixed those and I had good vacuum.

[PRH]

It seems obvious to me that if you have low flow numbers for a given port area, the velocity will also be low.

Imo, for 1hp/ci............ you’re way over thinking it.

Put some bigger valves in it, get all the easy flow you can....... cam it accordingly for the intended use/desired power level...... and iiwii.

My feeling is....... it would probably require a cam with more duration than might be “optimum” for the application to reach the hp goal.
Which means some compromise will likely have to be made.

I just don’t get much Ford stuff in my shop.
Aside from the 69/70 351w heads, what are some of the preferred later model castings that are supposedly the “better” ones to use for this type of application?
Are they readily available and easy to get?
How much better are they than what you’re using?
If they’re noticeably better, and fairly easy to procure...... and you’re going to do this type of job regularly....... it might be worth trying to stockpile a few sets of cores.

I’m spoiled from working on mostly Mopar stuff, where just about any 360 head can be made to flow in the 220-230 range with fairly minimal work.

[RevTheory]

What vehicle is this being built for? A lighter car with a looser converter and good gears vs. a heavy truck with 35s may help you prioritize your efforts.

[PackardV8]

Suggest, since it's your engine, you have the flow bench, do your work with the intake manifold and carb installed. We've all seen heads with hours in them and perfectly even flow go wonky when the intake and carb is installed.

We're finishing up an obsolete engine with many hours making the heads as good as we know how, but it's racing in a "factory-appearing" class which requires the OEM iron intake and exhaust manifolds. The dyno says most of the head work was unproductive, as the unreachable interior of the iron intake seems to be the choke point. (Yes, years ago we paid Extrudehone big bucks to do an iron intake. It was shiny inside, but the improvement was not commensurate with the dollars spent; no surprise, extrudehoning cuts aluminum much better than it cuts iron.)

[PRH]

I don’t flow the heads with the manifold installed very often, but I have done it for some F.A.S.T. R&D on a BB Mopar.

8 cylinder average of the head with a radius plate and the valve @.700 lift vs the head with an untouched 383 4bbl manifold(and a few others) and a 1000hp Holley carb on a 1/2” spacer.

Head= 286.7
Stock manifold= 212.8
Mild reworked manifold= 230.7
Heavily reworked manifold= 245.6
Port matched M1 single plane= 263.1
Ootb RPM= 250.6

The stock manifold on a stock head drops the flow about 35-40cfm.
Porting the head is worth a roughly 50cfm gain.
After bolting the stock manifold in place, that 50cfm gain is cut down to 18cfm.........194cfm on the stock head, 212cfm on the ported head.

In an application like what the OP is building, an aftermarket manifold is likely money well spent.

[GARY C]

Here are a few other old threads on this subject
viewtopic.php?f=15&t=43856
viewtopic.php?f=1&t=480
viewtopic.php?f=1&t=2109

[PackardV8]

The stock manifold on a stock head drops the flow about 35-40cfm.

Thanks for the very detailed real world numbers.

Now, here's the question I've never answered. If the intake manifold is the choke point, are there real gains to be made by porting the heads to ever increasing CFM? Would the net gain be greater if most of the porting effort was expended to make the intake manifold flow at least as much as the head intake port before beginning to port there?

Porting the head is worth a roughly 50cfm gain.

Again, assuming the reworked OEM iron intake flows 230 CFM, how much of that 50 CFM taking the intake port up to 287 CFM would actually be realized in net flow gain?

[PRH]

I’m pretty sure if you flowed the manifold by itself it would flow more than 230.

For example, on that head the M1 single plane flowed 263......... but on a better head I’ve seen it flow over 280.
Yet, if you bolted it to a stock 230cfm head, it would still flow less with that manifold attached.

I haven’t done enough of that type of testing to be able to come up with any kind of correlation between the flow of the head itself vs what it flows with the manifold attached in terms of power output.

My testing with the F.A.S.T. stuff was primarily to see if the BBM stock manifold was as bad as it seemed.

[quickd100]

I've got a mild 605 Hemi in my old truck, it's a street machine that on occasion makes a pass or two at the strip. I first used a set of Indy SR-426 heads on the motor. Indy cnc'd them, MCH gave them a bowl blend. 2.25/1.94 valves, 212cc intake port, they flowed 387cfm@.700. The motor made 853hp&818 ftlbs. It made 765 ftlbs at 3500rpm at the start of the Dyno pull. A couple years later I got the bug to make more hp. I bought a set of Indy Hi-Po heads, 2.4" intake valve, 266cc intake port. Out of the box they flowed 443cfm@.700 corrected. The heads being the only change it made another 100 hp but lost about 100 ftlbs of torque and I had to spin the motor up another 1000 rpm. Basically lost port velocity, and made it a pooch. If it were an all-out race motor I wouldn't care. I'm going to put the small heads back on.

[GARY C]

After listening to all the ST interviews and reading everything on this subject I have come to the conclusion that you need a square round oval rectangle port with a mcs somewhere between the plenum and valve seat, a fast slow air speed with very abrupt long smooth gentle radius's with good cfm as long as it not at low, mid or high lift and an intake that is port matched exactly as long as it is bigger or smaller then the head... Oh and a smooth rough texture or was that rough smooth?

I feel like I have spent a month listening to politicians tell me how to do a budget.