Cam selection vs Velocity and Flow.

-
Cuda416

Cuda416

Well-Known Member
Joined
Jun 12, 2014
Messages
3,068
Reaction score
2,996
Location
South, TX
My limited understanding of cam selection is you want to match the cam to the heads. I know there is more to it, but was wondering, since velocity is so important according to many people, whom I agree with , how that ties in if at all.

In other words, is velocity taken into account at all when choosing a cam, or is port velocity just something to maximize regardless of the flow and cam selection?

I hope I'm asking the question correctly.
 
I do not find that important at all but rather it is the cylinder head that should be maximized for the application or job at hand. That in itself will enhance the performance of the cam and engine.

I find matching a cam to a cylinder head stupid if not a jerky big Corp gimmick.

While it is possible to put to large of a head on an engine, it would have to be pretty far out of wack (stupid or crazy matching of parts) for it to ether not perform well or run out of steam early.

When I’m planning an engine, the camshaft dictates the rpm range or window the engine will operate in. The cylinder heads flow will match the target of the performance goal being sought.

Between the two, cam and cylinder head, I try and get the cam to lift as much as the cylinder heads port will flow. AKA, lift at the valve is the cam until the cylinder head stalls. This is for 99% of what I’m doing. Not race engines. But more of a high performance or a very hot street engine.

The cylinder head MUST be flowing good air for its port size & that must be good for the engine size & that must be good for the task at hand. Without pushing the issue to hard, it can simplify a cam lifting approximately.450/.500 to needing a great valve job. At or Over .500, a bowl porting, 550/.575 ish areas will want more port work. Narrowing the pushrod pinch and opening up the port window. Above that, whip out all the tricks.

Does this mean you can use a higher level of my descriptive porting on a small lifting cam? No, but I wouldn’t pay for the work to be done.

The cylinder heads port window should also fit the size of the engine and for the target goal. Putting a fully ported W9 or Indy on a NA 273 and expecting 25 mpg and run low 10’s isn’t just going to happen.
 
Thanks @rumblefish360 !

The only cam choosing info I've had or researched is that of an existing engine so I think that's where the "head first" method, at least from what I was looking into, came from. I only realized that after reading your response, which makes loads of sense to me, that the first choice is the cam since it sets the stage. It easily stands to reason if you're building an engine from the ground up, start with the part that controls the breathing, then build to support that part.

Everything basically makes sense, but this statement

rumblefish360 said:
lift at the valve is the cam until the cylinder head stalls.
Click to expand...

Can you elaborate on that?

Thanks again
 
Ooooo, that was missed and poorly worded I think but I can explain. Grab your coffee, have a seat.

When I’m building an engine that I would label a more serious build for a stout street or racing engine, I want to lift the valve up as high as possible. I would rather do it by the cam lobe lift rather than add a 1.6 rocker. But ether way is fine if your good with it.

Why?

On the cylinder head. When you have the head tested for the amount of air it flows, your looking at a number of things. Choosing a cylinder head based on flow alone is a bad idea. You must look at the entire head!

But! For the sake of the K.I.S.S. method here today, we are just going to look at the flow rates of the cylinder heads ports.

As you lift the valve up, .050, .100, .150, .200 etc….
The amount of air will be recorded by the head tester. If the tester only does the CFM amount of air, your missing information. But to continue on the KISS method…

At some point the valve will open up to a certain lift and the amount of air will ether decrease (stall) or become turbulent. (Not that good of a quality air and fuel mixture going past the valve.) If the flow rate remains the same, keep testing, keep lifting the valve.

Now! I’ll take a cam that will lift the valve as high as possible to take advantage of the heads ability to flow more air as the valve opens up more. Sometimes this is possible, sometimes it is not. But the idea is to get in as much air and fuel as possible.

Some have argued that lifting the valve higher than the stall point is a good thing. Since it will only see max lift once and see all other lift points twice. There is proof this will produce more power or quicker times. There maybe a time where this is OK to do. I have not found the need to push this issue. Even more so on a street beast engine.

Why?

Lifting the valve up high on something other than a heavy hitter or race engine is only introducing more stress. The valve springs take the load (abuse) and will wear out quicker. The higher the performance of the engine and the higher lift of the cam Leeds to more maintenance and adjustments and/or down time for spring changes. The cam is the next in line and can be worn out with heavy spring loads and possibly not enough oil when your put putting around the street light to light around the parking lots.

How much power is in taking advantage of this way of thinking? This is a highly variable answer that is build dependent. There is some power there that will be found on the top end of the RPM scale. But IMO it is only useful on the described engine settings above.

Id like to point out that Pittsburghracer runs very quick without a high lift cam. Pay attention to the details and tricks to apply to an engine build. No one trick or method will produce Stella results by themselves so keep expanding and applying modern techniques.

Most of all, build Sensibly!!!!
Now sit back and ponder.
Then ask yourself the below question.

How much street time will the engine see?
 
rumblefish360 said:
Ooooo, that was missed and poorly worded I think but I can explain. Grab your coffee, have a seat.

When I’m building an engine that I would label a more serious build for a stout street or racing engine, I want to lift the valve up as high as possible. I would rather do it by the cam lobe lift rather than add a 1.6 rocker. But ether way is fine if your good with it.

Why?

On the cylinder head. When you have the head tested for the amount of air it flows, your looking at a number of things. Choosing a cylinder head based on flow alone is a bad idea. You must look at the entire head!

But! For the sake of the K.I.S.S. method here today, we are just going to look at the flow rates of the cylinder heads ports.

As you lift the valve up, .050, .100, .150, .200 etc….
The amount of air will be recorded by the head tester. If the tester only does the CFM amount of air, your missing information. But to continue on the KISS method…

At some point the valve will open up to a certain lift and the amount of air will ether decrease (stall) or become turbulent. (Not that good of a quality air and fuel mixture going past the valve.) If the flow rate remains the same, keep testing, keep lifting the valve.

Now! I’ll take a cam that will lift the valve as high as possible to take advantage of the heads ability to flow more air as the valve opens up more. Sometimes this is possible, sometimes it is not. But the idea is to get in as much air and fuel as possible.

Some have argued that lifting the valve higher than the stall point is a good thing. Since it will only see max lift once and see all other lift points twice. There is proof this will produce more power or quicker times. There maybe a time where this is OK to do. I have not found the need to push this issue. Even more so on a street beast engine.

Why?

Lifting the valve up high on something other than a heavy hitter or race engine is only introducing more stress. The valve springs take the load (abuse) and will wear out quicker. The higher the performance of the engine and the higher lift of the cam Leeds to more maintenance and adjustments and/or down time for spring changes. The cam is the next in line and can be worn out with heavy spring loads and possibly not enough oil when your put putting around the street light to light around the parking lots.

How much power is in taking advantage of this way of thinking? This is a highly variable answer that is build dependent. There is some power there that will be found on the top end of the RPM scale. But IMO it is only useful on the described engine settings above.

Id like to point out that Pittsburghracer runs very quick without a high lift cam. Pay attention to the details and tricks to apply to an engine build. No one trick or method will produce Stella results by themselves so keep expanding and applying modern techniques.

Most of all, build Sensibly!!!!
Now sit back and ponder.
Then ask yourself the below question.

How much street time will the engine see?
Click to expand...


Man I love this place..!!!... I'll re-read this and your other message a few times. overall it makes total sense.

The engines we'll build around here will be on the "strong street engine" end of the spectrum, and nothing too crazy.
 
Below is the flow report for asked for. The numbers after the flow number reported at the various lifts is the amount of swirl. While I’m not very knowledgeable on this, I asked and was given the reply that at .600 lift the numbers started to climb but in the head porters testing, he still considered it good and flowing quietly. At .700, the port made a racket and flow took a dip.

His advice was to keep the lift in the .600 area and not much more. While he didn’t write the numbers down to the exact point or say the exact point, he did say it’ll take another .030-.040 no problem. I was the higher 600’s where it started to get noisy. I know that’s gauge but it wasn’t really super important to me and I didn’t ask for lift tests for every .050.
I would not have a problem & would feel just fine running a .600+ lift cam with these flow reports. As you can see the exhaust never quit all the way to .700.

These are my W5’s I plan to end up running a MP tunnel ram on with an Ultradyne NF lobe solid cam with .640 lift, then minus lash of course.

Some will say there’s more power in lifting the valve more. I’m OK starting here. I’ll also be starting with a 4bbl. (1 step at a time folks.)

522C6FF9-B620-47D7-9187-EB13B42CF2BF.jpeg
 
Keep in mind the adage, there’s more than one way to skin a cat. You can really well with a good bit of gear and stall. If it is a street car designed with the thought of a lot of driving, (street gears and converter) I suggest a cam with easier ramps and modest lift.
 
Cuda416 said:
My limited understanding of cam selection is you want to match the cam to the heads. I know there is more to it, but was wondering, since velocity is so important according to many people, whom I agree with , how that ties in if at all.

In other words, is velocity taken into account at all when choosing a cam, or is port velocity just something to maximize regardless of the flow and cam selection?

I hope I'm asking the question correctly.
Click to expand...

My Guess is when it comes to race engines, they are more single purpose something like nascar everything is taken to it max ability. For a hot street street strip or mild performer there's a lot of ways to accomplish your goals, look at modern hemi or LS engine they use heads that flow enough to make 500-600 hp out of the box but use smaller cams so they can make a streetable 350-450 hp or you could use lesser performing heads and a more aggressive cam to squeeze every last hp out of but end up with a less streetable car. Take a magnum 5.9l with stock heads and 10:1 cr you probably can get 425 hp or so with a fairly big cam 285-295 but put a set of trick flows and probably could do it with 20-25 less duration.

As for port velocity you have little choice we only have so many heads available to us Like the smaller cam trick flow magnum 5.9l eg.. above since your not using a lot of the ports ability you probably could add some resin in the right spots to bring up velocity but that's not gonna be for most of us. Plus a lot of people confuse more velocity with small restrictive ports. But if you look at builds with high lbs-ft to cid ratio like 1.3-1.45:1 lbs-ft:cid there gonna be big roller cams ports carbs etc.. but yes be higher in the powerband in the lowest rpm 2000 rpm and under maybe high as 2500 rpm stock heads cam etc.. might have a slight advantage but is gonna make less lbs-ft:cid above that. I feel people overly worry about velocity yes it's important and if your doing a custom $$$ build then it should be part of the plan but if off the shelf heads like edelbrock or trick flow I'd rather the trick flows even with a small ish cam.
 
Last edited:
As long as the port is efficient, the velocity will be good.
As long as the ports cross sectional area isn’t to large for the displacement on hand, it will be good.
If the CS is low, the engine runs out of steam early.
If the flow capability of the head is low, it runs out of steam even earlier.
As YR would say, it’s almost impossible to out to large or a head on a stroker. Mostly because there is t a head large enough to be to big for a SBM.

Im using TF heads on a 11.3-1, 360 CID engine with 1.6 rockers and a small Hyd cam lifting the valve .573 on a 224@050 cam.
 
And again I'm shown why this place rocks. I'm constantly amazed at the information, know how, and experience people here have and the willingness to share and educate.

A hardy Thank You...

Now, a rather simple observation on port profiles. Before I heard of people talking about velocity, it always seemed odd that the shape was small at the valve, then opened up so much as you look at a side view.... However, as I think about this, it resembles a venturi...

For instance taking into account the area of the intake valve (3.2 sq/in), vs the area of the intake port (2.63 sq/in), then the total size of the open area the intake is open to 0.7" (4.44 sq/in for a 0.7" tall cyl curved surface are, not including the ends), it seems like squeezing the volume from the cyl through the valve and bowl do indeed form a venturi. If the bowl opening is 86% of the valve size, which would be 2.75 sq/in, that puts the cross section from the bowl "pinch" to the gasket surface, almost the same.

So, is one of the goals when porting to try and make the cross section as close to constant along the length as possible to keep the velocity up?

My numbers may not be spot on, but this is just a thought I had this morning after reading some of the responses from @rumblefish360 and his comments on port cross section.

Finally, is there a formula that helps decide port C/S as it relates to CID?
 
There is a formula. Look at the Wallace site.
There is ideal, for a certain size engine but it will also correlate to a certain rpm.

When porting a head, it is desirable to eliminate the pushrod pinch as much as possible if not entirely. (This is not possible on stock heads) this keeps the air and fuel traveling down a uninterrupted road instead of a wide/narrow/wide/narrow road. Air and fuel have a better mixture.
 
rumblefish360 said:
Below is the flow report for asked for. The numbers after the flow number reported at the various lifts is the amount of swirl. While I’m not very knowledgeable on this, I asked and was given the reply that at .600 lift the numbers started to climb but in the head porters testing, he still considered it good and flowing quietly. At .700, the port made a racket and flow took a dip.

His advice was to keep the lift in the .600 area and not much more. While he didn’t write the numbers down to the exact point or say the exact point, he did say it’ll take another .030-.040 no problem. I was the higher 600’s where it started to get noisy. I know that’s gauge but it wasn’t really super important to me and I didn’t ask for lift tests for every .050.
I would not have a problem & would feel just fine running a .600+ lift cam with these flow reports. As you can see the exhaust never quit all the way to .700.

These are my W5’s I plan to end up running a MP tunnel ram on with an Ultradyne NF lobe solid cam with .640 lift, then minus lash of course.

Some will say there’s more power in lifting the valve more. I’m OK starting here. I’ll also be starting with a 4bbl. (1 step at a time folks.)

View attachment 1715762163
Click to expand...
OK Rumble I am with you on the lift/stall scenario.
I would like to see your take on selecting the correct duration and lobe separation, please.
 
512Stroker said:
OK Rumble I am with you on the lift/stall scenario.
I would like to see your take on selecting the correct duration and lobe separation, please.
Click to expand...

No, because this becomes a build dependent thing for which is designed for the engine/car duties it is to be expected to perform in. There is also another problem which comes in taste of the cams characteristics when at idle and driving. Furthermore, you can even taylor the lobes rate of ramp lift and back down to the base circle, the ramp jerk or double jerk, lobe separation and overlap, etc…

So the correct cam duration and lobe separation is an impossible thing to be “Correct” on.

What I can suggest is a duration that the cam will perform in on and a LSA that would be (or should be IMO) good for the targeted build on hand. However, this suggestion is often met with others chiming in with there suggestions which quite frankly be excellent as well, but what the general public see are are huge differences in just a few degrees of duration without consideration of the particular lobe I would spec and why.

This is why I only make general recommendations that will be in the ball park as close as possible to what I myself would do. And I don’t always do that.

You also would be hard pressed to feel the difference in such small changes in the wide verity of suggest cams that are within a few degrees.
 
Cuda416 said:
Quick followup, I was able to find the math for port c/s determination at

Porting Pointers - Hot Rod Engine Tech

Ac/s=CYLvol x RPM)/88,200

Looks like a 408 want's around 3.78 sq/in of intake ?
Click to expand...
OK! Let’s say this is dead on for your needs. What is the ports dimensions? A stock port is in the 1 x 2 area-ish or so? In an aftermarket head, what are those sizes? And from what head? Then, after you look through all the heads available, what are there port size in cc and there flow rates? If that doesn’t become muddy enough for you…. The further down the rabbit hole of “Optimum” parts you go down the he more frustrating it can become. But it is something you can solve. It can depend on the amount of money you’d willing to spend as well. It “Adjust/Modify” parts. LOL!

Here is the KISS method;

Select the style of lifter for the cam and the cams duration you want your displacement to perform in and equip the the short block with the best flowing cylinder heads & intake your pocket allows and top it with a great carb. (Tunability with enough cfm.)

Remember!
There is no need to go crazy on lift and valve springs on a street engine. You can still have a great performer with a modest lift cam and decent heads for a nice street machine.

If you want a heavy hitting street machine, you know the trade offs. I’m willing to live with them on some of my cars. Not so much with other ones.
 
The numbers for my calcs, were based on the TF 190 heads just to have something to use as a model.

I like the KISS theory, but I do like to understand why it works which is why I'm interested in getting down into the weeds. It's easier to make compromises when I know what I'm giving up or gaining.
 
Cuda416 said:
Quick followup, I was able to find the math for port c/s determination at

Porting Pointers - Hot Rod Engine Tech

Ac/s=CYLvol x RPM)/88,200

Looks like a 408 want's around 3.78 sq/in of intake ?
Click to expand...

I went down the same rabbit hole when I was designing a project that never happened 400-425 hp 273 for Time Attack 65 Cuda, Nothing wrong trying to figure all this **** out but for the most part these are all ballpark formulas and make builds way more complicated and expensive if you let it. Most of the formulas and info out there are for full race engines so unless that's what your doing a highly competitive class or something they might be overkill or even wrong for a street car especially a daily driver. To tell how efficient your builds is by peak torque engines fall into a fairly narrow lbs-ft-cid which is mainly effected by VE% = Volume Efficiency Percent, how well you fill the cylinders plus compression ratio chamber shape etc.. most low powered engines are 0.9-1.1:1 lbs-ft:cid engine most average guy can do 1.1-1.25:1 good builders 1.25-1.35:1 and top of the heap builders 1.35-1.45:1. So if you can do around 1.2:1 your doing good for a 365 that would be 438 lbs-ft.
 
I think the port size is a little small but is well made up with an effective port. When considering this head for a street machine on a stroker, I believe that you will find excellent throttle response with this head.

There maybe some limitations to the head with a street machine and light/medium track cars. The power difference, I think the drivers will not really see or notice much. The switch from already having that head vs a W2/5 or 9, Brodix or Indy head would probably not really be what I say is, “The juice is not worth the squeeze” for most people.

Racers, would probably not use the TF in a OOTB condition anyway. I’m not saying (at all) that the head isn’t capable of running hard and fast or producing a low ET and fast trap speeds. I have not seen or heard of anyone porting the TF head yet though I’m sure SOMEONE has.

I think this would all go back to my basic KISS method for at least the base angle of attack on building an engine for a strip only car. Depending on the level of speed/power your looking at, the head choice starts to become clearer.

If I had a 408 for track only in a faster bracket or heads up deal, the TF would not be my personal head choice. I’d look for the bigger port window head. Even if there both flowing the same cfm or slightly less.

Personally, I don’t think you could wrong with the TF’s up top a street bound 408. You could do better but at what cost and is it worth it to you? This is the common question in every ones head on builds.
 
Last edited:
Cuda416 said:
The numbers for my calcs, were based on the TF 190 heads just to have something to use as a model.

I like the KISS theory, but I do like to understand why it works which is why I'm interested in getting down into the weeds. It's easier to make compromises when I know what I'm giving up or gaining.
Click to expand...

You Never know what your given up, Nascar F1 etc.. engines are so expensive cause all of the dyno time trying different things, Say your building a hot street strip car even after all your efforts you might have to try 2-3 different cams carbs rear gears stall converter tires etc.. until you dial in the best combo. Especially if your shooting for more than just acceptable.
 
Last edited:
My GAWD. You can already send yourself to the funny farm just making a cam choice based on simple stuff like cubic inch, compression, induction, exhaust, converter, gears, weight and so on, I choose not to complicate it further than that. This is supposed to be a FUN hobby. What's next? Wantin to know how many molecules are in each individual metal in the engine and use that as a determining factor? I guess this is good for discussion's sake, but I'm not gonna think that hard.
 
To be clear I would really just like to have some fun. My day job involves high performance computing. So I tune systems of systems, but different. I like to think it's similar in nature since it's all input and output. I just work in terms of bandwidth, throughput and operations per second etc.

At the end of the day I'd like to be able to make good, solid educated decisions on a given engine build and remove most of the guessing game aspect that's common amongst beginners like me.
 
Cuda416 said:
To be clear I would really just like to have some fun. My day job involves high performance computing. So I tune systems of systems, but different. I like to think it's similar in nature since it's all input and output. I just work in terms of bandwidth, throughput and operations per second etc.

At the end of the day I'd like to be able to make good, solid educated decisions on a given engine build and remove most of the guessing game aspect that's common amongst beginners like me.
Click to expand...


Best thing is to look at tons of builds with dyno result no matter the engine brand and look for the trends. When thinking about my 273 I looked at a lot of Ls4.8l and 2v ford 4.6l and even 302/305 since there's little info on 273's. You'll see similar head flow cams cr cid will have similar results across brands and similar head flow cams cr with different cid changes where power is made.

Here's a thread I started with all the mopar builds I've found

Magazine Builds
 
You must log in or register to reply here.
-

Back
Top