Author Topic: Flow numbers: Cylinder head vs Intake manifold?  (Read 4139 times)

0 Members and 1 Guest are viewing this topic.

plovett

  • Guest
Flow numbers: Cylinder head vs Intake manifold?
« on: November 15, 2020, 12:31:58 PM »
When is your intake manifold too much for your combination and more specifically, the heads?

Do you look at cross sectional area?  Do you look at flow?  Vehicle combination?  All of it?

I am going to mention flow first.  We know that both cylinder heads and intake manifolds are flowed on a flow bench.   In neither case do the conditions on the flow bench, e.g. a steady 28" of vacuum, replicate what happens in a running engine.  But using the flow data is obviously very useful.  There are various rule of thumb calculations relating cfm and potential horsespower.   You can look at flow vs . cross sectional area and/or volume.  You can look at discharge coefficient which is a ratio of actual flow vs. an ideal theoretical flow.  In engines I think the ideal theoretical flow is based on the valve size area.  The ideal at 28" depression is supposed to 146 cfm per square inch of valve area minus the stem area. Anyway, flow numbers are very useful from a flow bench even though they don't mimic a running engine.  You could say the dyno is the ultimate flow bench.

When an intake manifold is flowed on a bench I believe 7 of the ports are plugged up and a fixture is used to pull air through the remaining port in the direction of the cylinder head.  This gets measurable and repeatable numbers, but it certainly doesn't replicate the chaotic environment in an intake manifold in a running engine.

Improvements in flow on a bench have fairly well understood effects in a running engine, so the bench is a great tool.  It is not always the case that more flow equals more power.  There are other factors like port speed, port volume, average cross sectional area, minimum cross sectional area, port shape, and on and on and on.

So,  back to cylinder heads and intake manifolds.   You can get a flow number for each, but how do you decide if they are compatible? 

I've read a rule of thumb that the intake by itself should flow about 25% more than the head by itself.  What about port size?  Does it matter?

What about flowing the intake and head together?  What does that tell you?  Do you look for the flow to diminish as little as possible after the intake is bolted on to the head?

I have a specifc question.   I have some ported Edelbrock heads.  They flow 302 cfm with 2.09" valves.   I plan on getting bigger valves put in, 2.19" or 2.20" and maybe open up the throats to match.  I expect the flow to pick up some.  I might get 315-320 cfm, I dunno.  The ports themselves are roughly a slighty larger medium riser size/ I can't measure them right now because they are at the machine shop. 

I have two intake manifolds.  Both are ported. One is a BT 1x4 dual plane.  It's average flow is 401 cfm per runner.  It also has slightly large medium riser size ports.   The other intake is a Dove Tunnel Wedge. 2x4 single plane.  It's average flow is 474 cfm per runner.  The thing is, the ports, while still roughly medium riser size, are actually smaller than the ports in either my dual plane intake or my cylinder head.

So applying the 25% rule, the dual plane intake would seem like the easy choice.  If flows enough to feed my heads given that measure, plus the added benefit of long separated runners.

On the other hand the single plane intake actually has smaller runners, but flows a butt-ton more.  Is all that extra flow just "wasted"?

I don't know the current manifold volumes.  Per Jay's book an unmolested BT1x4 in the medium riser version is 4950cc.  An unmolested Dove is 4125cc.    I don't think my intake's volumes have changed too much.

I understand there are vehicular constraints, too.  You might put the dual plane in a 4000 car with higher gearing even though it made less power.  Or you might put the single plane in your Cobra kit car even though power below 4000 was much lower.  I get that.  I mostly want to avoid getting into the particulars of a vehicle and concentrate on the engine, though.

Certainly the cam and intended power range of the engine are key.  Would that be the deciding factor?   Or is "enough" intake flow simply enough, regardless?

Just looking at an intake manifold and cylinder heads,  How do you pick the intake?   

Cylinder heads:        320 cfm     MR port (larger)
Dual plane intake:    401 cfm     MR port (larger)
Single plane intake:  474 cfm     MR port (smaller)

thanks,

paulie




« Last Edit: November 15, 2020, 01:08:47 PM by plovett »

427John

  • Sr. Member
  • ****
  • Posts: 343
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #1 on: November 15, 2020, 01:05:35 PM »
I think a lot of it depends on the intended usage for the vehicle,and your priorities of street manners vs max power.Its kind of like 429-460 guys comparing a wieand stealth intake to the edelbrock RPM,until you start getting into the bigger strokers the e-brock rpm seems to be favored for better low rpm response while the stealth is preferred for WOT power due to the larger internal volume.As the engine gets bigger the though the stealths street manners improve.Disregard this after rereading your post it becomes clearer that your main concern is WOT.
« Last Edit: November 15, 2020, 01:09:35 PM by 427John »

My427stang

  • Hero Member
  • *****
  • Posts: 3918
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #2 on: November 15, 2020, 01:15:28 PM »
Paulie, dig into some some math on runner length and plenum volume on line, it's out there, and there are as many assumptions and crazy ideas as there are good ones. Even the old Ramcharger stuff can be eye opening.  Get on excel build a spreadsheet and play

In the end, a runner can be too long, but not likely in the RPM we play with.  It can definitely be short. In terms of plenum volume, I read a comment a long time ago, Plenum volume = 1 cyl is likely the minimum, volume = all 8 probably too much LOL

I would say that likely your big intake will make the HP a sharp and higher peak and hurt the torque peak.  I think the single 4 is likely plenty, and would broaden/flatten both torque and HP, but my gut tells me, the biggest tradespace will be the torque curve

I think if I was playing with a streetable Edelbrock head 600+ hp motor, I would lean toward a ported Victor or TFS for a single plane, or a med riser dual plane dual quad with some rubbing, as applied to your combo (well at least what I think your combo is going to be without cam, compression and solid head flow yet) 

Of your two, I can't say the single 4 would make more power, but it would make more power per dollar, and likely be happier part throttle
---------------------------------
Ross
Bullock's Power Service, LLC
- 70 Fastback Mustang, 489 cid FE, Victor, SEFI, Erson SFT cam, TKO-600 5 speed, 4.11 9 inch.
- 71 F100 shortbed 4x4, 461 cid FE, headers, Victor Pro-flo EFI, Comp Custom HFT cam, 3.50 9 inch

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #3 on: November 15, 2020, 01:17:24 PM »
I think a lot of it depends on the intended usage for the vehicle,and your priorities of street manners vs max power.Its kind of like 429-460 guys comparing a wieand stealth intake to the edelbrock RPM,until you start getting into the bigger strokers the e-brock rpm seems to be favored for better low rpm response while the stealth is preferred for WOT power due to the larger internal volume.As the engine gets bigger the though the stealths street manners improve.Disregard this after rereading your post it becomes clearer that your main concern is WOT.

Yes, my main concern is WOT.  I think your post is still pertinent, though.  We often look at cylinder heads using flow vs volume as a measure.  That is one reason I listed the volumes of the two intakes I am talking about.  I am just not sure how to wrap my head around intake manifold vs. cylinder head flow. 

thanks,

pl

 

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #4 on: November 15, 2020, 01:29:39 PM »
Paulie, dig into some some math on runner length and plenum volume on line, it's out there, and there are as many assumptions and crazy ideas as there are good ones. Even the old Ramcharger stuff can be eye opening.  Get on excel build a spreadsheet and play

In the end, a runner can be too long, but not likely in the RPM we play with.  It can definitely be short. In terms of plenum volume, I read a comment a long time ago, Plenum volume = 1 cyl is likely the minimum, volume = all 8 probably too much LOL

I would say that likely your big intake will make the HP a sharp and higher peak and hurt the torque peak.  I think the single 4 is likely plenty, and would broaden/flatten both torque and HP, but my gut tells me, the biggest tradespace will be the torque curve

I think if I was playing with a streetable Edelbrock head 600+ hp motor, I would lean toward a ported Victor or TFS for a single plane, or a med riser dual plane dual quad with some rubbing, as applied to your combo (well at least what I think your combo is going to be without cam, compression and solid head flow yet) 

Of your two, I can't say the single 4 would make more power, but it would make more power per dollar, and likely be happier part throttle

It has occurred to me that the BT 2x4 MR intake would be good.  I have commented in the past how amazing it seems to run on a very wide range of engines, like from 350 hp up to 650 hp or so.  Freaking amazing.    On the other hand,  it's hard not to have enough low end power in a well built decent sized FE.   That's not what I wanted to get into, though.  LOL!

The question is more about comparing intake manifold flow numbers and cylinder head flow numbers.     I know fairly well where my head flow is going to be and the specific number won't vary enough to affect the answer to the question.  Trying to keep it on track. 

My Tunnel Wedge flowed about 410 cfm untouched if I remember right.  What then?  I assume that would make the 400 cfm dual plane an even easier choice?  Of course the runners were even slightly smaller then, too.

thanks,

pl
« Last Edit: November 15, 2020, 01:31:31 PM by plovett »

Joe-JDC

  • Hero Member
  • *****
  • Posts: 1489
  • Truth stands on its own merit.
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #5 on: November 15, 2020, 02:16:07 PM »
What most folks forget is that airflow changes with volume and velocity with the addition of length of the intake manifold.  Many years ago, I worked with EFI manifolds on my flow bench trying to answer your question to my satisfaction.  I found that the airflow cfm in a head flowed at 28" would pull down quite a bit when bolting the intake manifold to the head.  I also tried it some by checking the difference with different lifts of the valve, and was surprised how much things change.  It is VERY difficult to get a manifold to bolt up to a head, and the flow NOT drop a little/lot with a cast manifold, especially dual plane intakes.  The FE is one of the hardest to get the flow to stay up because of the turns and twists of the intakes with the exception of the single plane intakes.  A sheetmetal intake is designed with the idea that you not only can match the head flow, but actually increase the volumetric efficiency to upwards of 125% as in a Pro Stock engine.  To get the FE engine to perform at the maximum potential, you will find the best engines use sheetmetal intakes.  The reason I came up with the 110% street/125% race is directly related to flow bench testing, and David Vizard's formulas in his flow bench booklet from the '80s.  Lots of folks don't like DV's attitude, but he is successful with his builds, though now things have begun to pass him up with all the computerized flow simulations, and CNC porting.  The fact is, though that if you look at what works, line of sight, taper, parallel walls, and port volume all help to fill a cylinder to maximum.  A manifold should not pull the head down much if any, and it takes a lot of work to get all the runners to flow close to even in the FE dual plane intakes.  Edelbrock has done a great job with the RPM, and the new TFS Track Heat is simply the best single plane available out of the box.  They can be improved upon, and not lose performance with the right port work.  The Blue Thunder intakes are very good as cast, but have a lot of variance between the different ports in flow cfm, but can be corrected with flow bench work.  The BT 4V can be made better than the RPM with work.  Having said all that, most folks don't realize that the original FE heads like the CJ only flow ~250 cfm, and the PI intakes 270 cfm average, as cast.  The Sidewinder is less than 20 cfm better, and the RPM out of the box is ~310 cfm average.  Very few folks have the ability to actually flow intake manifolds and very few actually know what their work does flow.  Most just gasket match, and if it looks good, sell it.  I have flowed Wilson CNC'd manifolds that had a 80 cfm difference between their runners, and cost $2500.00 for their work.  Joe-JDC 
Joe-JDC '70GT-500

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #6 on: November 15, 2020, 02:35:08 PM »
What most folks forget is that airflow changes with volume and velocity with the addition of length of the intake manifold.  Many years ago, I worked with EFI manifolds on my flow bench trying to answer your question to my satisfaction.  I found that the airflow cfm in a head flowed at 28" would pull down quite a bit when bolting the intake manifold to the head.  I also tried it some by checking the difference with different lifts of the valve, and was surprised how much things change.  It is VERY difficult to get a manifold to bolt up to a head, and the flow NOT drop a little/lot with a cast manifold, especially dual plane intakes.  The FE is one of the hardest to get the flow to stay up because of the turns and twists of the intakes with the exception of the single plane intakes.  A sheetmetal intake is designed with the idea that you not only can match the head flow, but actually increase the volumetric efficiency to upwards of 125% as in a Pro Stock engine.  To get the FE engine to perform at the maximum potential, you will find the best engines use sheetmetal intakes.  The reason I came up with the 110% street/125% race is directly related to flow bench testing, and David Vizard's formulas in his flow bench booklet from the '80s.  Lots of folks don't like DV's attitude, but he is successful with his builds, though now things have begun to pass him up with all the computerized flow simulations, and CNC porting.  The fact is, though that if you look at what works, line of sight, taper, parallel walls, and port volume all help to fill a cylinder to maximum.  A manifold should not pull the head down much if any, and it takes a lot of work to get all the runners to flow close to even in the FE dual plane intakes.  Edelbrock has done a great job with the RPM, and the new TFS Track Heat is simply the best single plane available out of the box.  They can be improved upon, and not lose performance with the right port work.  The Blue Thunder intakes are very good as cast, but have a lot of variance between the different ports in flow cfm, but can be corrected with flow bench work.  The BT 4V can be made better than the RPM with work.  Having said all that, most folks don't realize that the original FE heads like the CJ only flow ~250 cfm, and the PI intakes 270 cfm average, as cast.  The Sidewinder is less than 20 cfm better, and the RPM out of the box is ~310 cfm average.  Very few folks have the ability to actually flow intake manifolds and very few actually know what their work does flow.  Most just gasket match, and if it looks good, sell it.  I have flowed Wilson CNC'd manifolds that had a 80 cfm difference between their runners, and cost $2500.00 for their work.  Joe-JDC

Thanks Joe!  Good info. 

Let me ask this,  If you were looking at an intake and it flowed say 370 cfm, and your heads flowed 250 cfm, could you categorically say that the intake is too much for the heads?  That it is simply not optimal? 

Or would you say hold on,  let's look at the CSA and the runner length, and the plenum volume, etc?   

Or something else?

pl

My427stang

  • Hero Member
  • *****
  • Posts: 3918
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #7 on: November 15, 2020, 03:01:46 PM »
Not Joe but I would do the latter

An extended, long smaller ID tapered pipe flowing 370 is different than a short larger one, in terms of power production

That’s why so many say don’t live or die by flow numbers. I like Joe’s estimate a lot, but it isn’t meant to determine how and where two different manifold designs behave in power, It’s created to let him know where he needs to be to provide enough to feed the head

However I have also been caught in the Paulie trap that I didn’t answer the question you asked....If I went back to the original question what I would say is first intake must flow more than head, then go dig into the other stuff to make it really do what you want
« Last Edit: November 15, 2020, 03:08:22 PM by My427stang »
---------------------------------
Ross
Bullock's Power Service, LLC
- 70 Fastback Mustang, 489 cid FE, Victor, SEFI, Erson SFT cam, TKO-600 5 speed, 4.11 9 inch.
- 71 F100 shortbed 4x4, 461 cid FE, headers, Victor Pro-flo EFI, Comp Custom HFT cam, 3.50 9 inch

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #8 on: November 15, 2020, 03:37:30 PM »
Not Joe but I would do the latter

An extended, long smaller ID tapered pipe flowing 370 is different than a short larger one, in terms of power production

That’s why so many say don’t live or die by flow numbers. I like Joe’s estimate a lot, but it isn’t meant to determine how and where two different manifold designs behave in power, It’s created to let him know where he needs to be to provide enough to feed the head

However I have also been caught in the Paulie trap that I didn’t answer the question you asked....If I went back to the original question what I would say is first intake must flow more than head, then go dig into the other stuff to make it really do what you want

It's not a trap.  It's a question.  LOL!  We use the tools and the information we have, combined with our experience.  Flow benches are a big part of that in my opinion.   

Question is, using my specific parts, is this.  Is an intake manifold flowing 474 cfm automatically too much for a head that flows about 320 cfm? That's a real question.  If you don't know the answer that is fine.  I don't know the answer.  But don't take it personally.  I'll try not to make up more "traps" for you.  :)

pl

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #9 on: November 15, 2020, 03:44:12 PM »
You can say the original question doesn't make sense.  That is an answer and one of the possible outcomes I am looking at.

pl

427John

  • Sr. Member
  • ****
  • Posts: 343
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #10 on: November 15, 2020, 03:53:07 PM »
A case of quality being as important as quantity,while quantity is measurable with access to a flow bench quality not so much.

My427stang

  • Hero Member
  • *****
  • Posts: 3918
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #11 on: November 15, 2020, 03:56:20 PM »
Fair enough, here is my answer after two very good points you just made

474 cfm automatically being too much?

I do know the answer, but this isn't cut and dry like you want, it's closer to not the right variable being measured

1 - Simple answer... no, 474 isn't automatically too much.   

2 - More complex answer... unanswerable given the information of the simple question and without knowing the goals of the engine

3 - Jaded answer knowing the two intakes and presumed build....I think the TW "could" be overkill and be less efficient in more places than it is efficient.   Note, not BAD, just more inefficiencies than gains.

However, if you are trying to run in a very narrow RPM range and match everything to the peak of that manifold and head, then it could actually work OK at that peak even being too big (but my experience with an untouched TW is that it is real big, never mind one ground on).

So why do I not just answer your specific question?

Because of what I thought you wanted to dig into, in fact it's fascinating stuff, port length tuning, water hammer effect, cross section, plenum volume stuff, overlap and 5th cycle it's all cool stuff.  When I try to help what we call a "drinking duck" (compliment) I send them to info and they eat it like lions.  That's what I was doing and I am not offended

Again though, without looking at the whole manifold, flow alone, with no other variables only has to "be enough"   However, let me tickle it a bit, I'd be looking at primary pipe and cam design to include overlap characteristics  to consider where that engine would start being happy with either intake. 
---------------------------------
Ross
Bullock's Power Service, LLC
- 70 Fastback Mustang, 489 cid FE, Victor, SEFI, Erson SFT cam, TKO-600 5 speed, 4.11 9 inch.
- 71 F100 shortbed 4x4, 461 cid FE, headers, Victor Pro-flo EFI, Comp Custom HFT cam, 3.50 9 inch

My427stang

  • Hero Member
  • *****
  • Posts: 3918
    • View Profile
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #12 on: November 15, 2020, 03:57:02 PM »
A case of quality being as important as quantity,while quantity is measurable with access to a flow bench quality not so much.
 

Certainly well said and to the point
---------------------------------
Ross
Bullock's Power Service, LLC
- 70 Fastback Mustang, 489 cid FE, Victor, SEFI, Erson SFT cam, TKO-600 5 speed, 4.11 9 inch.
- 71 F100 shortbed 4x4, 461 cid FE, headers, Victor Pro-flo EFI, Comp Custom HFT cam, 3.50 9 inch

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #13 on: November 15, 2020, 04:11:06 PM »
A case of quality being as important as quantity,while quantity is measurable with access to a flow bench quality not so much.

Hmmm.  I'm listening.  I'm not trying to be funny here, but how would one describe this quality?  Highest density of air/fuel in the cylinder at spark, at any given rpm?  I think it would it ultimately be "measured" by the dyno and/or the dragstrip? 

A big plenum short runner intake might affect this quality negatively at low rpm and vice versa with a small plenum long runner intake.  But a flow bench just takes a "snapshot" at 28" and gives you one number.  Engine rpms, loads, throttle positions, etc.  can all vary at that same snapshot of info. 

Like I said,  I am just trying to wrap my mind around the significance of intake manifold flow numbers with respect to cylinder head flow numbers. 

pl

plovett

  • Guest
Re: Flow numbers: Cylinder head vs Intake manifold?
« Reply #14 on: November 15, 2020, 04:24:04 PM »
Fair enough, here is my answer after two very good points you just made

474 cfm automatically being too much?

I do know the answer, but this isn't cut and dry like you want, it's closer to not the right variable being measured

1 - Simple answer... no, 474 isn't automatically too much.   

2 - More complex answer... unanswerable given the information of the simple question and without knowing the goals of the engine

3 - Jaded answer knowing the two intakes and presumed build....I think the TW "could" be overkill and be less efficient in more places than it is efficient.   Note, not BAD, just more inefficiencies than gains.

However, if you are trying to run in a very narrow RPM range and match everything to the peak of that manifold and head, then it could actually work OK at that peak even being too big (but my experience with an untouched TW is that it is real big, never mind one ground on).

So why do I not just answer your specific question?

Because of what I thought you wanted to dig into, in fact it's fascinating stuff, port length tuning, water hammer effect, cross section, plenum volume stuff, overlap and 5th cycle it's all cool stuff.  When I try to help what we call a "drinking duck" (compliment) I send them to info and they eat it like lions.  That's what I was doing and I am not offended

Again though, without looking at the whole manifold, flow alone, with no other variables only has to "be enough"   However, let me tickle it a bit, I'd be looking at primary pipe and cam design to include overlap characteristics  to consider where that engine would start being happy with either intake.

I wasn't looking for a cut and dried answer to all situations, Ross.  You know that.  In fact, I think engines are far too complex for simple questions and answers.  But that is why I start out asking simple questions.  The "drinking duck" analogy is apt and appreciated.  But answering one question directly can lead to a better understanding.   If we first rule out that there is a simple relationship between two numbers, then we can move on.  That is where I am trying to go.   Again, I honestly don't know the answer.  Just asking the question.   Is this a "Kung Fu" episode?  :)

pl
« Last Edit: November 15, 2020, 04:30:42 PM by plovett »