Your Intake Manifold and You

[Mazdazilla6]

Okay, so as some know I've been doing some flow testing of parts on a flow bench. Some of these parts are intake manifolds. The main reason why I'm doing this is because it's never really been done before. Flow testing parts is just like dynoing a vehicle. It gives you a good idea on how something is performing but it's fairly hard to accurately compare to results of others. Just like dynos, bench results can differ based on ambient conditions, bench calibrations, accuracy of the device, test pressure, etc. So when a company claims that their part flows xxx CFM it honestly doesn't really say much unless they can give more details, and most of the time they don't.

Before I get to my results I would just like to say, PLEASE do not compare my results to that of what others have posted. I have found little in the ways proper documentation for flow tested pieces and I really don't want this to turn into a, "well ______ said that their part flows xxx though". I don't care what a company has to say about their part, the point of this thread is to give an unbiased opinion, with properly documented results.


I'm going to start by covering some of the basics of the tests that pertain to all of the manifolds tested. The flow bench being used is a SuperFlow SF-600. This bench has a FlowCom computer attached to it which allows for great accuracy and ease of testing. All tests were done at a test pressure of 25" of H2O with a max variation of .3" H2O. The industry standard pressure for flow testing in the performance industry is 28" H2O. I chose to use 25" H2O because that is the pressure that the bench is calibrated for. I could have tested at 28" H2O but I decided I would rather have the most precise numbers I could get. The only real difference between 25" H2O and 28" H2O is that the test results would be a bit higher if the tests were done at 28" H2O. All results are in Cubic Feet per Minute (CFM).

To set a baseline I started with a totally stock intake manifold. The manifold was sealed with modeling clay to an adapter plate and sealed to the bench with clamps and a rubber gasket. I can't say for sure there weren't any leaks but if there were, it was less than a couple of CFM (aka mostly negligible). The VTCS was left intact and tests were done with the flappers open as well as closed. All ports not being tested were covered in tape so all air would be pulled through the throttle body opening. Lastly, every port was given a small exiting radius made of clay. This was done to help seal as well to keep variation of adapter plate placement a non factor.

Other ports sealed off:


The adapter plate sealed to the manifold:


Here are the results:
stock.png

Next up was a gasket matched port and polished manifold. VTCS was removed but the holes were not sealed between runners. The same sealing methods were employed as before.

Manifold ports:


Sealed:


Here are the results:
gasket match.png

The last stock variation of a stock manifold was a single runner manifold. Essentially a gasket matched manifold with a bit of throat porting and the runner divider removed. The same sealing methods were employed as previously.

Manifold ports:


Here are the results:
single runner.png

@maisonvi; was generous enough to lend me two JMF manifolds. One is a single runner set up for Pi and the other is a split runner set up for meth. I was able to get the single runner manifold tested but not the split runner because I ran out of time but I will test the split runner when I return from my break next week. All of the PI bungs were taped off as well as the holes for the vacuum lines so all air would be drawn through the throttle body opening. The same sealing methods were used as the other tests.

Manifolds:


Here are the results:
JMF single runner.png


So what are my thoughts on this? I discussed it with some of the IL Nator guys and I think that the reason for the flow imbalance between the runners on the stock manifold is because of the throttle body location. In order for air to go up the cyl 4 runner, it must make an almost 180* turn; that is something air does not like to do. On the other hand, to get to cyl 1 runner air has a mostly straight shot, not having to bend much. As to why 2 and 3 flow more than 1 and 4 I think is because of how the runners are connected to the plenum and the path they take is shorter than 1 and 4. Here's an illustration of what I mean:

mni.jpg

As for why the JMF flows more on 1 and 4 I'm thinking it has something to do with this:
FST Intake Flow 1 No Dividers.jpg

FST Intake Flow 2 No Dividers.jpg

This flow imbalance was noted by SP63 on JMF's mani for the Focus ST. I believe that we are seeing the same thing on our manifold.

UPDATE: I've finished the testing for the JMF split runner mani by itself. Now that all the mani's I have have been tested by themselves I can finish testing with them connected to the head. The results for it are in the picture below. To my surprise the manifold flows ~15-20% less than the open runner manifold. I'm not really sure why the flow is so much less from this other than maybe it is causing turbulence in the port. Though I'm not really sure about that because the divider is CNC knife edged so it doesn't seem like it would really cause a lot of turbulence. Maybe the results with it connected to the head will tell a bit more.

Edit for the update: Actually after looking at the difference between the pnp vs the single runner stock manifold the flow differences are fairly similar to the differences between the JMF manifolds so it looks like the divider is causing some turbulence of a sort.

Here is a more comparative view of the results. Please let me know what you guys think on this! If anyone has any non stock manifolds that I haven't tested yet and would like it tested please let me know!

Screen Shot 2016-03-17 at 11.24.50 AM.png

[Vansquish]

Looks like you need to get your hands on a CS manifold and a CP-E manifold to complete your testing. Great data so far, thanks for your efforts!

[g00s3y]

Awesome work, I was always curious to see actual numbers. Thanks a bunch for taking the time to do this. I'm sure many will appreciate it.

[speedfreak44]

What do you think of one of these single runner designs and what effect on flow the head will have when it hits the divider in the head? I know the flow bench shows peak flow when there is no obstruction at the end of the port. These are some pretty interesting results. Thanks for the work

[Mazdazilla6]

Originally Posted by speedfreak44 What do you think of one of these single runner designs and what effect on flow the head will have when it hits the divider in the head? I know the flow bench shows peak flow when there is no obstruction at the end of the port. These are some pretty interesting results. Thanks for the work

I've already put considerable thought into that. I've already begun testing the combos of manis mounted on heads. I can assure you I will report my findings after I've compiled the data.

[speedfreak44]

You the man!

[Mazdazilla6]

Originally Posted by speedfreak44 You the man!

Don't expect to see it too soon though. Testing these manis by themselves took me a couple of days. Testing them mounted on heads will take me a couple of weeks.

[speedfreak44]

Hey free is free I can not complain. I already have a ported stocker installed so it isn't like I will be changing within the week anyway. Just curious!

[GCPhotographers]

Some strong results. I agree that that the results from CPe and Corsport would be great. I think even after Justin from Freektune completes the test, we will somewhat know the direction the market on our platform will turn. He loves his JMF mani so numbers from Cp-e would be very enlightening... then again, I don't think Corksport will compare in the league with JMF or Cp-e (if Cp-e numbers are effective)..

[Mazdazilla6]

Originally Posted by GCPhotographers Some strong results. I agree that that the results from CPe and Corsport would be great. I think even after Justin from Freektune completes the test, we will somewhat know the direction the market on our platform will turn. He loves his JMF mani so numbers from Cp-e would be very enlightening... then again, I don't think Corksport will compare in the league with JMF or Cp-e (if Cp-e numbers are effective)..

Well the inbalance can be overcome on the JMF. They fixed it on the Focus ST manifold but chose not to alter it for ours. I think CS needs to be competitive with Cp-e and JMF in order for them to really sell. Realistically, I don't see them being able to sell a cast manifold for much less than $600. At that point, if it doesn't flow so great, you're better off spending an extra $100 and buying a JMF.

[GCPhotographers]

Originally Posted by Mazdazilla6 Well the inbalance can be overcome on the JMF. They fixed it on the Focus ST manifold but chose not to alter it for ours. I think CS needs to be competitive with Cp-e and JMF in order for them to really sell. Realistically, I don't see them being able to sell a cast manifold for much less than $600. At that point, if it doesn't flow so great, you're better off spending an extra $100 and buying a JMF.

Great point. Truly CS I think it on the level of producing parts for enthusiast that are looking for just a little more power so yes, the mani that they will produce will not be on the scale like JMF and the potential in Cp-e. Great notes. Looking for an update with the Cp-e mani if you get your hands on it or someone else.

[cletusb]

Great results keep them coming for sure

Has there been any word from JMF about matching ours to the new focus design or radio silence?

[Justin@Freektune]

Awesome test results. Really curious to see the divided runner results for the JMF as well.

Have to give it a little more thought. But even the least flowing runners on the JMF will outflow the cylinder head. In that case the head would become the equalizer of balanced airflow to the cylinders.




Sent from my iPhone using Tapatalk

[speedfreak44]

therefore, JMF or gtfo?

[GEE1989]

Noob question haha. Based on the stock mani flow rates. Couldn't that be balanced out by just porting 1 a little and a full ported double runner on 4? Im just looking at the numbers and seeing that could balance the flow right around 200

Sent from my SM-G920P using Tapatalk

[Mazdazilla6]

Originally Posted by cletusb Great results keep them coming for sure Has there been any word from JMF about matching ours to the new focus design or radio silence?

I've seen nothing from JMF. Maybe consider contacting them to see if they've done it one their newest version.

Originally Posted by Justin@Freektune Awesome test results. Really curious to see the divided runner results for the JMF as well. Have to give it a little more thought. But even the least flowing runners on the JMF will outflow the cylinder head. In that case the head would become the equalizer of balanced airflow to the cylinders. Sent from my iPhone using Tapatalk

I talked with @xfeejayx; about that very same thing. The stock head flows very evenly, but the imbalance in the runners will cause different pressures on a running engine, thus causing imbalanced flow again. Er that was the conclusion that we came to.

Originally Posted by GEE1989 Noob question haha. Based on the stock mani flow rates. Couldn't that be balanced out by just porting 1 a little and a full ported double runner on 4? Im just looking at the numbers and seeing that could balance the flow right around 200 Sent from my SM-G920P using Tapatalk

It's hard to say for sure because a flow bench doesn't really fully replicate an operating engine. The goal is to keep cross sectional area similar between ports at the same places, doing that still promotes imbalance with our manifold.

[silvapain]

It's not just air flow rate that's important, but also air speed. The air needs to be moving fast enough to be turbulent, to promote even fuel dispersion and atomization in the combustion chamber. That's why one of the intake ports has a flapper and is closed at low engine load; to increase air speed into the combustion chamber.

Differences in port cross-sectional area will affect each cylinder's efficiency and still be less than ideal.

[Mazdazilla6]

Originally Posted by silvapain It's not just air flow rate that's important, but also air speed. The air needs to be moving fast enough to be turbulent, to promote even fuel dispersion and atomization in the combustion chamber. That's why one of the intake ports has a flapper and is closed at low engine load; to increase air speed into the combustion chamber. Differences in port cross-sectional area will affect each cylinder's efficiency and still be less than ideal.

Very much agree. But turbulence before CC is not desirable (to a point). I believe a Reynold's number right around 3000-3500 would be good for airflow and atomization (if you have PI or meth).

I have Pitot tubes to measure air velocity. I've thought about it but I'd need to make an adapter to mount in on the bench when it's set to blow.

[GroceryGtr]

Hmm this is going to be a really good thread. With the imbalanced runners right now it seems we may have some hidden power just by getting them balanced out. Think about it right now we are tuning, knocking, etc to the cylinders getting "full air". If we could ensure all are delivering equally that's a few extra ponies locked up I the non-full runner/cylinder. I would love to see the a/f ratios for each cylinder on these cars, be neat to see how much they vary with the airflow differences in the runners.


Sent from my iPad using Tapatalk

[sheston]

I saw a comment on outflowing the heads. Doesn't anyone know off the top of their head what that number is?

Surprising to see that the single runner doesn't get a bigger flow increase over the gasket matched. I'm feeling better about just doing the gasket match.

Do any of the brainiacs on here know if we would have to measure these connected to the head to really be representative? Seems part of the equation in impedance matching, not just flow. I always wondered if the single runner would have problems with that?

[Mazdazilla6]

Originally Posted by sheston Do any of the brainiacs on here know if we would have to measure these connected to the head to really be representative?

That is stage dos. I've already started collecting data. Will be a couple weeks til testing is complete.

[sheston]

Excellent, and thanks

[Mazdazilla6]

OP updated with JMF split runner results. Results were a bit more drastic than I thought they'd be. Let me know what you guys think. Both of the JMF connected to the stock head results should be done in a week or so. Been a bit slow recently, trying to catch up on school work since I'm back from break. Happy St. Pattys Day guys.

[sh4d0w]

You my friend rock! This info is amazing.

[Easter Bunny]

Originally Posted by Mazdazilla6 OP updated with JMF split runner results. Results were a bit more drastic than I thought they'd be. Let me know what you guys think. Both of the JMF connected to the stock head results should be done in a week or so. Been a bit slow recently, trying to catch up on school work since I'm back from break. Happy St. Pattys Day guys.

*Paddy's*

[WetzMS3]

Originally Posted by Easter Bunny *Paddy's*

Good to see that engineering license paying off again around here.

[3deepsadzam]

This is a great topic I can understand a quick portmatching and polishing the runners on a stock mani and a vcts delete, the results with a single runner On the stock mani showed a postive gain, but say your the averag joe with out getting you mani bench tested after you ported/machined, your doing more harm then good, correct ,possible you would make the flow rates more inequal than what they already are and is this worth the hassle from say buying jmf you proved its just about 80-100% more flow?

[Mazdazilla6]

Originally Posted by 3deepsadzam This is a great topic I can understand a quick portmatching and polishing the runners on a stock mani and a vcts delete, the results with a single runner On the stock mani showed a postive gain, but say your the averag joe with out getting you mani bench tested after you ported/machined, your doing more harm then good, correct ,possible you would make the flow rates more inequal than what they already are and is this worth the hassle from say buying jmf you proved its just about 80-100% more flow?

Sorry dude, not to be an ass but I can't really make out what you're saying or asking due to your punctuation er lack thereof.

[sh4d0w]

Originally Posted by 3deepsadzam This is a great topic I can understand a quick portmatching and polishing the runners on a stock mani and a vcts delete, the results with a single runner On the stock mani showed a postive gain, but say your the averag joe with out getting you mani bench tested after you ported/machined, your doing more harm then good, correct ,possible you would make the flow rates more inequal than what they already are and is this worth the hassle from say buying jmf you proved its just about 80-100% more flow?

I think I get what he's asking i'll take a stab at answering you. From what I gather from this thread, in short NO. The P&P on the stock intake is not doing more harm than good. It flows more even though the flow might be a little more uneven the JMF flows very uneven on the 1st and 4th runners. I can't see why it would be an issue on the stock one. (someone please correct me if I'm wrong here)

Just my 2 cents.


One question I would have is what's the main difference between a single runner mani and the dual runner one? Obviously the single runner flows better so what's the disadvantage, if there is any in running a dual runner vs a single?

Just trying to figure out why anyone would run a dual runner mani in the first place if the single one flows so much more.

[3deepsadzam]

if your selfporting you stock intake manifold ,is it even worth it, if you have to get it flow tested to check equal flow in every runner and the gains are minimal the question isnt it worth it to just buy jmf intake

[sh4d0w]

The numbers are all there in this thread. Numbers don't lie. Look at the difference % for each mani. The ported one isn't that far off from even the stock difference %.

[Mazdazilla6]

Originally Posted by 3deepsadzam if your selfporting you stock intake manifold ,is it even worth it, if you have to get it flow tested to check equal flow in every runner and the gains are minimal the question isnt it worth it to just buy jmf intake

Yes it is still worth it to port the stock manifold. The manifold is more restrictive than the head so porting the manifold will still pick up airflow gains even if it doesn't necessarily even out the flow a bunch.

Originally Posted by sh4d0w I think I get what he's asking i'll take a stab at answering you. From what I gather from this thread, in short NO. The P&P on the stock intake is not doing more harm than good. It flows more even though the flow might be a little more uneven the JMF flows very uneven on the 1st and 4th runners. I can't see why it would be an issue on the stock one. (someone please correct me if I'm wrong here) Just my 2 cents. One question I would have is what's the main difference between a single runner mani and the dual runner one? Obviously the single runner flows better so what's the disadvantage, if there is any in running a dual runner vs a single? Just trying to figure out why anyone would run a dual runner mani in the first place if the single one flows so much more.

Thanks. To answer your question the advantage of the dual runner manifold is to keep the port runner divider a longer length. Theory says that this would help with low RPM performance. Keep in mind these flow numbers tell max flow but now how the piece will perform overall.

More of this will be made clear after I am done testing with these connected to the head because the valves are kept in it so I am able to measure flow at low lift points.

Also, the JMF by itself outflows a stock head by a pretty sizable amount so the difference between a split and single runner mounted to a stock head *should* in theory flow fairly similar. Tests will tell tho.

[speedfreak44]

I believe the only way to make our flow even out is to run an intake manifold that is designed like a tri Y header. But to be fair, we have seen 800whp out of a jmf...flow imbalance or not shit is working

[sh4d0w]

This thread actually gives me more reason to pnp my stock mani. Can't wait to see the rest of your results. I love this kind of information.

[Mazdazilla6]

Originally Posted by speedfreak44 I believe the only way to make our flow even out is to run an intake manifold that is designed like a tri Y header. But to be fair, we have seen 800whp out of a jmf...flow imbalance or not shit is working

Not necessarily. Take a peak at this thread about the JMF mani for the FoST. They had a similar imbalance on it and fixed it without altering exterior dimensions. The goal is to have the plenum flowing the same to each runner.

SPEED PERF6RMANC3 & JM Fabrications Intake Manifold

But yes I agree the JMF is by no means a bad manifold.

[3deepsadzam]

I understand you obviously get more flow p&p and with a singal runner,how do you know your getting the consistent flow +- with out flow testing I see all these guys porting ther manifolds and eyeballing what material there grimding away and not knowing what the flow is I'm just saying Is there a max tolerance a runner can flow from another before it causes a problem

[sh4d0w]

Hey I'm curious I just found this on a website. Would this math apply here?

The horsepower potential of an engine can be calculated by the airflow capability of the cylinder head and intake manifold. Airflow testing of cylinder heads and intake manifolds can be conducted on a flow bench. Horsepower calculations can be estimated from the flow bench test data as follows:
The standard measure for flow testing is 28" of test pressure on a Superflow 600 flow bench.
The formula for calculating horsepower from flow test data measured at 28" of test pressure is as follows:

HP = 0.255 x flow test data at 28 inches of test pressure

Example:
Intake airflow of 250 cfm at 28" of test pressure can produce 510 hp.
250 cfm x .255 = 63.75 hp per cylinder
63.75 hp per cylinder x 8 cylinders = 510 hp.

I guess the fact that we run forced induction would make the point moot but I'm still kind of curious.


O.K. just re-read your original post. You tested at 25'' Which basically answers my question.

Sweet!

[speedfreak44]

I did read about that jmf design but I feel it is more of a bandaid than a resolution to the issue. They did not really balance it all the great in my opinion. Best bang for the buck here is the pnp stocker from what I see and have gathered from past expirements as well. Your testing is definitely top notch though and greatly appreciated. It is a worthwhile read for sure. Now someone design me a tubular tri y intake manifold and let me test it out

[sh4d0w]

Am I right in assuming just porting the stock mani would basically net a gain of around 30hp? Without even touching the tune?

[speedfreak44]

Negative. Gained about 5-10whp peak tuned