DMS Performance intake manifold

[stuntman]

Billy, why are you trying to compare the two? The test is not apples to apples because you have a stock ECU and he has an infinity system. I'm not sure how well you are familiar with Infinity, but it makes a big difference in the dyno comparisons.

The size of a 68mm vs a 76mm tb is no where near as significant as comparing the two different ECUs that you two have. If the motor doesn't need more than the 68mm tb then the gains would be very minimal. I am doubting the motor is asking for much bigger than a 76mm at most, and I have seen test between 62mm and 72mm TBs(2.6L motor) that have shown a 1whp gain with almost identical curves. People make TB size out to be a lot more significant than it really is.

I'm comparing intake manifolds. What are you trying to imply? If you take two competent ECUs and tune them for the same motor, peak power really shouldn't change much since there the motor is still getting the same ideal fuel/air/timing. While the stock ECU wasn't adjusted, a tuned ECU might be able to get a little more power out of the Ported Stock Manifold, but the AFRs were in check and there was nothing standing out as an issue. Ideal, probably not, but still decent for a comparison.

 

[jones83]

I'm comparing intake manifolds. What are you trying to imply? If you take two competent ECUs and tune them for the same motor, peak power really shouldn't change much since there the motor is still getting the same ideal fuel/air/timing. While the stock ECU wasn't adjusted, a tuned ECU might be able to get a little more power out of the Ported Stock Manifold, but the AFRs were in check and there was nothing standing out as an issue. Ideal, probably not, but still decent for a comparison.

I feel like you think I may be attacking you because I disagree. I'm not implying anything, but simply disagree. You aren't dealing with two competent ECUs. You have one competent ECU and a stock ecu. Your comparison is invalid because your stock ecu can't adjust timing, add fuel for the new volume of air entering, change vtec engagement point, etc. You can absolutely change the peak whp number in a Honda by going from a stock ECU to a capable ECU. I have seen it time and time again for more than a decade.

And I didn't even get into how much the area under the curve is affected with the different ECUs.

 

[stuntman]

No, not at all. I understand and agree that an aftermarket ECU can improve the area under the curve and the peak power of the Ported Stock Manifold run. The extent isn't known until tested but its still a fair comparison because a optimized tune for the Ported Stock Manifold would yield a somewhat favorable power graph and would most certainly not be less.

But then you need to ask dmscrx if he optimized both of his tunes for the stock and aftermarket manifolds. If he didn't then it's even more comparable if both tunes were optimized for the stock manifold.

 

[jones83]

He should have tuned the car with the Infinity on the stock manifold first. I want to say that he did. The Infinity system doesn't come with a stock basemap from the factory for the NSX(afaik), and completely setting one up from scratch is part of the install process. I thought he made a thread on being the first with the infinity system or something for an NSX on this site as well.

The extent isn't known until tested but its still a fair comparison because a optimized tune for the Ported Stock Manifold would yield a somewhat favorable power graph and would most certainly not be less.

100% agree but how is your tune optimized is my concern with the comparisons that you bring up? Only thing you are working with is a stock ecu which isn't remapped for your mods. It is my understanding that the only thing that is going on with your car is that the a/f is acceptable correct? Simply changing the vtec engagement point(to a more favorable engagement point) on your car would change the outlook of your map unless you have used some type of tuning program I am unaware of?

Different variables - ECU tunes, dyno calibration, elevation, and headers are the first that come to mind on why it would not be good to compare. I'm not saying you wouldn't gain power by switching to this manifold, but without physically taking your manifold and giving it to DMS then there is a slew of reasons(to me) why your data isn't worth comparing.

For example: A different ECU than stock might make your curve look much better than his in the low end and mid range. You will never catch him on the top end due to the volumetric efficiency of his manifold at high rpm.

- - - Updated - - -

I have to leave for a lunch meeting, but I would love to chat more about this later. I love these types of discussions.

 

[stuntman]

Dynojets are pretty reliable and consistent tools. In the BMW world, in NC I can dyno a stock E9X M3 and one with a tune and test pipes and make within 5-10hp as a car with the same mods in CA with identical curves. The SAE correction factor helps.

His stock manifold pull does not look like the VTEC engagement point was changed and its a bit hard to tell for his dsmcrx Manifold pull. Changing the VTEC point just changes when you get on the (slightly) larger cam profile, so at say 6,500rpm the power level really wouldn't be changed in a lowered engagement point vs. stock because you're on the same cam at that rpm in both examples. The NSX's VTEC is a relatively simple crossover and not a whole lot changes compared to an iVTEC engine.

An optimized tune can improve parts of the graph but it probably won't do much to the general shape which is dictated by the VE of the motor. There aren't very many back-to-back header comparisons out there and most aftermarket ones don't make much more than the stock NA2s or change the graph a ton. They all have a very similar shape.

Due to all of this, I think it's a fair comparison and I don't think an optimized tune will get the ported stock manifold to the same mid-low end as stock or dsmcrx's inake, nor the top end breath ability.

You can start a different thread if you want to go off on a tangent, I don't want to hijack dsmcrx's thread.

 

[RYU]

Since you guys are getting into the weeds here these were the two questions I had that would affect the result.

1. What happens when you use the crappy CA 91 oct gas?
2. What total timing was added given the use of the EMS?

Sorry if dmxcrx already answered this. Good work here!

 

[BATMANs]

Just wondering, has the shape of that intake manifold been used in other applications?

 

[jones83]

Just wondering, has the shape of that intake manifold been used in other applications?

That shape is as old as dirt and used across the world for many decades on motors.Ferrari has thought for a long time that separating the cylinders into pairs and adding a bridge between plenums as the superior design. *little fact*Back in 86' they used a plenum that was ahead of it's time for a number of design reasons, and looked to tap into the different harmonics. It was a good design for getting each cylinder the same amount of air.

Since you guys are getting into the weeds here these were the two questions I had that would affect the result.

1. What happens when you use the crappy CA 91 oct gas?
2. What total timing was added given the use of the EMS?

Sorry if dmxcrx already answered this. Good work here!

Depends on what octane he used. Using e85 will definitely give him more power as he will be able to advance the timing more. As of now he is gaining a little over 91, but nothing as substantial as going from 89 to 93.

I'm curious as well on how much timing he was able to use.

Dynojets are pretty reliable and consistent tools. In the BMW world, in NC I can dyno a stock E9X M3 and one with a tune and test pipes and make within 5-10hp as a car with the same mods in CA with identical curves. The SAE correction factor helps.

His stock manifold pull does not look like the VTEC engagement point was changed and its a bit hard to tell for his dsmcrx Manifold pull. Changing the VTEC point just changes when you get on the (slightly) larger cam profile, so at say 6,500rpm the power level really wouldn't be changed in a lowered engagement point vs. stock because you're on the same cam at that rpm in both examples. The NSX's VTEC is a relatively simple crossover and not a whole lot changes compared to an iVTEC engine.

An optimized tune can improve parts of the graph but it probably won't do much to the general shape which is dictated by the VE of the motor. There aren't very many back-to-back header comparisons out there and most aftermarket ones don't make much more than the stock NA2s or change the graph a ton. They all have a very similar shape.

Due to all of this, I think it's a fair comparison and I don't think an optimized tune will get the ported stock manifold to the same mid-low end as stock or dsmcrx's inake, nor the top end breath ability.

You can start a different thread if you want to go off on a tangent, I don't want to hijack dsmcrx's thread.

I won't hijack it either. I will just say that VTEC engines have had iVTEC added to them before, and moving the VTEC point of engagement and adding headers can give you an iVTEC type of gain on a graph. Having true iVTEC is just a lot more adjustable. Even more affective for turbo cars than NA ones.

 

[dmscrx]

Thank you. Do you have AITs?

I'm getting a 7whp difference going from STD (276.12whp) to SAE (269.08whp) and I've always liked to look at SAE since it makes different runs a bit more comparable than STD.


Overlaying the images isn't as clear and took more time than just hitting 'export', but here's my analysis:


Stock Manifold: (Red line)
-Custom DMS direct intake piping with DF scoop
-Custom DMS Upper turbo manifolds & Route KS titanium exhaust
-AEM Infinity
-Stock 65mm TB

dsmcrx Manifold: (Blue line)
-Custom DMS direct intake piping with DF scoop
-Custom DMS Upper turbo manifolds & Route KS titanium exhaust
-AEM Infinity
-BLOX 74mm TB

Ported Stock Manifold: (Green line)
-Stock piping with DF scoop
-TopSpeed Headers & small muffler (look at MotoIQ article)
-Stock ECU
-Stock 65mm TB

Conditions:

Stock Manifold: 61*F with 36% humidity.
dsmcrx Manifold: 55*F and 26% humidity.
Ported Stock Manifold (MotoIQ Article): 85.56*F 13% Humidity

-The dsmcrx Manifold had slightly more preferable temperatures and humidity, but they are close. The Ported Stock Manifold pull had much less ideal temperature but had better humidity conditions.

Power @4,000rpm:

Ported Stock Manifold: 138.26whp, 181.28tq
Stock Manifold: ~148whp, ~193tq
dsmcrx Manifold: ~150whp, ~195tq

-The lack of the VVIS plates in the Ported Stock Manifold greatly hurt the low end torque of the engine below 5,250rpm. The larger TB and dsmcrx manifold equaled or bettered the power and torque up to 5,250rpm.

Power @5750 rpm:

Stock Manifold: 211.95whp, 193.52tq
Ported Stock: 224.64whp, 205.22tq
dmscrx Manifold: 228.19whp, 208.35tq

-The Ported Stock and dsmcrx manifolds really pull away from the stock intake manifold above 5,250rpmand overlay very similarly with the dsmcrx manifold slightly having more power between 5,600-6,300rpm but the stock ported manifold has a pronounced strength from 6,300-7,000rpm. They then overlay from 7,000-7,500rpm.

Power @ 7500rpm:

Stock Manifold: 258.93whp, 181.26tq
Ported Stock Manifold: 274.70whp, 192.19tq
dsmcrx Manifold: 277.60whp, 194.32tq

-Both the Stock Ported and dsmcrx manifolds have noticeably more power than stock at higher engine speeds. If you look at the general shape of the ported manifold vs. stock, they have the same dips and shape from 5,500-8,000rpm with the ported manifold having ~10-12hp more than stock during this range.

Power @7,900rpm:

Stock Manifold: 247.29whp, 163.64tq
Ported Stock Manifold: 268.90whp, 178.65tq
dsmcrx Manifold: 281.31whp, 186.14tq

-At high rpm is where the dsmcrx really shines. After 7,500rpm where the stock TB and intake (even the ported one) start to drop, the dsmcrx manifold takes off and keeps making more power until ~8,200pm I wonder if a larger TB would keep this power increase going all the way to 8,500rpm.


Overall, it's impressive that the 74mm TB and dsmcrx manifold match and better the stock VVIS system at low rpm. This is huge for a daily driver since the Ported Stock manifold would feel very sluggish for daily driving rpm where you're not revving the motor out all the time. For the most part the dsmcrx and ported stock manifolds are similar between 5,250-7,500rpm with each having a slightly stronger band, however the combination of low end power and very high RPM power, the dsmcrx manifold & 74mm TB would make for a great well-rounded and more powerful gain for an NA car, and even more so for those who are going to rev their motors out and increase the redline.

I would be eager to see both a 65mm and 76-80mm TB pulls to see how they affect the top end breathing, ut either way the 74mm throttle body should make for a very good performance gain.


At ~$2,400, will the fuel rails also come with the manifold and TB?

I also wanted to note I was hitting 8500RPM when I hit the rev limit on that pull. For some reason the dyno is off by roughly 200rpm. I always get my signal directly from the coil so I don't know where the latency is coming from; could be the really long pickup wire for the dyno. I appreciate the kind words regarding the manifold as real world results are my motto. I did compare again the difference from STD and SAE on my graph and it is still -4hp. Every dyno stack is calibrated to the location's elevation and other factors can attribute to this but oh well. The price yes is a complete manifold package with Throttle body, fuel rails, IACV remote plate. For the DBW guys it will be the same but with an adapter plate to accommodate the OEM DBW throttle body. For users who wish to change to a bigger DBW TB I can make custom adapters in house. I am just tyring to make it easier on the manufacturer to mass produce these with little alterations to make things go smoothly and to keep costs down. They are going to send me the retail version of mine that was copied and I am going to make sure everything is to my liking including another dyno session to make sure the numbers are on par. This version will be with the 76MM throttle body. Once I am happy with everything I will give them the "OK" and start a group buy.

I am pretty sure if I was running a larger exhaust I would have picked up a little more. At 2inches, there is definitely room to grow on that front. My exhaust manifolds have a 2.25 inch outlet (for turbo) and I just made 2 inch downpipes in to the 2 inch exhaust. The outlets are 2.5 inch with 4 inch tips. The exhaust is doing some funky stuff before exit so I'm sure with a proper header and exhaust a bigger difference could be seen. Nevertheless, being naturally aspirated is only temporary for me


AITs at the beginning of the run with the engine at operating temperature for about 15 minutes were 38C (about 100F) on the Infinity with surface temp being 110degrees F. By the end of the dyno pull AITs dropped down to 29C (about 84F) and the surface temp dropped down to 101F.

- - - Updated - - -

No, not at all. I understand and agree that an aftermarket ECU can improve the area under the curve and the peak power of the Ported Stock Manifold run. The extent isn't known until tested but its still a fair comparison because a optimized tune for the Ported Stock Manifold would yield a somewhat favorable power graph and would most certainly not be less.

But then you need to ask dmscrx if he optimized both of his tunes for the stock and aftermarket manifolds. If he didn't then it's even more comparable if both tunes were optimized for the stock manifold.

Yes, I tuned both setups with the Infinity. Trying to change the VTEC point with the stock manifold was fruitless from the stock setting. I think it may have varied only by 100 rpm. The power went one way or the other and just didn't like it. I always try to get the most ignition advance safely from a setup, but again my stock manifold setup just didn't like more timing. Made no power difference. I'm currently making peak power numbers with less than stock ignition advance which tells you something else about the efficiency of the setup. Some some info to chew on

- - - Updated - - -

That shape is as old as dirt and used across the world for many decades on motors.Ferrari has thought for a long time that separating the cylinders into pairs and adding a bridge between plenums as the superior design. *little fact*Back in 86' they used a plenum that was ahead of it's time for a number of design reasons, and looked to tap into the different harmonics. It was a good design for getting each cylinder the same amount of air.

Depends on what octane he used. Using e85 will definitely give him more power as he will be able to advance the timing more. As of now he is gaining a little over 91, but nothing as substantial as going from 89 to 93.

I'm curious as well on how much timing he was able to use.



I won't hijack it either. I will just say that VTEC engines have had iVTEC added to them before, and moving the VTEC point of engagement and adding headers can give you an iVTEC type of gain on a graph. Having true iVTEC is just a lot more adjustable. Even more affective for turbo cars than NA ones.

I am using 93 octane. Considering the lack of ignition advance I was using on my stock manifold, I don't think 91 will make much of a difference compared to 93. I do think however that this manifold will make better use of lower octane considering my ignition advance is lower now than it was with my stock manifold (except at high rpm where it needs it). I am going to be posting results here soon on what it will do when I convert my car to flex fuel with the Infinity. I will get the pump E85 here and nothing laboratory grade. I know it will make more, but I'll have to wait and see how much more

 

[jones83]

Dmscrx, what is your duty cycle on your injectors? I figured you moved the VTEC point and I still believe Billy would as well if he wasn't on a stock ecu.

How come you didn't add more timing advance? Knock was rising? Saw no gains so you pulled it back?

What diameter is your intake piping? I've seen that affect power gains more than TB size also. I would not expect anything less than 3.5" on this setup. More ideal being 3.875"OD to 4"OD.

 

[stuntman]

FYI my exhaust has a 2.5" collector that's necked down to 2.25" after the flex joint. We are doing the same with temporary NA tests.

How much for just the larger fuel rails? Will they work with the stock manifold and what size tube/threads is it? (-6, -8?)

I wanted your file because if you lost less from STD to SAE, it would have shown a bigger gain over my Ported Manifold....

 

[RYU]

I would like to voice the same kudos Billy mentioned. It's a great accomplishment yielding gains with this IM in both the high and low end curve. IM solutions for the NSX thus far tends to sacrifice one or the other.

If this was a perfect world, dmxcrx would have a set of ITBs to test as well.

 

[jones83]

A perfect world for NA is ITBS with variable runner length.

 

[dmscrx]

Dmscrx, what is your duty cycle on your injectors? I figured you moved the VTEC point and I still believe Billy would as well if he wasn't on a stock ecu.

How come you didn't add more timing advance? Knock was rising? Saw no gains so you pulled it back?

What diameter is your intake piping? I've seen that affect power gains more than TB size also. I would not expect anything less than 3.5" on this setup. More ideal being 3.875"OD to 4"OD.

I only moved the VTEC point after the manifold upgrade because I was able to get more power out of the primary lobes up to a little bit higher rpm. There is no dip/jump in the power during the crossover because I optimized the transition point. Wasn't able to do the same on the stock manifold because of its flow characteristics. I pulled timing back not because I saw knock, rather because there was no improvement. You should always run a little less than more. My intake piping is 3 inch from the scoop to the throttle body. Transitioning from a 3.5 inch at the scoop. There is no point making the intake pipe bigger than the TB at this point. Unless the engine is trying to flow more air than the 3 inch pipe can handle, the airflow will stall inside of the bigger pipe until the car reaches a decent MPH and the scoop starts working. Perhaps a 3.5 inch may suffice with a slightly bigger TB. I may play with that a little later.

 

[jones83]

At what rpm did you move the engagement point to?

I have seen many people say that before, and real world testing prove them wrong. I have seen 3.5"(no transition) work over 3"(no transition) intakes on 68mm, 70mm, and 74mm before. All motors were between 2.0L-2.6L.

Also, the transition in your intake is severely important to whether the intake makes positive gains or not.

 

[Rope]

Which version of the AEM Infinity is used/ needed?

 

[dmscrx]

At what rpm did you move the engagement point to?

I have seen many people say that before, and real world testing prove them wrong. I have seen 3.5"(no transition) work over 3"(no transition) intakes on 68mm, 70mm, and 74mm before. All motors were between 2.0L-2.6L.

Also, the transition in your intake is severely important to whether the intake makes positive gains or not.

I have proven myself right in this regard on the NSX engine. I developed these intakes for the NSX with two different variations of pipe diamteters and found the 3 inch to work the best; this was on the stock intake manifold though. The bigger pipes did net a few hp more up top but lost low end to the point where I didn't like it, most likely due to the VVIS. However, given the new intake manifold going to a bigger TB and bigger intake pipe could prove to be useful. I will need to do that testing when I have finished everything else. My VTEC engagement is set at 5300 RPM.

- - - Updated - - -

Which version of the AEM Infinity is used/ needed?

I am using the Infinity 6 on mine. It is all that is needed for my current/future setup.

- - - Updated - - -

A perfect world for NA is ITBS with variable runner length.

That looks yummy

 

[jones83]

Oh I was talking about with the new plenum.

I honestly have no idea how the VVIS system messes with intake gains on an NSX, but I could definitely see sizing being minimized due to the butterflys. It makes sense, and I am glad you have real world data on it.

 

[dmscrx]

Oh I was talking about with the new plenum.

I honestly have no idea how the VVIS system messes with intake gains on an NSX, but I could definitely see sizing being minimized due to the butterflys. It makes sense, and I am glad you have real world data on it.

Yes sir.. And I didn't change the intake with the new manifold because I had already made the same thing for my previous setup and was trying to keep the variables to a minimum. I just had to alter the bend a little to cope with the taller manifold and add roughly 4 inches to the length. Didn't take long.

 

[stuntman]

How much for just the larger fuel rails? Will they work with the stock manifold and what size tube/threads is it? (-6, -8?)

 

[BATMANs]

That shape is as old as dirt and used across the world for many decades on motors.Ferrari has thought for a long time that separating the cylinders into pairs and adding a bridge between plenums as the superior design. *little fact*Back in 86' they used a plenum that was ahead of it's time for a number of design reasons, and looked to tap into the different harmonics. It was a good design for getting each cylinder the same amount of air.

.

The OP design has sharper angles that I'm concern if it will create vortex from the manifold to the runner if you follow me.

That one you've illustrated is more smooth.

Also a D shape port has better flow than a circular shape port - according to a NASA engineer that I've met over a decade ago.

 

[jwmelvin]

Also a D shape port has better flow than a circular shape port - according to a NASA engineer that I've met over a decade ago.

My understanding is that:

One can expand the flow from a noncircular (e.g. elliptical) section harder (less loss). But it's not clear what port you are implying in this manifold would/should be noncircular. The expansion in an intake with a scoop generally happens before the filter, to manage the pressure drop across the filter.

As I believe greenberet has pointed out in the past, NA intake performance is generally dominated by acoustics within the intake. And FI performance is more about avoiding loss, as you allude to. I suppose that your comments make sense in light of your FI setup. And perhaps the loss characteristics of this intake will become more important for a FI setup.

 

[jones83]

The OP manifold does have edges that aren't smooth, but that is part of the price you pay to keep the price down. The edges aren't detrimental to losses all across the curve. They would be minimal unless I saw otherwise. I've seen square manifolds produce on different cylinder engines before. Especially when a turbo is introduced into the equation.

The cost of something like this to remove the edges of the manifold would not be worth the minimal gains you received.

The other thing to remember is that your fears and theories mean nothing because of his real world data. There are books that contradict many techniques and designs in the racing industry, but real world test will always be the absolute. His manifold is showing good gains across the board. That can not be disputed. Definitely needs more testing however.

It is from my understanding that oval ports flow better than circular ports but they give up velocity unlike a circular port. I believe nascar runs oval exhaust for instance over circular. Circular pipe is only really made due to it's strength, not flow rate.

 

[rj45]

I remember at my old work it was 10k to do a cast for something like this, then pretty much nothing after that point. Although I'm not sure the metal used by the cast is strong enough and this piece looks more complex that what we were doing. If enough people ordered ??

 

[stuntman]

10K cast. 10 orders =1K/each.

NASCAR exhaust designs are heavily influenced by ride height and under car aerodynamics.