Why metallic intakes do not work

[IceFyre13th]

I could call this post "the $1000 intake", that is how much I have spent to find out the below.

First intake I bought from a good vendor here seemed to be the way to go. Most people at the time said it worked great, the vendor said it would not pop the dreaded CEL, and it looked like the best design of the intakes available.

But, it did pop the P0172 CEL (rich condition) at least once a week, even with UniChips tune for it. Using the Torque App the MAF reading would jump around 50 CFM....the LTFT and STFT readings would hover around -15% to -20% while driving, at idle they would go to -30%. the CEL would only happen at a stop. This is when the STFT and LTFT readings were the worse. It also would heat-soak badly, at a stop intake temps went as high as 160 F. While moving it would never get less than 10 F over outside temperature. (note: while accelerating or coming off throatle the LTFT and STFT will read higher or lower depending upon what the engine needs in way of fuel at the time).

My first thought was the heat issue. The intake was ceramic coated, also I tried intake wrap, a reflective insulating wrap claimed to reduce temps by 30% or more. Stopped temps did drop, moving temps would start out lower, but after a few miles the 10 F over outside air would be the norm.

I thought the intake design was sound, but the MAF sensor tube diameter was smaller than the factory intakes, not by much......but it was too small. Heat did not seem to be the problem as the wrap did not stop the CEL from happening.

I modeled the vendors intake in SolidWorks 2013 (I am a CAD engineer) and ran SolidWorks Flow Simulation. What I found was the location of the MAF sensor and the small diameter was effecting the readings of the MAF sensor. Due to the bends and turbulence in the tubing there was no way the MAF sensor could work right.

So the next step was to take the model and improve the design. I increased the diameter of the tubing, reduced the radius of some of the bends, and added a flow straightener in front of the MAF sensor. Ran Flow Simulations, made changes based upon the data, then had this new intake built. The new intake was coated with Jet Hot, a better coating than the original one had and thicker.

It was better, heat was still a problem, and I had to have the UniChip custom programmed to work with it, LTFT and STFT now at 10% to -10% driving, STFT at a stop would go to 12% and LTFT would be at -17%.....but. MAF was steady as well.

After a few months the CEL would start showing up, not too many times at first, but got more frequent as time went by. The MAF now was swinging 100 CFM too........something must of gone bad, first thing I tried was to remove the UniChip, but no it ran worse........so back on it went.

By now I wished I kept the factory air box, but it was donated to a good cause, and I decided to go get the TRD intake. What a surprise, this intake seems to make more power at the low end. The MAF is rock steady, STFT and LTFT are between 5% to -5% running or stopped. Gas mileage has gone from 31.7 MPG to 35.5 MPG average. Its only been on a week, and I hope everything stays this way...........

Now for the reason the metal intakes failed. Both used a support arm that bolted the MAF tube to the radiator support. Look at the MAF sensor, the areas where the MAF gets bolted to the tube are metal. The metal tabs are electrically connected to one of the five pins the MAF sensor connector has.............both the vendors intake and my intake were coated, after a while the bolt that supports the MAF tube and under the support arm rub though the coating and the paint on the radiator support.

This now means the MAF sensor is electrically connected to ground, something the plastic of the factory air box or TRD intake will not allow to happen..........and why I think that metallic intakes do not work. I see no easy way to insure the MAF sensors mounting tabs can be isolated on metal intakes. The only solution I can think of is to make the MAF sensor mount tube out of plastic.......this would be expensive due to the plastic mold that would have to be made.

I also think this is why most people are saying the factory intake with a drop in (or the TRD intake) are the best solution, not because the other intake designs are bad, but due to the MAF sensor being connected to an unintended ground.

Finding this out I cannot say the original intake I bought is bad, and why I am keeping the vendors name out of the discussion.

I will recommend to all makers of metal intakes to find ways to insure the MAF sensor is isolated from this unintentional ground.

[wootwoot]

Wow... Nice work!

[FR-S Matt]

My money is the first intake was an Injen. Did I win?

[Cjymiller]

sub'd for interest.

[Spartan65]

Originally Posted by FR-S Matt

My money is the first intake was an Injen. Did I win?

I think you're probably right, they changed the ID on the injen and the MAF location distance if I am thinking straight.

[wootwoot]

Originally Posted by FR-S Matt

My money is the first intake was an Injen. Did I win?

I am almost positive it wasn't.... Who offers an optional ceramic coating on their intake... Hmmm....

In any event it doesn't matter. And this is a very interesting insight!

[IceFyre13th]

Originally Posted by wootwoot

Wow... Nice work!

Thanks, wish I could afford the molds to make a plastic tube version of the first intakes.......but I think they will run close to $100K due to the complexity of the design required. It would have to be a rotational mold to make the full tubing version, machine the MAF mount location as a secondary operation.

It might be possible to have a short straight section made that has the MAF mount and flow straightener built in, then the rest could be bent out of aluminum tube.....one or two extra silicon couplings would be needed. That would bring the mold cost down to $30K to $50K...

Or be machined out of plastic stock (probably cost $200 - $500 each part if you made only a few, less if you had a higher mass production run).

I hate being an engineer sometimes, I can see the solutions but cant afford to produce the finished parts.

[FR-S Matt]

I almost wish we had a vendor or someone with access to all of these intakes to temp test them to see how they truly perform whether its metal or roto mold. I'm sure we'd see some interesting results. I'd like to see full CAI vs SRI.

[Chee-Hu]

Originally Posted by FR-S Matt

My money is the first intake was an Injen. Did I win?

Who cares, really? The point of this thread wasn't to bash the vendor. It's to point out the mistake.

[IceFyre13th]

Originally Posted by wootwoot

I am almost positive it wasn't.... Who offers an optional ceramic coating on their intake... Hmmm....

In any event it doesn't matter. And this is a very interesting insight!

I only said it was ceramic coated, I never said it came that way...

Like I said earlier, the original design is good...no use guessing or saying who they are...its not the design, but the possible unintentional ground that may be the issue.

[FR-S Matt]

Originally Posted by Chee-Hu

Who cares, really? The point of this thread wasn't to bash the vendor. It's to point out the mistake.

I care and I care about my car not having a shitty intake throwing CEL's, don't you?

[Chee-Hu]

Originally Posted by FR-S Matt

I care and I care about my car not having a shitty intake throwing CEL's, don't you?

It's a pretty simple solution, sir. Don't purchase metallic intakes?

[FR-S Matt]

Metallic doesn't have anything to do with it really. It's the MAF location and the flow tricking the MAF into making it run more rich. He even said it in his first post.

I modeled the vendors intake in SolidWorks 2013 (I am a CAD engineer) and ran SolidWorks Flow Simulation. What I found was the location of the MAF sensor and the small diameter was effecting the readings of the MAF sensor. Due to the bends and turbulence in the tubing there was no way the MAF sensor could work right.

The only intake that I know which does this and people that have gotten the CEL's and the rich codes are Injen users. The only ones that have corrected it tuned the car. Perrin, AMR, and FA20club full CAI's are metallic and have no flow issues nor do they relocate the MAF from its stock location. The only thing in question with those CAI's are temperatures which I have not seen yet.

[IceFyre13th]

@FR-S Matt You are wrong, not saying who anyway.......but please stop guessing, and please don't turn this into a Vendor bashing thread.

Yes the original one showed issues using SolidWorks Flow Simulation.......even my optimized design still had some, there is no way to make a perfect flow in a automotive environment.

I think the original and mine would work fine, but there has to be a way to isolate the MAF sensor from an unintentional ground.......that is all!!!! It has everything to do with the metal intake causing the unintentional ground.....

Tuning via UniChip or Flash Tuning is only really needed after this to get the most out of the modifications.....not to correct poor flow, but to tune the combination. Like I just said, there is no way to make a perfect flow in a automotive environment.