Airflow discussion

[shnazzle]

A rather interesting discussion started on the 2zz Facebook group about MAF adapters, airflow, meshes, venturi etc.
Thought it'd be good to get it in here as well as there are a fair few engineers.

In short the discussion started about an EliseParts MAF adapter used for 2zz conversions. Then it went into whether the MAF would read properly and not cause throttle issues on tip-in and close, because of the lack of vanes and/or mesh.

Obviously the go-to statement in the "tuning" world is to get rid of the mesh because it blocks airflow!
And yes, it does. No arguing that. But at what point? And at what point does it actually start to choke your engine?

Black/white; yes, anything you put in the path of flow will (at a certain point) become a hindrance to the maximum amount of air you can push through. This is how the whole "remove the MAF mesh!" trend was sold. And it's not a false statement. But we have a MAF to think about
Dig back to your high school physics in fluid dynamics and it all makes sense.
A venturi is great as long as you can follow through. A venturi opening up into a wider space causes massive amounts of turbulence (see airflow behind cars). So, that's why you see some intakes with a venturi at the beginning and then a long bit of uniform pipe all the way to the end. The venturi becomes a lot more effective at higher amount of flow (think venturi tunnels on cars and how they only work at speed).

All meshes are not made equal. A square mesh is very inefficient for laminar flow. Needing tons of tiny holes to be effective but that leaves you with a massive blockage. An offset honeycomb is very effective and can leave you with over 75% of available area (see pro-made aftermarket meshes). So you lose 20-25% but the airflow is straight as an arrow.  Great for MAF and great for air entering a wide open throttle.

So if you want a short massive air intake with no restrictions because the VE of your car is mahoooosive, you're going to have to slap on that 3-4" pipe with a big cone and go speed-density with a MAP sensor.
Or a massive pipe with a MAF and ways to straighten the airflow (see Supra intake) and accept loss.

On the 2zz (which does flow a feckload to be fair) the EliseParts adapter will do just fine given that there's at least a decent amount of pipework before and after the MAF to allow air to straighten up. Which is why nobody sees issues with them. The rest will be compensated for by fueling adjustments. Sub-optimal, but fine.

The 2zz intake runners are also purposely long after the throttle body to straighten up airflow before going into the cylinders. Have a look, there's about 6" of runner before it splits off. The magical 6" rule of thumb for straightening airflow



Come on engineers... Discuss

[Topdownman]

Sadly all of that airflow has gone over my head

[1979scotte]

Fluid dynamics at school nope don't remember that

[shnazzle]

So.... It seems I read too much and need to get a life...

[Carolyn]

Ah.  Fluid dynamics.  Done a bit of that.

The length required to properly smooth flow is directly proportionate to the area of cross section.  It took 200 mm (8") to smooth liquid N2O through the injectors in my rocket motors.  Cross section? Well diameter was 8- 10 mm (78.5 mm2).

Not that it matters, 'cos the intake is best left stock!

[shnazzle]

So it seems then that the golden rule of 6" of straight pipe before a MAF is a bit hopeful for straightening the flow on a 63mm ID air pipe

[Carolyn]

Ay lad, but it's better 'n nowt.

[Dev]

 Maybe I can contribute something from my own testing based on what was discovered on the Celica board many years past.

  Lets start with a little history. 
  Aftermarket intakes for the 2ZZ Celica was causing some to experience high fuel trims, bogging when cold and in some cases a check engine light for lean conditions.  Someone by the screen name Bluebomer  discovered that hacking up the stock intake, removing a short section of tube that has the vanes and adding it before the MAF they were able to resolve all of the issues, and the drivability of the car was improved.
Many on that site were hacking up their old intakes and performing the modification with success.

A few years later someone discovered that the Subaru WRX shares a similar Denso MAF sensor and the same intake diameter.  An aftermarket company named Cobb made intakes for the WRX where they incorporated vanes as part of its design which was also flow tested so the MAF reads correctly for the amount of air flow entering the engine. Someone adapted the intake for the 2ZZ and it worked as well as the hacked up modification but the advantage with the Cobb was a less resistive vane and a velocity stack design.

The idea behind the vanes and the honeycomb straighteners are different. Prior to 2000 most car manufactures used MAF sensor wires that ran near the full length of the tube and the honeycomb was used to reduce turbulence for a laminar air flow.  After 2000 most MAF sensor wires were incorporated in the body of the MAF. In order to make this configuration work effectively they used the vanes in the intake to speed up the velocity of air by simulating a smaller tube. This would effectively cool down the heated MAF sensor wires effectively so the MAF sensor would report the correct amount of metered air to the ECU. The ECU would then know how much fueling and timing to advance or subtract. I have a feeling the reason why this was done rather than having no vanes is to make it possible for the Denso MAF sensor to be adaptable to many cars with different intake diameters but that is just a guess. 

  If you don't have the vanes in the intake what happens is the MAF wires will not cool down enough for  the amount of air flow and as a result will underreport. The 02 sensors will register this as lean condition and the ECU will then add fuel to the mixture to offset the deficit.  This can actually have a negative effect on how the ECU is optimized for performance and drivability at part throttle.

When I switched out my previous aftermarket intake for the Cobb intake with the vanes it made all the difference especially at part throttle. The cars fueling is well within specifications only adding a little fuel trim probably due to my less restrictive intake and exhaust but its a genuine reading that is taking advantage of the modifications. 

[shnazzle]

So basically it seems the vanes are put in to compensate for the lack of cooling effect the airflow in the stock tube has on the chosen MAF.
Which is odd as the MAF is always touted as the better fueling strategy for low load, i.e. low airflow.
So in effect it's fluffing up the airflow to work within specs of the MAF.

I asked this a while ago somewhere on here but it'd be very handy for the Reference section to have a table of the stock system MAF readings across a rev range in a certain gear. There will be differences between cars but all in all it'd be a good way to graph an aftermarket intake against the stock intake to see how far off it is. Then do the necessary to get it back up to spec

(yes, I know Carolyn, just use the OEM system... That's not the point)

[Dev]

So basically it seems the vanes are put in to compensate for the lack of cooling effect the airflow in the stock tube has on the chosen MAF.
Which is odd as the MAF is always touted as the better fueling strategy for low load, i.e. low airflow.
So in effect it's fluffing up the airflow to work within specs of the MAF.

I asked this a while ago somewhere on here but it'd be very handy for the Reference section to have a table of the stock system MAF readings across a rev range in a certain gear. There will be differences between cars but all in all it'd be a good way to graph an aftermarket intake against the stock intake to see how far off it is. Then do the necessary to get it back up to spec

(yes, I know Carolyn, just use the OEM system... That's not the point)

I think it was put there as a way to work in tandem with the MAF design. Probably to lessen the foot print of the sensor so its not in its own way to optimize air flow and get a precise reading for the design of the MAF to be reliable.   
  As to your solution its an easy fix if you run a stand alone or piggy back ECU and have the car dyno tuned. When you re tune the car it will not matter if you have vanes or no vanes as the end result from the turning will be what it is.

So the solution comes down to two options.
1. Incorporate the stock intake or buy the Cobb intake.
2. Get a stand alone ECU and tune for the car as a whole.

There was someone that printed the vanes based on my measurements I gave him and it did bring the fuel trims closer to 0 but I don't thing that is the way to do it because it needs to be flow tested on a bench using instruments so it can be calibrated.

[shnazzle]

That makes sense. Smaller sensor blockage but then compensate by adding vanes.
See, it all falls into place once you get the right people involved! As usual Dev you're a wealth of knowledge.

Interesting that the Cobb MAF adapter never made it to the mainstream whereas the vaneless EliseParts adapter is the go-to for swaps.

What I do find a bit odd for me personally is that when I had my DIY short cone intake (now sold to someone else on here), it used the stock MAF adapter with vanes and all, but the readings were well out of whack.
I'd max out on about 75g/s on a full 3rd gear pull. Whereas the Hurricane goes over 100 and the stock just short of about 98. I posted those results a while back (before I got the Hurricane) and the conclusion was that stock was easily best overall.

So the vaneless Hurricane either over-reads or my setup genuinely has increased the VE of the engine by a few cfm. But I've never seen any associated increase in fueling to suggest over-reading. In fact, much the opposite. I've had to bump up the MAF voltage on my emanage to get it to stop underfueling.
On a side note, I think the MAF adapter used on the 1zz Hurricane kit is actually the same EliseParts MAF adapter... It must be. Its from the same shop.

[Dev]

That makes sense. Smaller sensor blockage but then compensate by adding vanes.
See, it all falls into place once you get the right people involved! As usual Dev you're a wealth of knowledge.

Interesting that the Cobb MAF adapter never made it to the mainstream whereas the vaneless EliseParts adapter is the go-to for swaps.

What I do find a bit odd for me personally is that when I had my DIY short cone intake (now sold to someone else on here), it used the stock MAF adapter with vanes and all, but the readings were well out of whack.
I'd max out on about 75g/s on a full 3rd gear pull. Whereas the Hurricane goes over 100 and the stock just short of about 98. I posted those results a while back (before I got the Hurricane) and the conclusion was that stock was easily best overall.

So the vaneless Hurricane either over-reads or my setup genuinely has increased the VE of the engine by a few cfm. But I've never seen any associated increase in fueling to suggest over-reading. In fact, much the opposite. I've had to bump up the MAF voltage on my emanage to get it to stop underfueling.
On a side note, I think the MAF adapter used on the 1zz Hurricane kit is actually the same EliseParts MAF adapter... It must be. Its from the same shop.

You get so called self proclaimed experts or those that throw around their degree to seem like they are an authority with overcomplicating tech when the logic is very simple. Be weary of complicated explanations, the best explanations are those that make sense and are simple to understand. I had some push back from this and after contacting the engineers at Cobb it became obvious once they told me they flow test as part of the design of the intake so its calibrated to the MAF, not for turbulence reduction.

You might find this interesting. In my opinion this gets the job done but its counterintuitive. Using vanes like the OEM is the best option.
https://www.injen.com/mr-technology/?SID=hagnpg8j4a715mfude9vs9gaj7

[delhusband]

when I had my DIY short cone intake (now sold to someone else on here, it used the stock MAF adapter with vanes and all, but the readings were well out of whack.
I'd max out on about 75g/s on a full 3rd gear pull.

Oh now you tell me  >>:| :protest:

[Dev]

What I do find a bit odd for me personally is that when I had my DIY short cone intake (now sold to someone else on here), it used the stock MAF adapter with vanes and all, but the readings were well out of whack.

If you try to adapt vanes to your intake  you never know what you are going to get because it could be imprecise compared to bench flow testing and computer molding  like the Cobb. There is a possibility that the design is calibrated with varying air flow from idle to WOT and everything in between. I really don't know what goes into it but all im concerned with is keeping the ECU happy the way it was designed to run.

[shnazzle]

when I had my DIY short cone intake (now sold to someone else on here, it used the stock MAF adapter with vanes and all, but the readings were well out of whack.
I'd max out on about 75g/s on a full 3rd gear pull.

Oh now you tell me  >>:| :protest:

Have you had any issues Del?
Remember, my car has the piggyback and bored throttle body as well as an owner obsessed with fuel trims. "out of whack" to me is +/-5

[delhusband]

when I had my DIY short cone intake (now sold to someone else on here, it used the stock MAF adapter with vanes and all, but the readings were well out of whack.
I'd max out on about 75g/s on a full 3rd gear pull.

Oh now you tell me  >>:| :protest:

Have you had any issues Del?
Remember, my car has the piggyback and bored throttle body as well as an owner obsessed with fuel trims. "out of whack" to me is +/-5

I jest. It didn't hinder me in shaming an S3 a couple of weeks ago, and I like the noise too much now. I'd never have noticed, I'm barely capable of bringing the trend up on torque   

[mikek]

One word MAP

[shnazzle]

One word MAP

MAP has its shortcomings at idle and low load. And the small detail that it has to be attached to a rather expensive standalone

Other than that. Yup. With you on that

[Gaz mr-s]

More reading.....

http://planetsoarer.com/resonator/resonator.html