blow through stock maf questions

[94awdcoupe]

There is a guy in evo forums that posted his success with using evo 8 mass in blow through mode. he was making well over 600whp with it. said it was working great. spools turbo faster. engine itself is already restriction after turbo. so moving mass after turbo results in faster spool by removing restriction before turbo.

I have been contemplating trying this myself on my TS 20g setup. I am having huge drivability problems at part lift. Blow through would solve this as I could put BOV before mass.

Anybody here have any new info on this? I am huge anti fan of using GM maf conversion tables. I have never seen them work that well in the drivability areas.

I tried to do it this past weekend but couldnt find the coupler sizes needed to install mid pipe. 4/2.5 and 3.25/2.5 are needed.

You must be logged in to view this image or video.

 

[stefanOst]

That guy is going to have a fun time running the upper rad pipe

 

[bling5tatus]

What's wrong with using speed density?

Oh... and everyone knows that the maf is not an airflow restriction in front of the turbo oo.... just kidding, I had to say that because the forum local to my area is full of people that think it doesn't restrict air flow.

 

[Ceddy]

Using a stock MAF like that will only work at very low boost levels.

I've heard of Lancer turbo conversion guy using them this way, but only at 5 psi.

A Karmen Vortex MAF sensor will not work in blow thru setup with any real amount of boost. The Karmen MAFs have been around a least 20 years, people have tried blow thru and it doesn't work. I would have to see it working to believe it.

Also the EVO guys have problems sucking honeycombs in with a draw thru setup, I would hate to see what blow thru does to them.


Maybe someone can explain the physics of a Karmen system.

 

[need2boostpsi]

seems to me like it would just blow apart at high boost, much like plastic intake manifolds. I would love to see this done as i agree with your reasoning for removing a draw through type maf.

 

[tkelly27]

I would imagine that the vastly changing density of the air would make that difficult to tune, if it was tunable at all. By the time you're done getting something to tune with, building a custom enclosure for it, and figuring out how to tune it, you could have just gotten a maf translator, stand alone, or a native speed density system like Jackal for 1g's or link/evo ecu flashing for 2g's.

Kármán vortex street - Wikipedia, the free encyclopedia
http://upload.wikimedia.org/math/7/c/2/7c21167031c3738f95728facf4d459d7.png
You can see the frequencies of the vortecies is based on the Reynolds number, which in draw through is pretty much only based on the velocity of the fluid (with a known cross sectional area this gives you volumetric flow rate). When you get substantially different pressures the kinematic viscosity also changes
Viscosity - Wikipedia, the free encyclopedia
due to a change in density.
http://upload.wikimedia.org/math/d/f/4/df4ecd6ca6a4c3237ade69b9655c58ee.png
I imagine this is compensated for using the baro sensor in the draw through location, but I don't think the baro can read much positive pressure, nor do I think the ECU can compensate for what is going on at those higher pressures.

I think I got that right... looks like there might be another velocity thrown in there, that doesn't change the fact that changing the pressure changes how the sensor reads.

Even if that guy got it to work, it's a bad idea. 600hp is nothing to crow about with an EVO maf, they flow enough to make that in draw through. Even the 2g maf is good out to 500hp in draw through.

 

[94awdcoupe]

I am not sold on the hearsay that it wont work. There are numerous posts being spread that it wont work. These are not coming from people who have tried it. From my understanding of how the meter reads airflow there is no reason it wouldnt work in the IC pipe. The baro sensor reading would be incorrect. You will likely need to keep that part of sensor pre turbo. To say the meter wont read karmann because the air is compressed is kinda silly. That would be like saying the meter reads at sea level but at 5000 ft it doesnt work. Just look at the formula for vortex. The V is for viscosity of fluid. The fluid in this case is air. Thicker air (compressed) will have higher viscosity. Thus it will read more vortexes for the slower moving air. The body looks like it will handle 40psi.

If someone has actual testing data of why it wont work I would sure like to see it. I havent been able to find that myself.

I am not interested in making more power or faster spool. That is not my interest here. I like my car quiet so recirculating is a must. problem comes on part throttle lifts. The air being dumped into intake pipe causes air to change directions. Since the maf reads in both directions it gets a false reading of air going into the engine. Air is pushed backwards and the mixture goes dead rich causing sputter and stutter.very similar to having a hose brake. metered air does not make it into engine so mixture goes rich. I can watch this happening live on the AFC. the karmann counts go high as air is dumped into intake pipe.

 

[tkelly27]

Kinematic viscosity changes non-linearly with density
http://www.mathsrevision.net/gcse/1overx.gif
and density changes with pressure and has a nonlinear variation with temperature. So you're trying to tune for two nonlinear variables plus a third linear one. Relative under boost condition, these variables remain static or in a range that is reasonably linear at the inlet of the turbocharger. So it's not hearsay, it's the laws that govern the functioning of your mass air sensor. It's not so much that it won't work, but if you change any of these variables it won't be linear and it will be exceedingly difficult to tune.

Yes, flow occurs both ways through the MAS, however the bypass from the charge piping goes into the turbo creating a loop, theoretically bypassing the MAS. Flow is also not technically additive, but if you have one with a pressure differential driving it one way, and another driving it the opposite way, they should cancel out. That is, air coming into the mas, should be opposed by air coming out of the mas to net a lower flow rate.

If you are getting a rich condition when you let off the gas, that would point to a leak somewhere rather than a faulty function with the MAS. I'm not saying you're wrong, but I've never heard of this unless someone is venting. Specifically the sputter and stutter, every DSM ever made is setup with the recirculation dumping behind the mass air sensor, as well as every hot wire nissan, flapper door/karman toyota, VW, etc. None of them get a rich spike unless there is a problem with their recirculation setup.

If you haven't done a boost leak test from the inlet tube back, I'd advise it. Sometimes you can get a leak in the recirc fitting that closes itself under normal operation, but opens when the bov releases.

I used to have a page on why karman vortex mass air sensors can't be used in blow through bookmarked, but once I learned that I could use speed density I stopped caring : D Try it if you must, makes no difference to me. If it works, you've solved a problem nobody else with a mass air sensor is having.

 

[brian9397gsx]

i to would like to see how this works. i can understand how it works with a hotwire maf but from what i understand a karmenn vortex operates differently making this way not feaseable. try it if it doesnt work keep us posted but i think changing how close your bov vents to the maf would be easier or even adding a tube like the 2nd gens used stock.

 

[94awdcoupe]

Tom, the car does not have boost leaks. And the recirc setup is designed well. The return tube is about 12" and dumps close to turbo. The air intake pipe is mostly bellowed rubber to buffer the MAF from returning air. Yet the problem is still there. Especially noticable in my car as exhaust is about 1/4 inch from car. So exhaust hitting car in rapid fire mode lets me know I have a huge problem to correct. I have built some 50 setups in last ten years. I have not encountered issue like this. But I have never had a 20g spool this fast either. The TS setup plus cyclone makes for an extremely fast spooling setup.

I am just a good tune away from really having an unbelievable fun car to drive. There is no common way to tune and evo 3 ecu. Currently the AFC is all I have. Tune is actually pretty dang close. Just need access to timing maps to get it perfect.

guess its time to throw the Motec in and be done with it.

You must be logged in to view this image or video.

 

[tkelly27]

That's strange that there is still a question about AFC and mass air sensors when there is a motec sitting around your house. you're going to love it.

Starting fluid always helped me dial in the initial start with a stand alone.