difference between intakes

[locoman99]

Bernoulli Principle

[AaronGTR]

Originally Posted by MercenaryX2

@Mike, I have a question about your CAI. If I'm reading it right, the outside is rigid for flexibility but the inside is smooth like metal piping, correct?

Originally Posted by MilzyZ34

yeah, the tube has a steel helix as it's skeleton (think big thin spring), which keeps it pretty rigid, but very flexible without binding, the inside is very smooth, and corrugation is barely noticeable inside.

It doesn't matter how smooth the inside is. If it's normally straight pipe but is corrugated (however this is accomplished) and able to be flexed, then it will no longer be a perfectly smooth surface inside once it is bent. In order to achieve the bend one side has to be shorter than the other. It's simple geometry. The fact that the material has to zig zag inside to take up a shorter distance is going to create breaks in the surface, and no matter how small they are they will cause turbulence by the laminar air flow detaching from the surface at break points. This is a standard principle of fluid dynamics.

The difference may not be very much depending on the severity and the peak air speed/volume we see, and you'd have to do some pretty detail flow testing to quantify the difference... but based on known air flow principles a hard pipe that is mandrel bent into the correct shape will always have smoother flow than a flexible pipe. How much this matters is a different question though.

....What this means is that the flow through the MAF is not consistent, therefor to match the tune in the pcm, the MAF has to be clocked at a certain angle. This is probably hard to understand, so think about it this way ... ideally with the MAF in the long straight pipe, the peak of the airflow would be right in the center of the tube, but in a curved tube, this peak is not in the center, it's offset along the axis of the curve, between center and the outside edge of the tube. If you had a constant airflow through the length of the tube, and twisted the MAF, you would see the flow numbers go up and down as you twist it, eventhough the actual flow rate hasn't changed. So to avoid issues with the MAF being slightly misaligned, we leave the tube in place, which is keyed for the MAF and for the throttle body, locking their relative position.

Originally Posted by MilzyZ34

for arguments sake, lets say you are flowing exactly 50 lb/min through the curved intake tube. the maf measures flow by the vibration of the sensor inside of it, which spits it out as a frequency (Hz). The pcm then assigns lb/min value of air based off the MAF freq vs MAF flow table. When you rotate the sensor in the tube, it will only read 50 lb/min at one position, any other way you clock it, it will either read high or low.

Um, I think you are a little off on function of the MAF. That doesn't match anything I've ever read about it. Here's a quick link with an explanation similar to others I've seen. http://www.sensorland.com/HowPage060.html

The MAF does not function on vibration. It is a hot wire sensor like any other MAF, and is basically a thermistor, and it converts voltage to a digital Hz signal for the PCM. It is true that having the MAF in the straightest air flow possible is optimal, and putting it in an extreme curve could possibly affect the reading, but they also put that screen there on the front for a reason. It straightens the air flow and mitigates some of the effects of air turbulence. Putting it in front of or behind a gradual curve isn't going to affect it much, if at all. Also GM doesn't use lb/min as a unit of flow measurement in their PCM's. They've gone metric like everyone else. It's grams/second.

GM also used that MAF on a bunch of different cars with different engines and different air boxes. The MAF was probably calibrated in a lab with a piece of straight pipe, then the fueling tables were probably tweak in individual cars it was installed in. If you change the intake and change the position/orientation of the MAF, you can tune for changes (if any) with a tuner the same way.

[MilzyZ34]

Originally Posted by AaronGTR

It doesn't matter how smooth the inside is. If it's normally straight pipe but is corrugated (however this is accomplished) and able to be flexed, then it will no longer be a perfectly smooth surface inside once it is bent. In order to achieve the bend one side has to be shorter than the other. It's simple geometry. The fact that the material has to zig zag inside to take up a shorter distance is going to create breaks in the surface, and no matter how small they are they will cause turbulence by the laminar air flow detaching from the surface at break points. This is a standard principle of fluid dynamics.

The difference may not be very much depending on the severity and the peak air speed/volume we see, and you'd have to do some pretty detail flow testing to quantify the difference... but based on known air flow principles a hard pipe that is mandrel bent into the correct shape will always have smoother flow than a flexible pipe. How much this matters is a different question though.


Um, I think you are a little off on function of the MAF. That doesn't match anything I've ever read about it. Here's a quick link with an explanation similar to others I've seen. http://www.sensorland.com/HowPage060.html

The MAF does not function on vibration. It is a hot wire sensor like any other MAF, and is basically a thermistor, and it converts voltage to a digital Hz signal for the PCM. It is true that having the MAF in the straightest air flow possible is optimal, and putting it in an extreme curve could possibly affect the reading, but they also put that screen there on the front for a reason. It straightens the air flow and mitigates some of the effects of air turbulence. Putting it in front of or behind a gradual curve isn't going to affect it much, if at all. Also GM doesn't use lb/min as a unit of flow measurement in their PCM's. They've gone metric like everyone else. It's grams/second.

GM also used that MAF on a bunch of different cars with different engines and different air boxes. The MAF was probably calibrated in a lab with a piece of straight pipe, then the fueling tables were probably tweak in individual cars it was installed in. If you change the intake and change the position/orientation of the MAF, you can tune for changes (if any) with a tuner the same way.

I could be wrong about how the frequency signal is generated, but the rest of what i said still stands. I think what you're trying to say is the corrugations would cause a boundary layer to form. even if it did, this boundary layer would have to be over 1/4" thick for it to cause any loss of flow compared to the 3" pipe.

the screen is there to provide a more consistant signal, but it can't physically move the point of peak flow back to center just by having a screen there.

i agree you can tune out any difference you create by twisting the maf, but are you going to re-tune the whole MAF table every time you remove and replace your cold air intake, assuming that you don't mark exactly where it was and put it right back? All I'm saying to you guys is if you go through all the effort of doing a dynotune or other wideband tune, either always make sure the MAF is always in the exact same clocking, or put it right in the middle of a 3ft section of straight pipe so orientation doesn't matter.

in HPTuners, I'm pretty sure the MAF flow vs freq table is entered in lb/min. Since I don't directly read the serial data off the pcm, I guess I refer to the data I deal with in HPTuners. I couldn't tell you which one the pcm itself uses, my guess is both.


the MAF itself is probably calibrated to have the proper frequency output at a given flow rate in a straight piece of pipe, but i bet you the maf flow vs freq table is actually dialed in on a vehicle with a curved intake tube.

I know about fluid dynamics, it was a requirement to get my mechanical engineering degree.

[AaronGTR]

Originally Posted by MilzyZ34

I think what you're trying to say is the corrugations would cause a boundary layer to form. even if it did, this boundary layer would have to be over 1/4" thick for it to cause any loss of flow compared to the 3" pipe.

No I'm not taking about boundary layers. There is always a boundary layer with moving air. Air has viscosity and has a tendency for some of it to stick to the surface it's against just like water. The boundary layer (the laminar layer) moves slowly and allows the inner layers to move smoothly and faster. But air, like anything else with momentum, also wants to keep moving in the same direction. If you change it's direction smoothly and gradually it sticks to the wall and moves smoothly. If you change direction to drastically or with a sudden angle (or a bunch of small sudden angles like in a corrugated tube) the air wants to pull away from the surface and tumble. At a certain point you move from laminar flow to turbulent flow which isn't as good for velocity, or for MAF readings.

the screen is there to provide a more consistant signal, but it can't physically move the point of peak flow back to center just by having a screen there.

i agree you can tune out any difference you create by twisting the maf, but are you going to re-tune the whole MAF table every time you remove and replace your cold air intake, assuming that you don't mark exactly where it was and put it right back? All I'm saying to you guys is if you go through all the effort of doing a dynotune or other wideband tune, either always make sure the MAF is always in the exact same clocking, or put it right in the middle of a 3ft section of straight pipe so orientation doesn't matter.

Seriously? It's not that hard to put a MAF back in the same place. If you mod the engine and get your setup finish and have a good intake and get it tuned as is, it would be a piece of cake to do maintenance on the car and put the MAF back the same way every time. Or at least close enough that it's not going to make any difference. Also you'd hardly need a 3 ft section of pipe. Air velocity would normalize (or get darn close) after a much shorter distance than that.

Also proper design of the intake can negate some of the affect of velocity difference across a bend. As an example, my SC setup requires me to use a large bend from the TB to reach the fender. I used a 3" large radius 180, cut off at around 160 degrees to get the right angle. That connects to a 3" to 3.5" reducing coupler on the back of an LS1 MAF. In front of the MAF though is a 3" long 3.5" diameter coupler hooked to a 3.5" large radius 45 degree bend with a 3.5" opening K&N filter on the end. I used a large filter and large diameter pipe with a smaller angle to reduce the velocity and turbulence in front of the MAF. The difference in velocity between the inner and outer curves increases with a greater angle and smaller radius, and it also is affected by total velocity. Lower total velocity equals a smaller differential.

in HPTuners, I'm pretty sure the MAF flow vs freq table is entered in lb/min. Since I don't directly read the serial data off the pcm, I guess I refer to the data I deal with in HPTuners. I couldn't tell you which one the pcm itself uses, my guess is both.


the MAF itself is probably calibrated to have the proper frequency output at a given flow rate in a straight piece of pipe, but i bet you the maf flow vs freq table is actually dialed in on a vehicle with a curved intake tube.

In DHP it's in grams per second. The injector table is metric too. It's msec/gram. Charles was a GM software engineer at one point, so I'd imagine he'd use the same measurements in his tuning software that he used in his job. Pretty much every major company that sells anything internationally has gone to metric these days. HP probably converts to standard because it's what people in the U.S. are used to.

And about the MAF calibration, I'm almost positive you are wrong about that. They had a whole bunch of MAF files on the tuning forum and they were all for specific engines, ie LA1, LS1, LQ4, etc. Those engines were used in multiple cars with different intakes and the MAF file was always specified by engine, not by car. They probably did any fine tuning for specific cars in the VE file for that car.

[MilzyZ34]

dude, you're typing makes my head hurt.

air does not always move through a pipe all at the same rate, and just because air doesn't stick to the walls as it moves like you said, that doesn't make it turbulent throughout the tube. air will always find the best line through the pipe called a streamline, just like a race car driver will choose a similar path to take a turn. he starts our wide, hits the apex and then carries the speed out wide again. this is the fastest way to take the turn, and air flows the same way. the presence of this affect does not keep it from being laminar.

as for what i said about MAF tables, I meant the maf tables are designed "in a car" meaning on an engine, with exhaust, intake box, etc, which would be the opposite of in a lab on a straight piece of pipe like you said.

as for units, I'm pretty sure the pcm does in fact have both sets of units as outputs, but in the end, who cares what one company uses or another company uses. it's all the same thing, one unit is the same as another unit times a coefficient and vice versa. HP Tuners is setup for both. I prefer imperial. I'm not the only one, or it wouldn't be on there. If you like metric, great. It really makes no difference in the end.

[MercenaryX2]

One thing I know about changing anything intake related is that once you do that, your MAF scalings are thrown off. It may not be a huge difference but there is still a difference. I saw it all the time when I went through a few different intakes on my DSM.

[AaronGTR]

Originally Posted by MilzyZ34

dude, you're typing makes my head hurt.

air does not always move through a pipe all at the same rate, and just because air doesn't stick to the walls as it moves like you said, that doesn't make it turbulent throughout the tube. air will always find the best line through the pipe called a streamline, just like a race car driver will choose a similar path to take a turn. he starts our wide, hits the apex and then carries the speed out wide again. this is the fastest way to take the turn, and air flows the same way. the presence of this affect does not keep it from being laminar.

My typing makes your head hurt? Look who's talking. You are putting words in my mouth now. I never said air moves through a pipe at the same rate. Any idiot who has read anything about fluid dynamics knows how laminar flow works. It's layers of air (or liquid) that slide along next to each other smoothly, and the velocity increases from the outside towards the inside (or the path of least resistance) with the highest speed normally being in the center of a straight pipe.

I also didn't say air not sticking by the wall made a flow turbulent. I said if there are breaks along the wall it can make the air tumble, and if there is a drastic enough change in direction and enough tumble added the flow will change from laminar to turbulent. Since you are a mechanical engineer and studied fluid dynamics you should know what I'm talking about. It's called the dimensionless reynolds number. It varies by specific geometry of the tube and substance that is flowing through it, and it's sensitive to introduction of turbulence and imperfections in the system... but once you reach that threshold flow changes over to turbulent. So my reasoning based on that theory is you should try to avoid introducing any unnecessary turbulence if possible.

as for what i said about MAF tables, I meant the maf tables are designed "in a car" meaning on an engine, with exhaust, intake box, etc, which would be the opposite of in a lab on a straight piece of pipe like you said.

Yeah, I understood what you said before and I'm saying you are wrong about that because different cars that use the same engine and MAF sensor but different intakes and exhausts have THE SAME MAF table. I thought I said this already?

as for units, I'm pretty sure the pcm does in fact have both sets of units as outputs, but in the end, who cares what one company uses or another company uses. it's all the same thing, one unit is the same as another unit times a coefficient and vice versa. HP Tuners is setup for both. I prefer imperial. I'm not the only one, or it wouldn't be on there. If you like metric, great. It really makes no difference in the end.

Why on earth would GM make the PCM output two different values for MAF signals? It would be completely unnecessary for them. You are right that it doesn't matter which tuning company uses which output. I'm just saying that's not how GM measures it. Any company can take the output though and do a calculation in their software to convert it to any measurement they want.

[AaronGTR]

Originally Posted by MercenaryX2

One thing I know about changing anything intake related is that once you do that, your MAF scalings are thrown off. It may not be a huge difference but there is still a difference. I saw it all the time when I went through a few different intakes on my DSM.

This can be true to a degree... but a lot of it depends on the specific car and the kind of MAF sensor it uses, and the design of it's intake. I've read a lot of magazine articles where they dyno tested intakes on a lot of different cars, and some made power with no tuning necessary. Others such as turbo Subaru's like the WRX were so sensitive to intake changes they would either run lean or actually lose power unless they were re-tuned with the new intake and then they would gain power.

[MilzyZ34]

remember when I said "I'm pretty sure the pcm does in fact have both sets of units as outputs..."

well the pcm obviously has one unit of measurement measured by the MAF, which would be an INPUT to the pcm. What I said is that both are outputs. You can monitor both standard and metric units with any scanner, not just HPT, which is why i said what i said. the pcm actually outputs multiple readings for maf numbers, not just two.

as for the peak flow being right in the center of the pipe, this is only true if the pipe is perfectly straight, otherwise this max flow will occur at the position of the main stream-line I mentioned in my previous post, the position of which is continually changing as the air moves through the intake tube from one bend to another.

[AaronGTR]

Originally Posted by MilzyZ34

remember when I said "I'm pretty sure the pcm does in fact have both sets of units as outputs..."

well the pcm obviously has one unit of measurement measured by the MAF, which would be an INPUT to the pcm. What I said is that both are outputs. You can monitor both standard and metric units with any scanner, not just HPT, which is why i said what i said. the pcm actually outputs multiple readings for maf numbers, not just two.

No, the pcm only receives and outputs one signal for MAF reading. It only needs one to do it's calculations. Why would it use more than one? It's the scanner doing the converting to different readings. Not the PCM. I can do this in the scanner on DHP. I can open a gauge and click on properties and modify the equation to make it read in any scale or unit I want. That's how I set up my analog input to show the reading from my wideband.

as for the peak flow being right in the center of the pipe, this is only true if the pipe is perfectly straight, otherwise this max flow will occur at the position of the main stream-line I mentioned in my previous post, the position of which is continually changing as the air moves through the intake tube from one bend to another.

uh, duh.... that's why I said the same exact friggin' thing you just said. Laminar flow increases velocity from the outside to the inside of a pipe until it is at the center or the path of least resistance... AKA the path through the tube where the most flow occurs. I guess you didn't both to actually read what I said?

[MilzyZ34]

Originally Posted by AaronGTR

No, the pcm only receives and outputs one signal for MAF reading. It only needs one to do it's calculations. Why would it use more than one? It's the scanner doing the converting to different readings. Not the PCM. I can do this in the scanner on DHP. I can open a gauge and click on properties and modify the equation to make it read in any scale or unit I want. That's how I set up my analog input to show the reading from my wideband

I agree there is one input from the maf to the pcm. i disagree about the outputs. You know what, think whatever you want, I dont care. The world is flat, the moon is made of cheese, and you are always right.

[AaronGTR]

Originally Posted by MilzyZ34

I agree there is one input from the maf to the pcm. i disagree about the outputs. You know what, think whatever you want, I dont care. The world is flat, the moon is made of cheese, and you are always right.

Now you are just being ridiculous. Show me ONE credible source showing that the PCM outputs more than one value for the MAF reading. Anything believable that's not nonsense. Please...


There is absolutely NO reason why GM would need it to do that. Given the limited computing power of the PCM, and the fact that the more signals you add to sample while scanning slows down the refresh rate through the OBD port, that would be a stupid thing to do. It makes more sense to let the scanner do that work, since it has more resources available.

[sleepyalero]

Jesus Christ this is still going on.