chuckster

I was looking over some old tuning posts struggling with KR on Aftermarket MAFs.. So I decided to put together a document to cover the facts and dispell some myths as to why they do not work.

Why Aftermarket MAFS and MAF ENDS introduce Knock Retard on the LS1 Engine


Aftermarket MAF’s and MAF Ends are a big cause of Tuning Problems and Knock Retard. The ones with recalibrated electronics are bad and the plain MAF ends are worse. It seems that GM really did know how to make a stock MAF work properly and keep the Velocity up.
First let’s understand how the MAF Works.

MAFS measure air volume by determining how fast it can cool down the “Hot Wires” inside it. The stock MAF is fixed in diameter and is calibrated at the factory to respond a certain way as the measured air moves through it. The faster the air moves across the wires the more volume is being interpreted by the PCM. It then adds fuel accordingly and determines the Timing needed. If you change the ends of the MAF to be bigger, you in effect increase the “ability” of the MAF to flow more air. But. The reality is that you have just slowed the velocity down and have not increased the volume at all. To demonstrate this I will use this example..
Lets use an engine spinning at a steady 3000 rpms. It will require a “fixed” amount of air volume (CFM) to continue to pump at that rpm. Lets say you had a Stock MAF that was 3” in diameter. If you measured the velocity and the CFM you would get some value for each. Then swap the MAF with a 6” one and then take the same measurements. You would see the velocity of the 6” MAF drop significantly and the CFM stay exactly the same.

This problem of KR is well known and misunderstood. Most everyone believes that the KR is caused because the MAF is Bigger and therefore it is Flowing more air. Since it is flowing more air then it is leaning us out. This is wrong. KR shows up in 2 places for different reasons.

Part Throttle KR - The KR that shows up during part throttle driving shows up because the Air is traveling though the MAF at a much slower VELOCITY due to the fact the opening is bigger. Now the PCM sees that slower velocity as low engine load and in turn jacks you up to a higher timing point. All we have successfully done with this bigger MAF is fool the PCM into doing something we don’t want.

WOT KR - The KR you see at WOT is most definitely due to the lack of Fuel. But again, the lean condition is not caused by the fact that MORE air is flowing. Actually, it is still flowing EXACTLY the same as the stock MAF. The problem is caused by the fact that the Velocity is slower than the stock MAF. Therefore the PCM now thinks it does not need as much fuel and starves the motor.

Velocity is King.

Dirty Howie

I found this very informative.

Could you do another lecture on ping/rattle with stock MAF and cold air intake.

Thanks

leaftye

I wish I read this a year ago before I descreened and ported my MAF housing. I've known for years that the dimensions of the intake runners would affect power delivery, but didn't extend my line of thinking to the MAF, so I thank you for the enlightening me. I've been struggling for the past couple months with KR while at WOT and low RPM, and this helps me understand why I'm having problems. I've strongly suspected the MAF to be the source of my trouble for months, and now this is enough to push me to get a new unmolested MAF sensor/housing.

nuke61

I don't agree with either explanation. Let me talk through it...

The MAF essentially tries to evaluate the mass of the air passing through the housing. It works, as you've said, by the cooling done to a heated wire. Since the CFM rate is the same for both MAF's, the cooling done to the wire should be the same, therefore, the calculated mass is the same.

A basic thermodynamics formula is Qdot=Mdot * deltaT, in other words, the heat transfer rate is equal to the mass flow rate time the temperature difference. Velocity is implied in mass flow rate... whether you have a 3" opening or 6" opening, air mass is constant. Where and how does the PCM determine air velocity? It doesn't; there is no air velocity sensor input to the PCM.

chuckster

Here you go..

Pinging on LS1 with Cold Air Intakes

Sometimes pinging is introduced when a cold is intake is installed on the LS1. For the most part it is NOT caused be the cold air itself.

It is caused by turbulence or the interruption of air flow through the MAF.

Some Cold air intakes have the filter mounted so close to the MAF it does not give the air enough time to form a uniform stream before it enters the MAF. Some Cold air intakes have a tube that the air must travel through before it enters the MAF. These usually work out better.

People who open the front ducts in the C5 to allow cold air in also introduce some heavy turbulence in front of the Cold Air Filter. So they are more likely to have pinging issues if their Filter is short and does not give enough time for uniformity of the air volume.

With a MAF, the Slightest interruption in flow will wreak havoc on the PCM’s ability to decide what is happening..

To prove this point try the following experiment.

Remove the entire air filter assembly leaving nothing but the exposed MAF. Start the car. Then while the car is idling go over to the front of the MAF and blow across it.. The motor will stumble.. Why? Because the PCM saw the interruption as less airflow and consequently starved the motor of fuel.

My personal opinion is that the MAF should have been mounted behind the Airbridge not in front of it. This way the Airbridge will ensure a uniforrn entry of Air.

This brings me to the next point. MAF Screens. The Big Debate….Pull em or Don’t Pull em. The Screens job is to keep a continuous stream of air directed at the MAF wires. That is it. The statements I see about it being a filter for large foreign object is pure BS. If you have a cold air intake that has a short plenum then I say Leave it. It will help fight off turbulence. If you have something with a longer plenum you can safely pull it.


Hope this helps..

chuckster

This is why there is so much confusion surrounding the great MAF Debate... Not seing the forest though the trees.. Keep one this in mind before before using complicated calculations....

the MOTOR has a FIXED pumping volume (Normally aspirated) and any given rpm. It is PULLING a fixed CFM through the MAF at a given RPM. It if it is fixed and you make the MAF opening bigger.. You effectively slowed down the velocity because it has nore room now. If the motor was pulling lets say 100CFM that means that in one Minute 100 cubic feet of air will pass through the opening. Whether the opening is 6 inches or 6 feet. But. if you were standing in front of that 6 foot opening you would not feel even a breeze of air movement. But if you put you hand in front of the 6 inch opening you'd get your hand sucked in. Again Velocity is what is being measured. If you put a hot wire in the 6 foot opening it would burn up. No air MOVEMENT to cool it down. For the bigger opening to correctly measure the CFM ALL of the air would have to be directly focused on the hot wires.

Using a simple room fan as an example.. Slow moving air cools less than fast moving air. Period. If a room is 70 degrees and the air is Still it does not have the same cooling effect as if there is a ceiling fan moving the air. You have not decreased the temperature.. YOu have increased the airs ability to remove heat surrounding your body.

All the MAF is doing is measuring how FAST the wires cool down. Period. It's calibrated to return a certain frequency for every temperature change. All the rest of the crap is done inside the PCM from a combination of the IAT sensor, Throttle position, etc..

Also there is one other flaw in that thinking....The Bigger MAF allows much more air to pass by the wires without ever having a cooling effect. So by diluting the flow of air with bigger ends ruins the cooling effect.



[Modified by chuckster, 12:38 PM 7/4/2004]


[Modified by chuckster, 12:42 PM 7/4/2004]

nuke61

Complicated calculations? Huh? It doesn't get much simpler than A=B*C

There is air movement. Since you said there's a volume of air being passed through the opening, there must be movement. If there is no movement of air, there is no CFM of volume. If the wire burns, its because it was improperly calibrated for the 6' opening. Nothing more, nothing less.

That is not true, what is needed is proper calibration of the output based on the amount of cooling the wires recieve.

Again, not true. Less volume of air moving past an object cools less than more volume. If you increase the velocity but keep the mass of the air volume constant, you cannot and will not increase cooling.

Heat transfer rate is equal to the mass flow rate times the temperature difference. You cannot get by this basic thermodynamic equation, no matter how hard you try.


That's because the fan is moving more cooling *volume* (mass) past the hot object.

That's a problem with the calibration and has nothing to do with the size of the opening. Chevy and Ford, over the years, have themselves increased the size of the MAF with no ill effects. Why? Because they have accurate calibration of mass flow rate based on hot wire cooling.

To compensate for inaccurate calibration due to aftermarket MAFs or ends, people have been using LT1 Edit and the MAF Translator for years. All they do is allow you to accurately adjust the MAF sensor input to the PCM and let the PCM know that the motor is getting more air.

You really need to do some study of basic thermodynamics, because your understanding of heat transfer is sorely lacking.


[Modified by nuke61, 12:27 AM 7/5/2004]

chuckster

George, you win.... your intellect has astounded me. You are correct I do not have a degree in physics (if it matters my father has a Phd in Physics). And I am sure my writeup does not meet your standards for a passing grade Mr. Instructor. :P

... However I am not trying to Impress anyone with my intellectual prowess or get an A on my understanding of heat transfer.

I am trying to help other resolve the KR Issues surrounding the use of Aftermarket MAF ends....I have used the MAFTranslator for years and then used LS1edit after that.. Yes, I did "TUNE" out the KR. I had no problems doing it. In fact I was the poster child telling everyone don't even bother to use maf ends if you don't plan to tune them.

But the point is simple here.. it only gets you back to square one with NO gain. I posted about how to tune properly to get rid of the KR. But realized that my car simply responded better and gave me more HP when I went back to the stock MAF.. I stand by my point that the Air velocity going though a stock MAF is higher than a ported MAF.

If you want to be helpful then answer the question SIMPLY so that everyone understands. WHY DOES THE AFTERMARKET MAF CAUSE PINGING AT PART THROTTLE AND AT WOT???

You have been a member here for 29 days and your response to me makes me think you are annoyed at my utter ignorance... If you read the responses I have helpful so far to others except you. Keep it light hearted my friend...


Please leave the formulas at the door.. They are of no interest to me. Everyone has their own areas of expertise.. But it is the True Intellectual that can convey his knowledge to others without them needing a degree in that field.

Dirty Howie

Thanks again...your comments and reasoning are very logical and seem to be backed up by experience.

One of the reasons I got the Haltech TRAP was because Andy at A&A Corvette pointed out that he thought the cold air retrieved from under the car would be smoothed out (less turbulent) than the STINGER which is the shortened version (same filter and air bridge). He talked me out of the only CA certified CAI, the FPIK because he felt the twin cone intake would cause obvious turblulence at the MAF.

Anyways.....should a proper LS1 edit result in ZERO ping/predetonation/KR no matter what CAI or stock air box is used???

427CPE

Chuckster,

I worked with Pro-M over a year on other projects directly related to the MAF. In fact, right now I am working on a flashable MAF for them to help alleviate some of these problems. Problems commonly enountered in MAF calibrations.

1. Changing an AIR BOX will automatically throw off the meter. The theory about the wires doesn't mean a thing if you create bends if front of or change the flow of air across the meter from a characteristic other than stock. If you create a situation where air flows down one side or the other of the sensor, guess what? NO METERING. Even clocking the meter at different angles can
affect it. No metering, no fuel commanded=lean plus lower load spark conditions causing ping. Also, VERY significant is the fact that the maf is also used to calc load. If the loads are lower than expected due to lower freqs sig to the pcm, expect lower load parts of the spark tables to be utilized.

2. Try a datalog, you will see aftermarket airboxes generally cause lean conditions and generally speaking so do most aftermarket mafs. The last
one I logged indicated less airflow at idle than stock maf. Then again, so did
changing my airbox.

nuke's points are all good, except they probably dont' have a lot to do w/the problem at hand.

nuke61

chuckster, the basic issue is that your explanation can't be backed up by basic science. My explanation of why these problems occur? Incorrect calibration and/or turbulence. How to fix it? Proper calibration and a straight section of tubing prior to the MAF or a flow straightener.

I'm not annoyed... just simply saying that your explanation makes no scientific sense.

I think you've posted an incorrect explanation of why aftermarket MAFs may cause problems. I disagreed with your explanation and told you exactly why I disagreed with it. It's just simple discussion... I never called you any names, or YELLED at you. All I did was expain where your explanation didn't make sense. If you have a problem with that, you've got a very thin skin indeed.

chuckster

I agree with everything above.. In fact you sould like you have quite a bit of experience..

I especially like your point of the air getting the hot wires unmetered...

If you create a situation where air flows down one side or the other of the sensor, guess what? NO METERING

This is exactly the condition I was trying to point out..No Air FLow = No Cooling of the wire = Lean.

Which points out one MAJOR flaw in the MAF design.. How does one make it big enough for maximum flow at WOT and still have it be sensitve enough at low speeds..

The last point I wanted to make was that Other than Forced Air or Larger Cubic inches, there is no point to a bigger MAF. Once you recalibrate with LS1-edit you are back to the same point you started.. But in my opinion less performance.. Due to the lack of velocity.
I have tested extensively both ways and had my MAF ends perfectly tuned with LS1edit. Whenever I went back to stock the car responded much better.



[Modified by chuckster, 12:51 PM 7/5/2004]

chuckster

Here is a little more supporting info for the Velocity Theory..
Although they are discussing Heads... The same theory applies..The smaller the the port the higher the velocity
http://performancetechniques.com/hardeyhead9/

The Excerpt is below...

Larry Hardy has worked hard at putting together various stages of headwork on Harley Heads, that we at Performance Techniques believe puts us at the top of this game. When dealing with the stock heads it should be understood that the ports start out too big. This fact is proven by the 883's heads with its comparatively tiny ports. With 140 CC.s less than a 1340 Big Twin, a 1200 with 883 heads can produce very similar torque and horsepower numbers. The key to street power is "Velocity". Even after the 883 heads are well ported, they flow close to the same cfm. as a stock unported set of Big Twin heads. Because the port is so much smaller, the velocity is substantially higher. This allow more cylinder fill and thus more torque and horsepower.
Making a broad powerband is accomplished by volumetric efficiency. The goal is to pack as much air and fuel into the cylinder over the widest rpm range possible. This is done by meeting the engines design needs with the smallest port we can get away with. This smaller port creates more velocity, which in turn creates more inertia. Visualize the principle of inertia as a moving train of air attempting to pack into any given area before it stops. The faster the train is going, the more force it develops, and the more mass it will pack before it stops. That is what happens in an operating motor. As the cylinder fills it starts to exert forces that want to stop the incoming train of air. The more inertia there is the longer that train continues before being stopped, thus providing more cylinder fill. When one realizes that inertia = velocity squared x volume (cfm.), it becomes apparent that velocity is more important to cylinder fill than volume.



[Modified by chuckster, 11:41 PM 7/5/2004]

Sandra Bigwoode

So what if you increased the size of the MAF to 12", the CFM remains the same but the velocity of the air slows even further, are the heated wires getting the same amount of cooling air?

The molecules in the air mass are less dense and therefore will cool the wires less... unless you assume longer wires to fit the 12" MAF...

leaftye

According to John Lingenfelter in John Lingenfelter on Modifying Small-Block Chevy Engines: High Performance Engine Building and Tuning for Street and Racing, having a smaller path for air to flow within can help more air flow at low speeds. This is why big valve, big lift valvetrain systems don't do well at low RPMs. This is also the problem that Variable Valve Timing solves. It does this by having one small cam open the valves at low RPMs, but at higher RPMs the big valve takes control and opens the valves as much as they will go. BMW or maybe Toyota has been working on another part of the solution as well. They've been fielding a system with essentially two intake styles in a single system: skinny & long runners for low RPMs, and fat & short intake runners for high RPMs.

We already know GM is working with Siemens to develop electronically actuated valve mechanisms, which with some fancy programming, should be able to do everything that Variable Valve Timing does, along with being able to shut down certain cylinders for cruising. If GM also starts to field Variable Intake Runners, we'll start to see engines that can get 40 mpg, have incredible low end torque #'s, and awesome top end horsepower.

If you can somehow right about how an extremely wide intake can flow the same amount of air as a properly tuned smaller intake, given the same demands, you could be a rich RICH person if you could convice GM, Honda, Toyota, and BMW of the same thing.

As for simply making sense, it makes much more sense that it's easier to measure airflow thru a smaller path. For example, a skinnier path would show greater fluctuations in air speed, where a much wider path would barely show any change in air speed, even with the same increase in air flow. This means the sensor would have to be much more sensitive to be able to "see" the changes. This could also be compared to o2 sensors. A normal narrow-band sensor only uses 1 volt to send it's signal, but a wide-band sensor requires 5-volts. I'm not totally sure why this is, but I believe it's so that the sensor pickup can "see" changes more easily, instead of trying to break 1-volt down soo much, it has much more to work with if given 5-volts.


As far as where my opinion is now...
I still think the stock MAF was good enough. I destroyed sensors ability to meter by removing the screen, and since the stock meter was already more than capable of flowing enough air, I unneccessarily ported it, and probably made my car perform worse because of it. Even if the air flow could've still been used, I still want a meter with the flow straighteners.

Right now the only problems I think the MAF has is really with the IAT sensor. One of the forum members, PhilSVT I think, posted about how he feels more power on the street, but feels less on the track. He believed his problems were because the underhood temperatures produced high IAT readings which was causing the ECU to introduce KR, and produce less power. His suggestion was that racers ice down their MAF while staging. This totally makes sense to me since a well sealed engine bay and a MAF located directly above the radiator can't be good for a temp sensor that's supposed to be getting intake air temps, but instead appears to be getting heat soaked.

My theory is confirmed when I look at the logs for my engine. I get knock only at low RPMs, and I get it when I first get into the throttle while under load. Eventually it goes away when the knock sensors catch the resultant detonation, but this shouldn't even happen if the MAF knew about the increase in airflow.

For now my attempts at tuning are over until I install a new stock MAF and reload my stock program.

blu00rdstr

That caught my interest. Once you go FI and start flowing a huge amount of air, what's the next step with a MAF? I'm running 3.5" round pipes, and outside of reaching the frequency range of the MAF, it and the throttle body have 75mm/2.9" ID.

427CPE

Anyway, after ~some research, a very brief discussion with LS1 Edit's programmer, and some real time dyno testing, a few ideas:

1. The PCM after reaching the max frequency will use an unidentified
error routine that throws huge amounts of fuel at the engine. I have a log from a blown Z06 that supports this.

2. RECAL MAF. The stock MAF was created with the airflow potential of the stock engine in mind. You would simply use a recaled MAF to fit your new configuration's airflow. As long as you keep the new scale near the real time output of the engine (like my car which is ~11,500Hz), load functions will not be skewed. At this value, I see a max of 0.76-0.80gm/cyl load.

3. REPROGRAM EXE CODE OF PROCESSOR(yeah right): The labels you see in LS1 Edit and HP Tuner are not real time data retrieved from the calibration area of the processors. They manually input that in to their programs. Seems it is tangled up in the EXE code in such as fashion that it is hard to edit. I suspect it won't belong before this function is available in one of the software packages.

4. At the 0 vacuum setting of the IFR table (WOT), one could adjust this one fuel injector rate to the "desired" outcome since it essentially represents WOT fuel. Funny thing I noticed, there are 17 IFR cells, yet 1-10 are really the basic ones used. At Cell 10, labeled 45kPa, the engine is at idle and thus all cells to the right really are not used giving you 9 cells to truly tune with.

Anyway, its late...recal the meter if you are over the maf limit, have it flowed, then input the real data in the transfer function.

nuke61

That's correct, I assumed that one would increase the size of the sensor element in that case. If you didn't, it would obviously heat up more, requiring the recalibration of the MAF output. I still think, like I've already stated, that the problem is due to incorrect calibration and/or turbulence and it can be fixed by proper calibration and/or a straight section of tubing (or a flow straightener) prior to the MAF.

Another alternative that the high horsepower LT1 folks have used is to ditch the MAF entirely and go entirely with Speed Density.

ArKay99

nuke61, I believe you are missing chuckster's point. The point being UNMETERED flow or mass. The wires are the dynamic part of the sensor and they are a fixed size. Therefore if a greater amount of air is passing through the body of an MAF due to some modification the PCM will recieve altered parameters. If the calibration of the MAF table is not altered to reflect this difference in velocity, contributing to greater wire cooling, the PCM will adjust it's timing and fuel delivery incorrectly for optimum performance. The other factor that hasn't been discussed here but implied is the relevance of the Volumetric Efficiency tables. During closed loop operation the VE tables are looked at in conjunction with the MAF inputs and load is calculated based on these two tables and MAP. Timing and fuel delivery are determined through calculations combining the lookups of these three tables, and actually several more. If the air mass transfer function has been altered the final product will result in altered fuel and timing. Also, during open loop mode, the PCM depends almost entirely on the MAF for it's air calcs. These are used with values in the IFR tables and PE RPM tables to determine AFR. If there is altered flow, the IFR table will be off because calculations based on lookups of it's cells are determined by MAP.


[Modified by ArKay99, 2:19 AM 7/7/2004]

nuke61

I've read throught the thread again and you're right, I didn't understand his point because I assumed that the sensor wire size increased along with the opening size. Chuckster, I aplogize for being so dense and such a