PDA

View Full Version : MAF housing restriction?



mb7050
11-15-2010, 07:44 PM
How far is the oem maf good for ?

I found this data
55mm (Stock 88-93 Mustang) 275 HP
70mm (Stock 94-95 Mustang) 350 HP
75mm Pro-M Bullet 550 HP
80mm (Stock Ford) 450 HP
Pro-M 85mm Flanged 650HP
Pro-M 95mm Flanged 750HP
80mm Pro-M 800 HP
92mm Pro-M 1500 HP



this is the data I have gathered this far
3s OEM maf = 650cfm = ~433 CHP ? at compressor Inlet Conditions
gm 3.5" MAF w/ screen = 759 cfm = ~506 CHP
gm 3.5" MAF w/o screen = 979 cfm = ~650 CHP
3.75" Zo6 maf = 1202 cfm = ~800 CHP

Keyan
11-15-2010, 09:12 PM
speed density = infinite powaaaaaaa

:p

Atrosity
11-15-2010, 09:14 PM
speed density = infinite powaaaaaaa

:p


Psht, you're an infinite powaaaaaaa

GM Maf ftw.

mb7050
11-15-2010, 09:16 PM
yeah speed density wins but the pro m 92mm supports about 1500hp that is plenty (same size as arc 2 big maf)

mb7050
11-15-2010, 09:21 PM
and if you think you need something bigger you can buy the 4"=100mm maf which will flow even more
http://cgi.ebay.com/ebaymotors/4-Inch-MAF-GM-LS2-LS3-LS7-L99-Camaro-Vette-TBSS-GTO-/180586863440?pt=Motors_Car_Truck_Parts_Accessories&hash=item2a0bd0db50
http://www.lingenfelter.com/mm5/merchant.mvc?Screen=PROD&Store_Code=LPE&Product_Code=L480000000&Category_Code=C162

mb7050
11-16-2010, 03:48 PM
JDM MKIV Supra ECU uses speed density for air-metering there must be a reason why all the others, US and Eurospec supras come with maf air-metering ?

mb7050
11-16-2010, 06:26 PM
good reading @ ls1tech.

I think the MAF is definitely the way to go if you can get it to work. Now no doubt that SD setup can run well, but I would still submit that under a variatey of conditions the MAF will come out superiour in drivability, etc. as well as virtually the same in power.

The potential advantages I see for SD are

1) Removing the MAF - "restriction" issues

2) Response time lag (mostly on FI cars with the maf on the intake side).

3) Freedom in routing the intake tract.

A for #1, the maf being a "restriction" - it depends on how much power you are making. On a stock or heads/cam car I don't think it is an issue at all. As for how much power the maf will "support" - well, you can flow as much air as you want through it, as the maf starts becoming a restriction you will just get a larger pressure drop. It's not that you will simply rail out at a certain point. Now as far as air metering, it will max out somewhere from 450-500rwhp normally (depending on your bsac, etc.) - you can easily tune past this point - it will be the same as tuning a SD car at WOT at that point anyway - use the PE vs RPM.

As for #2 I would keep the maf as close to the TB as possible - again, mainly an issue of FI setups. I would just move it to the pressure side. That way you can run a normal BOV also.

As for #3, well - if your intake prevents you from using a maf at all, then you are doing SD - but in a vehicle whose configuration allows for a MAF I would go for it.


SD tuning to me is old school, it's like tuning a an old school controller. You dyno and lock in all the parameters.

99% of our members can use MAF tuning and be very happy, do any of your SD tuners have any 10 second heads/cam cars? Any 9 second motor cars?

I know Ford guys who tune like that because they are using stuff like EPEC (sp?) and it's not because they want to. They end up changing the tunes depending on track weather and altitude.

And we have guys running 105-118mph with stock internals.


I've tuned my car both way's.SD ran a little better on the street but dyno'd the same.SD had a better throttle response for sure,but when the weather changed about 30 degree's the tune went off on me.Decided I liked ussing a MAF and tuned in the VE tables till it was just as responsive as SD.

just to share few posts :)

mb7050
11-16-2010, 06:33 PM
one more



However I feel SD is more suited for the track, than it is for the road. This for someone that just wants to do a few mods and tune it, and leave it for a long time like it is, maybe some minor fiddles, the MAF tune would be better. If you want the most outa your car, race etc and are always changing things, retuning etc then SD is more suited.
Why, one point only really will spend some time and try to explain better than I usually do, think a bit, it hurts however. The purpose of any EFI system is to inject an amount of fuel into the engine. To do this, the PCM must know how much air is going into the motor at ANY given point.
What you should know is how much oxygen is in the combustion chamber at the time of ignition. The only way to tell how much fuel to inject in to mix with the available oxygen ( counting the molecules is hard) So you need to calculate the mass of air in the engine.
The amount of mass of air/gas in a given volume (the cylinder) depends on the temperature and pressure of the air. So SD setups measure the pressure in the intake manifold, and the current engine speed (which gives volume), from calculating these the pcm knows the mass of air in the combustion chamber. Speed density.SD is very fast, pretty accurate, and because it measures the intake charge in the intake manifold (close to the combustion chamber) it gives very good results.
But they have one problem, unless I’ve missing something else. The temperature and pressure of the intake charge in the cylinders wouldn’t be really the same as it is in the intake manifold. The air passes through the cylinder heads and past the valves first, and other factors can reduce or increase the amount of air that actually makes it into the cylinder. To allow for these differences, a speed/density PCM needs more info, the correction table called volumetric efficiency.
With volumetric efficiency (VE), you run the engine at each load point on a dyno, and manually adjusting the table until you get the correct air/fuel ratio as measured by an exhaust gas oxygen sensor. Then, after its is all perfectly tuned and producing nice power anytime after this you do anything else to the engine that changes breathing, you must recalculate all the VE points again.
This means as the engine ages, accumulates carbon deposits, cat converters clog up, etc etc etc, the VE of the engine is constantly changing, the perfect tune is constantly deteriorating. Emissions are also increasing overtime. It also means that if you do something to make a big change in VE - like change catbacks, extractors, intake systems, for example - that perfect SD engine tune will suffer. You need to go and reset the tune again

So holden like GM have used a different method of determining the mass air charge. Instead of the speed density systems, they use a sensor that measures air mass directly, a maf.
From the point of view of the factory, a maf is a great. As air mass is measured straight off the sensor, the maf compensates for wear, or ppl modifing the car to improve it (why you may need to reset the pcm to relearn). Since it has been mapped a given air mass vs RPM in the computer's tables, you can do what you like to the engine's VE and the computer can adjust for it. So you can make small changes and not stuff up your maf tune to much.

So it all depends on what you want to do with the car which tune to get. A maf tune is more forgiving, so could be said to be safer due to the point about VE and SD tunes.

mb7050
11-16-2010, 06:45 PM
Speed density is definitely no go up here as the weather changes from ~ -50 degrees Fahrenheit to 100 degrees Fahrenheit and Iīm not the kind of a guy who wants to be tuning my car all the time.

mb7050
11-16-2010, 07:00 PM
... maybe thats why all the speed density cars run like crap here :)

Chris GTO TT
11-16-2010, 10:31 PM
Similar reasons why piggyback tuning is so limited on these cars. Also those running MAF-T's run into similar problems that those running speed density since the MAF-T locks in baro and IAT's.

mb7050
11-17-2010, 12:06 AM
how about if you use IAT and BARO signals from those late model gm mafs or hack up the stock maf just to get IAT and BARO reading,

Atrosity
11-17-2010, 12:34 AM
how about if you use IAT and BARO signals from those late model gm mafs or hack up the stock maf just to get IAT and BARO reading,

This is what bboyalan and I are doing with my car. I just bought a stock maf from Ground Zero Performance, we are going to hack up the stock maf and we have a pigtail for the harness to wire in to use the IAT and Baro sensor. This is going to be very nice with my GM maf and maf translator.

mb7050
11-17-2010, 08:17 AM
I was looking at some data from those late model GM mafs (3.75") and I found out that the IAT sensor in those is actually very close to our IAT specs

mb7050
11-17-2010, 08:31 AM
the GM MAF frequency output should be baro and temp compensated, so there should be no need for the ecu to calculate that again
'
The stock sensor is a volume sensor, so the ecu needs temp and baro data to compensate the airflow reading.

If we connect an IAT and/or baro sensor, then we would be double compensating for the air

and then this


The GM meter is a true mass airflow meter. The Mitsubishi in comparison is actually a velocity meter with a baro and IAT sensors used to calculate the mass flow rate. The GM sensor measures changes in density directly and the output reflects a "true" mass airflow. The Mitsubishi sensor outputs a frequency based on the air velocity, then the ECU calculates mass flow rate based on baro, intake temp, and MAF frequency. The caveat here is that the GM sensor also is to some extent a humidity meter as well, as humidity also affects the sensor output, where I think humidty only marignally affects the Mitsubishi sensor, if at all.

The issue with locking the baro and IAT sensors is you no longer have baro and IAT trims. While you will compensate for density changes because of these different ambient conditions, you will not get corrections that will adjust fuel and timing to accommodate these different ambient conditions.

On one hand, this SHOULD provide more stable AFRs and timing advance, provided the MAF accurately measures air mass under various ambient conditions. On the other hand, the car will not be tuned to best optimize different ambient conditions.

To take it one step further, it seems like the EVO ECU should be able to accommodate a GM MAF directly. The GM MAF uses a Frequency output. Provided the EVO ECU can accommodate the frequency range of the GM MAF, changing the MAF scaling table to match the GM MAF output should provide reliable fueling. You would want to disable any density compensation within the ECU however, which is seems like has been discovered recently.

If the EVO ECU can handle the GM MAF frequency range, I see no reason why direct GM MAF operation isn't possible? Futher, combining something like a Subaru MAF (hot wire 0-5V) with a voltage to frequency converter could allow a good post turbo MAF sensor solution, as the Subaru MAF has proven to be much more stable in a pressurized enviroment then the GM sensors by the Subaru community.

Also, I think this thread title should be relabeled as something to do with "Direct GM MAF Implementation Discussion," or something similar, as it would probably get a lot more knowledgeable people involved.
Read more at http://forums.evolutionm.net/ecuflash/389737-question-about-barometric-pressure-sensor-interesting-project.html?ktrack=kcplink

"If the EVO ECU can handle the GM MAF frequency range, I see no reason why direct GM MAF operation isn't possible? Futher, combining something like a Subaru MAF (hot wire 0-5V) with a voltage to frequency converter could allow a good post turbo MAF sensor solution, as the Subaru MAF has proven to be much more stable in a pressurized enviroment then the GM sensors by the Subaru community".

mb7050
11-17-2010, 08:44 AM
This is what bboyalan and I are doing with my car. I just bought a stock maf from Ground Zero Performance, we are going to hack up the stock maf and we have a pigtail for the harness to wire in to use the IAT and Baro sensor. This is going to be very nice with my GM maf and maf translator.

Trevor did this years ago I wonder what was his conclusion ??

mb7050
11-20-2010, 07:42 PM
Anyone replaced the stock intake bubble with something like this ?

4" piping

http://www.westcoastcorvette.com/images/Product/medium/2548.jpg

3.5" piping
http://imgs.inkfrog.com/pix/mqtautoparts/Dual_Intake_blue.jpg

mb7050
11-20-2010, 07:52 PM
good read about our maf

Programming the Mitsubishi MAF in AEM
http://farnorthracing.com/stealth/images/MAF.jpg

Mitsubishi uses a unique Mass Airflow Sensor (MAF) in most of its EFI cars - unique, because it isn't really a real MAF. Instead, it measures air volume, not mass.

This sensor uses a technique called the Karman Vortex Street to measure air volume. The inlet airstream passes over an airfoil. This airfoil generates vortices in direct proportion to the volue of air flowing over it, and a pressure transducer is used to count the vortices.

Accordingly, the output from the sensor is a pulsed (frequency) signal that is proportional to airflow volume. An Air Temperature Sensor (AIT) inside the sensor body is used in conjunction with a calibration table in order to correct volume for density.

On the MAF sensor used on a Stealth or 2G DSM, the sensor is capable of measuring roughly 2800 Hz of air, which corresponds to approximately 650 CFM.

On a Mitsubishi using a "MAF", what you really have is a volume, plus some corrections tables. This makes for a calibration which shares elements of a true Mass/Air system and a Speed/Density system - it is a hybrid of the two. You are less concerned about determining VE the way you are in a "pure" Speed/Density, because the volume sensor accounts for VE (you are measuring volume directly, not inferring it from Manifold Pressure and Engine RPM) but you still have to correct for air temp and air pressure.

That, however, is largely theoretical. You can treat the calibration as if it were a MAF-based system, once the air temp and baro pressure tables are set.

One immediate AEM "Weird Harold" is that the Load axis units used on the fuel map etc are completely bogus. If, for example, you set the Load units to PSI, the Load PSI will only correspond to actual manifold pressure through sheer coincidence. DO NOT believe the units! Instead, connect a real MAP sensor (see http://www.farnorthracing.com/stealth/MAP_sensor.html for more info) and use it to determine what actual boost pressure corresponds to what Load axis.

Note too that the AEM software will display the scale on all Load related tables and maps using the scale defined for your MAP sensor - in other words, if you connect a 2 BAR map sensor to the AEM, then the Load units range will be 0-2 BAR (or equivalent in PSI or kPa) even though there is no direct relation between the Load value (as measured by the MAF) and the MAP sensor.

It doesn't actually scale the Load axis - don't panic! It just scales the units associated with MAF_as_Load. This is a tie-in with speed/density mode (where MAP units really are the Load axis) and strictly speaking, it's a bug. But it only affects display, not operation. Bottom line: treat Load units as a dimensionless quantity, and treat MAP_as_Load as real units, even though the software makes them look like the same thing.

So then, the MAF sensor produces a frequency signal (in Hz) that corresponds to an air flow volume. The AEM transforms this frequency into a Load parameter which is then used as the Y axis in any number of tables. How does that work?

Firstly, there is an Option for the MAF called VMAF_Flow_Max. This option is used to determine the maximum amount of flow that the sensor can read. Note that this is a SOFT maximum, set in software, that has NOTHING to do with the actual physical flow limit of the MAF (the point where increasing air volume no longer produces an increasing signal.

VMAF_Flow_Max works in a non-obvious way which we will describe later.

The other Option that controls the MAF is VMAF_Flow_Sum_Revs, which is used to determine over how many engine revolutions the MAF pulses are counted per "timeslice" or "window" in which we calculate load. Note that the AEM works in terms of pulses, not in terms of frequency which is how most of the systems that talk to the OEM ECUs describe airflow.

So here's the description of the AEM's operation:
1.The MAF vortex pulses (and fractional pulses) are counted between a number of engine revolutions, set in option VMAF_Flow_sum_Revs, this sum is shown in parameter MAF_Flow_sum_raw

2. This is then corrected for barometric pressure and air temperature as shown in parameter MAF_Flow_sum_cor

3.MAF_Flow_sum_cor / VMAF_Flow_Max = MAF_as_Load as long as MAF_Flow_sum_cor in is the range defined by VMAF_to_TPS_Above to VMAF_to_TPS_Below

4.Since MAF_Flow_sum has a range of 0 to 4095.9, @ 6000 rpm and sum revs of 1 would give a maximum frequency of way over 100kHz.

5.Parameter T4PER will indicate the period = 1/frequency of vortex meter, expressed in microseconds.

6.MAF_as_Load is a 0-100 value indicating the percentage corrected load range of the flow meter.

7.Engine Load is the same value converted to an engineering unit.

8.The scaling for humans can be adjusted via the options MAF_scalar, MAF_offset, Load_scalar, and Load_Offset (I haven't tested this yet - it might be possible to tweak MAF units to match - or at least approximate - MAP units. Or maybe it would be better to just display 1-100)

The trick is that, unlike what it says in the manual and in the Context Sensitive Help, VMAF_Flow_Sum_Revs is CAMSHAFT revs, NOT CRANK revs.

So at a given engine RPM, divide by 2 to get cam revs. Then divide by VMAF_Flow_Sum_Revs (in my calibration, "4") to get the number of "windows" per minute in which pulses can be counted. Multiply by VMAF_Flow_Max to get pulses/min and divide by 60 to get pulses/sec (AKA Hz).

That gets you the maximum MAF frequency at that RPM that the calibration can support.
It's the at that RPM part that's so tricky. There is no timer loop in the AEM that just reads the MAF pulses and puts the current MAF frequency into a register or whatever. Instead, the program just counts MAF pulses per "window" and scales that as a 0-100% load value. (Represented as MAF_as_Load) and then further presented in engineering units (as Load)
But because the "window" gets shorter in time as the engine revs higher, the physical volume of air increases at the same MAF count at higher RPM. A count of 30 pulses in a window is less air at 3000 RPM than it is at 7000 RPM. So if the max pulses number is too low, you run out of "flow" at a low RPM.

Perhaps an example is best:

Using VMAF_Flow_Sum_Revs = 4 and VMAF_Flow_Max = 40

8000 RPM = 2667 Hz
7000 RPM = 2333 Hz
6000 RPM = 2000 Hz
5000 RPM = 1667 Hz
4000 RPM = 1333 Hz
3000 RPM = 1000 Hz
2000 RPM = 667 Hz
1000 RPM = 333 Hz

This calibration ensures that physical MAF overrun takes place at 8000 RPM - up where it is safe. But it is very easy to flow ~ 1500 Hz of air in the 4500 RPM region and hit the VMAF_Flow_Max limit. This is manifested by triggering "fuel cut" which effectively shuts the car off.

So if we change the value of VMAF_Flow_Max to 50, we get:

8000 RPM = 3333 Hz
7000 RPM = 2917 Hz
6000 RPM = 2500 Hz
5000 RPM = 2083 Hz
4000 RPM = 1667 Hz
3000 RPM = 1250 Hz
2000 RPM = 833 Hz
1000 RPM = 417 Hz

It is now possible to physically outflow the MAF starting at 6500 RPM or so (assuming the engine will flow that much air) but we bought ~400 Hz of headspace at 5000 RPM.

Given that VE starts dropping at higher RPM, this tradeoff might be worth it.

The missing piece of the puzzle at the moment is a table that equates CFM to airflow Hz. With that, one could work out an airflow demand curve and pick the value of VMAF_Flow_Max accordingly.

By using the data from http://www.stealth316.com/2-mas_liter-per-hz.htm I was able to work out a polynomial best-fit equation to approximate the Hz to CFM data from the MAF.

y = -1E-08x3 + 2E-05x2 + 0.2452x - 2.8571 if you are curious.

With that, I could equate the max Hz the AEM EMS can read at each RPM to a CFM number.

Then, using Engine Analyzer 2, I worked out some CFM demand values at various boost pressures. These numbers I don't trust very much, especially up top where I know errors in port size and camshaft specs and whatnot artificially choke flow - but the values returned in the midrange seem reasonable.

Plus, the calculated numbers for CFM I'm using are raw, not corrected for air temp and baro pressure. So there's about a 10% swing on the calibration side and a further 10%-ish swing on the calculated demand side.

But at 11 PSI, the default value of VMAF_Flow_Max should be good in most cases unless the air is cool-ish or dense. At 13 PSI, you're right in the middle of the error zone, and at 17 PSI, no way no how.

By changing VMAF_Flow_Max to 46 you reach the limit of the ECU to read increasing MAF frequency at 7000 RPM, meaning that above 7000 RPM, assuming the engine demand is there, it is possible to run out of headspace (something that is impossible with the default VMAF_Flow_Max) At this setting, 11 and 13 PSI are good to go with plenty of room. There is a danger spot at 17 PSI at 6000 RPM where demand is 560 CFM but the AEM can only read 540 CFM - but this is a case where the temp and pressure corrections work in our favour and it might squeak through.

And at VMAF_Flow_Max of 48, demand is always met, but now there is potential to run out of calibration headspace at 6800 RPM instead of 7000.

The bottom line here is that given my intent to run 16 PSI or so, changing VMAF_Flow_Max to 46 from 40 almost certainly buys me all the headspace I need in the MAF to keep me off the boost cut - and all I have to do to fix the fuel maps is multiply the entire table by 46/40 = 115%


More and updated info at ;www.farnorthracing.com/stealth/AEM_Mitsu_MAF.html

mb7050
11-20-2010, 08:31 PM
That means you can flow more air and get a lower Hz reading if you want.


Steve

yes but it doesnt change the fact that the housing 'becomes "bottleneck" after 650cfm' ?...

GTOJOE
11-23-2010, 08:58 AM
Did you know that there's a calibration screw on the stock AFM? Afterall the same AFM is used across a number of models, it's the calibration/ part no. that differs.

That means you can flow more air and get a lower Hz reading if you want.

Also there's no need to meter all of the air, just half of it.

Steve

Where is the adjuster screw on these? cus I have one apart on my desk. MAF sensor, barro and air temp all desoldered from the board. It's completely apart and there is no pot or adjuster anywere. I suspect this is changed by changing resitor values not an actual screw. Unless the plates in the lower assembly can be bridged to change the calibration/configuration.

mb7050
11-23-2010, 03:57 PM
Where is the adjuster screw on these? cus I have one apart on my desk. MAF sensor, barro and air temp all desoldered from the board. It's completely apart and there is no pot or adjuster anywere. I suspect this is changed by changing resitor values not an actual screw. Unless the plates in the lower assembly can be bridged to change the calibration/configuration.

I would like to know too my maf is on the table and I cant find this screw ??

KeithMac
11-24-2010, 07:44 PM
Turn the housing upside down, dig the silicone out of the round bung and you`ll see a round plastic screw head. To be honest all that`s usefull for is calibrating the MAS it idle, not worth messing with.

Have a search for how SnakeSkinner ports the stock MAS, I did the same and it works very well (you need to re-tune for it though), and there`s the Mellon Bypass mod (also worth a search).

I`ve done both the above with good results.

mb7050
11-24-2010, 08:11 PM
I have got an impresion that low load tuning is near impossible with ported stock maf.

GTOJOE
11-24-2010, 09:22 PM
Turn the housing upside down, dig the silicone out of the round bung and you`ll see a round plastic screw head. To be honest all that`s usefull for is calibrating the MAS it idle, not worth messing with.

Have a search for how SnakeSkinner ports the stock MAS, I did the same and it works very well (you need to re-tune for it though), and there`s the Mellon Bypass mod (also worth a search).

I`ve done both the above with good results.

I haven't done this but I can tell you that there is nothing on the airflow sensor or any other part that corresponds to this part of the housing. There is an air gap between the lower part of the housing [with the silicon filled bung on it] and the middle part of the MAF. The air flow sensor only reaches as far as the middle section. So i'm a little confused.

KeithMac
11-25-2010, 06:03 PM
I haven't done this but I can tell you that there is nothing on the airflow sensor or any other part that corresponds to this part of the housing. There is an air gap between the lower part of the housing [with the silicon filled bung on it] and the middle part of the MAF. The air flow sensor only reaches as far as the middle section. So i'm a little confused.

The air travels through every part of the MAS but only the center section is metered, the bung blocks the bottom bypass passage of the MAS, the less blockage (winding the screw out) the more unmetered air enters the engine. Using this passage and screw you can "tune" the MAS housing to a specific hz reading at a specific airflow, calibrating it.

Mb750, as long as you leave the middle screen in the tune is rock solid once set, it doesn`t drift at all. I played around with a completely de-screened one for a while and they don`t like the centre screen removed.

mb7050
11-25-2010, 06:11 PM
so how snakeskinner ports the stock MAS. i canīt find it anymore I remember something from memory but..

KeithMac
11-26-2010, 05:09 PM
Here`s a pic, hope he doesn`t mind.

http://www.dynamicracing.com/mw/images/c/c8/Ported_stock_MAF_002.jpg

mb7050
11-26-2010, 06:04 PM
thanks

MR2
12-04-2011, 04:22 AM
Well, how interesting...

I came up with the same conclusion looking at how to meter the air going into my engine, I also came up with a LOT of people having issues with the GM MAF and ended up going with a pair of these.
VMP Tuning - Buy Ford SCT & GM Custom Handheld Flash Tuners Performance Economy Towing: VMP Blow-through 75MM MAF housing and HPX MAF sensor (http://www.vmptuning.com/store/index.php?p=product&id=234&parent=99)

good for over 650hp each...small enough package that you can fit them just about anywhere AND they seem to be sworn by through out a very large amount of turbo tuning forums.

DG
12-05-2011, 12:24 AM
There is apparently a newer EVO MAF that flows a lot more air and has a wider metering range. That would make an ideal sensor for an AEM or similar.

One thing that nobody has ever done (AFAIK) is to put the OEM MAF on an honest to goodness flow bench and see exactly what the Hz curve looks like and when it tops out. That would be a Good Thing to do.

DG

MR2
12-05-2011, 12:26 AM
Negative, I looked at the Evo MAF's and the EVO X one (I assume your talking about) has been known to blow up ALL THE TIME... guys are now keeping spares in their glove box! seriously!!!

mb3000
12-05-2011, 12:35 AM
Figured I would post up a pic of my old maf.

http://i928.photobucket.com/albums/ad128/mb3000gt_photos/IMG_1007.jpg

Greg E
12-05-2011, 05:32 PM
Negative, I looked at the Evo MAF's and the EVO X one (I assume your talking about) has been known to blow up ALL THE TIME... guys are now keeping spares in their glove box! seriously!!!

I've read about this with the EVOX MAF but never the 8 or 9....

MR2
12-05-2011, 05:39 PM
from reading the limit of the 8 or 9 was only marginally better, been a few months since I looked but only 20-30% better than our own iirc.

you could just cut the whole stock GTO MAF up and cut a hole in a 4" pipe and pop it through you know :p

mb7050
12-05-2011, 05:43 PM
evo 8 maf reads up to 890cfm ~ 3khz IIRC

MR2
12-05-2011, 05:59 PM
maffrequencylimits [ECMTuning - wiki] (http://www.ecmtuning.com/wiki/maffrequencylimits?s=evo) yep

EVO 8 is 890 cfm

3000gt is 650 cfm??

the HPX PMAS:

Tube Size Est. HP range with HPX sensor
3" 300-750RWHP
3.5" 750-900RWHP
4" 1000+RWHP

I decided to just go with 2 x of the 3" in blow through gotta be way more than I'll ever need :) good rep...very reliable etc..(so a great deal of tuner forums say)

in theory 150cfm is 100hp, so....

mb7050
12-05-2011, 06:15 PM
EVO8 (it looks restrictive)

http://members.cox.net/rstrope/IMG_2413.JPG
http://i225.photobucket.com/albums/dd39/PHDinFast/parts%20for%20sale/DSC01867.jpg
http://i7.photobucket.com/albums/y275/Eric21/Album%20Stealth/IMG_0620.jpg

EVO 8 and 9 mafs are physically exactly the same but thy are calibrated a bit differently AFAIK

I recently swapped a Evo 9 MAF onto my 05 evo 8. The car went about .8 leaner overall. My Evo 8 MAF on the Evo 9 caused the 9 to go richer.

MAF scaling and Smoothing tables between a Evo 9 and Evo 8 they are different, with the Evo 8 Scaling and Smoothing with higher numbers through the whole tables. MAF size tables are the same.


3s OEM
http://www.hadesomega.info/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=74990&g2_serialNumber=1

EVO x maf is not a KV unit AFAIK which means it won't work with our ECU easily like the EVO 8 maf does ( sensor can be removed from the housing and the housing can be replaced with a larger housing easily )
OEM EVO X housing
http://farm6.static.flickr.com/5141/5622003000_533b5eae5a.jpg

3.5" aftermarket EVO X housing
http://www.mellonracing.com/images/ETSMaf-Housing-Large.jpg

just a flange
http://www.streettunedmotorsports.com/parts/b/jmf_evo_x_maf_flange.jpg




maffrequencylimits [ECMTuning - wiki] (http://www.ecmtuning.com/wiki/maffrequencylimits)

My Evo 8 Maf conversion! - 3000GT/Stealth International Message Center (http://www.3si.org/forum/f42/my-evo-8-maf-conversion-462718/)

:scared2:
Need Alternative: Evo 8 MAF honeycomb destroyed at 539whp/491wtq - DSM Forums (http://www.dsmtuners.com/forums/tuning-engine-management/216013-need-alternative-evo-8-maf-honeycomb-destroyed-539whp-491wtq.html)

MR2
12-05-2011, 06:21 PM
Ahhhhh yes, it was that last link that made me go elsewhere...I was wondering for a while.

could always run without the honeycomb but the air needs to be realllly straight before going past it...else you get all sorts of shit going wrong (hense the long tubes a lot of aftermarket ones use)

mb7050
12-06-2011, 01:23 PM
Judging by the photos I can't imagine EVO8 maf being much less, if any, restrictive than the 3s OEM maf, the good thing is its ability to read much more accurately at higher cfm of airflow and its able to read more HZ too .

Emilie@GZP
12-06-2011, 02:44 PM
We did some back-to-back dyno pulls, at the same PSI, on the same car, with a stock MAF, and a GM MAF and a MAFT Version 2 installed. The car gained about 40awhp from just changing the MAF


323AWHP


http://www.youtube.com/watch?v=C3vNcsUXJp0

360AWHP


http://www.youtube.com/watch?v=7A68W3ROllg

Jeff V.
12-06-2011, 03:24 PM
How much of that was JUST the MAF, and how much was due to tuning alterations imposed by adding the translator?

mb7050
12-06-2011, 03:35 PM
assuming all things equal I wouldn't be surprised if the 3s OEM maf is really that restrictive..

Emilie@GZP
12-06-2011, 03:37 PM
The first video, he was running an Emanage Blue, the second video, we changed to a GM MAF and Version 2 MAFT. There wasn't too much tweaking done, kept it at the same PSI, and did a rough tune on it, and it put down 360.

Emilie@GZP
12-06-2011, 03:39 PM
assuming all things equal I wouldn't be surprised if the 3s OEM maf is really that restrictive..

We were very surprised at the power difference on the car I posted above. We did not expect that much gain in power from adding the GM MAF.

mb7050
12-06-2011, 03:50 PM
Turbos hate Pre turbo vacuum, same way as boost/Hp can be effectively reduced by a compressor inlet or outlet restrictor, this increases heat as well, a low-restriction intake system will be rewarded with more power and less heat,of course going blow-thru changes the way the MAF reads the air

I would add this info here : you can use the info to estimate the size of MAF pipe and pre turbo piping you need ., second things to consider are of course MAF sensor reading capacity and restrictions inside the MAF unit like the honeycomb etc ...

Fluid dynamics - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Fluid_dynamics)
Speed of sound - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Speed_of_sound)
V = sqrt(k*R*T): k=specific heat ratio, R=gas constant, T=abs temperature



Tube size can easily be checked by calculating the maximum airflow attainable, dividing by the area of the tube in square feet, and dividing again by 60 to convert to feet, per second. An approximate value tor maximum airflow can be obtained by multiplying the desired bhp by 1.5.
Example:
Let power = 400 bhp, for which maximum airflow is approximately 600 cfm, and air tube diameter = 2.5 in. Then


http://img827.imageshack.us/img827/6247/yjuffffffffffffffffffff.jpg

the 2.5-inch-diameter tube will be adequate to flow 600 cfm without unreasonable drag.
Resist the temptation to use larger diameter tubes than necessary, as little drag is created in smooth tubes with gentle bends. Larger tubes will only add to the volume of the IC system, and that is not a good thing to do.


According to Corky Bell, Maximum Boost pg 61, ~~304 MPH or ~~0.4 mach is the point at which airflow meets increased resistance (drag) and flow losses are experienced.

You can use this chart to estimate and plan the size of piping you need. Actual values will vary 'little' depending on temp etc.



0.4 mach = 304 MPH

2"
1.57 x 2 = 3.14 sq in
300 cfm = 156 mph = 0.20 mach
400 cfm = 208 mph = 0.27 mach
500 cfm = 261 mph = 0.34 mach
585 cfm max = 304 mph = 0.40 mach


2.25"
3.9740625 sq in = 1.98703125 x 2
300 cfm = 123 mph = 0.16 mach
400 cfm = 164 mph = 0.21 mach
500 cfm = 205 mph = 0.26 mach
600 cfm = 247 mph = 0.32 mach
700 cfm = 288 mph = 0.37 mach
740 cfm max = 304 mph = 0.40 mach


2.5"
4.90625 sq in = 2.453125 x 2
300 cfm = 100 mph = 0.13 mach
400 cfm = 133 mph = 0.17 mach
500 cfm = 166 mph = 0.21 mach
600 cfm = 200 mph = 0.26 mach
700 cfm = 233 mph = 0.30 mach
800 cfm = 266 mph = 0.34 mach
900 cfm = 300 mph = 0.39 mach
913 cfm max = 304 mph = 0.40 mach


2.75"
5.9365625 sq in = 2.96828125 x 2
300 cfm = 82 mph = 0.10 mach
400 cfm = 110 mph = 0.14 mach
500 cfm = 137 mph = 0.17 mach
600 cfm = 165 mph = 0.21 mach
700 cfm = 192 mph = 0.25 mach
800 cfm = 220 mph = 0.28 mach
900 cfm = 248 mph = 0.32 mach
1000 cfm = 275 mph = 0.36 mach
1100 cfm max = 303 mph = 0.40 mach


3.0"
7.065 sq in = 3.5325 x 2
300 cfm = 69 mph = 0.09 mach
400 cfm = 92 mph = 0.12 mach
500 cfm = 115 mph = 0.15 mach
600 cfm = 138 mph = 0.18 mach
700 cfm = 162 mph = 0.21 mach
800 cfm = 185 mph = 0.24 mach
900 cfm = 208 mph = 0.27 mach
1000 cfm = 231 mph = 0.30 mach
1100 cfm = 254 cfm = 0.33 mach
1200 cfm = 277 mph = 0.36 mach
1300 cfm max= 301 mph = 0.39 mach



and here you can look how much CFM your turbos will flow at xx amount of psi.
Stealth 316 - Turbocharger Compressor Flow Maps (http://www.stealth316.com/2-3s-compflowmaps.htm)

http://www.3sgto.org/f13/smic-vs-fmic-3601-18.html

MR2
12-06-2011, 04:33 PM
yeah, I would have preferred twin 2.5" Blow through Maf over the 3" Blow through Maf, but 3" is the smallest they make :|

DG
12-06-2011, 04:44 PM
Judging by the photos I can't imagine EVO8 maf being much less, if any, restrictive than the 3s OEM maf, the good thing is its ability to read much more accurately at higher cfm of airflow and its able to read more HZ too .

I'd still like to see independant confirmation via a flow bench and an ocilliscope. Too much speculation, too little hard data.

DG

mb7050
12-06-2011, 04:45 PM
yeah, I would have preferred twin 2.5" Blow through Maf over the 3" Blow through Maf, but 3" is the smallest they make :|

then this is good to take into consideration ;)



http://img708.imageshack.us/img708/5483/aaaaaaaaaaag.jpg
A cone angle greater than 15° can cause disruption of the air stream's boundary layer, increasing drag.

mb7050
12-06-2011, 04:53 PM
I'd still like to see independant confirmation via a flow bench and an ocilliscope. Too much speculation, too little hard data.

DG

Me too.

mb7050
12-10-2011, 06:45 PM
I'm hoping I'm wrong about the EVO 8 maf being any less restrictive than the OEM. It sounds like a good upgrade especially with a flashable ECU..

I have been trying to find a bigger KV maf without any luck

Greg E
12-11-2011, 02:42 AM
I'm hoping I'm wrong about the EVO 8 maf being any less restrictive than the OEM. It sounds like a good upgrade especially with a flashable ECU..

I have been trying to find a bigger KV maf without any luck

Same here. The EVO8/9 was the best I could find. :(

KeithMac
12-11-2011, 03:57 PM
Probably all you can find.. Looking at the Evo MAF and what you have to do to fit it, it`s deffinitely more restrictive than the 3s MAF in my eyes.

Jeff V.
12-12-2011, 03:02 PM
Probably all you can find.. Looking at the Evo MAF and what you have to do to fit it, it`s deffinitely more restrictive than the 3s MAF in my eyes.

You're basing this on what, some photos you saw on the internet?

It's got a higher flow rate than the 3S MAF (288 vs 358 g/sec off the top of my head), and according to calibration data one of the DSM guys has, it has a lower pressure drop than a 3S MAF.

If you're someone who cares about something besides WOT, the fact that it's got better resolution down low and stays accurate into higher frequencies it nice too.

More accurate air metering = better fuel economy = smoother around-town drivability.

Or you can just stick a giant piece of Home Depot-spec plumbing on your intake and tune WOT with some knobs. It's your car.

MR2
12-12-2011, 04:21 PM
Jeff's post is unfortunatly quite correct.

when the tuners went to work on my car they discovered that the Y pipe made for my car (by them, without my guidance) actually had very poor stability especially with low airflow, so at this stage I cannot recommend the airflow meter I got.

keep in mind that the idiots did not use the pipes that VMP supplied, which are designed to make the airflow as laminar as possible, I Will be working on improving this situation but alas...as Jeff said, low air speed is very important.

if I cannot get the low air speed correct I'm contemplating seeing if the ECU guys can use MAP below a certain engine speed/CFM

mb7050
12-12-2011, 04:43 PM
are there any fins inside the VMP pipe ? you could use some to accelerate/smoothen out the flow

mb7050
12-12-2011, 05:31 PM
A quick googling revealed me that there are none? if the tube length and the width is the same they should flow the same as you know, you would want to locate the sensor as far away from the y joint/bends as possible, That turbulence at the maf opening can really mess up the air flow readings ..

Certainly a badly made y pipe can cause a lot of turbulence same way as any change in flow direction(bends)
http://www.appliedmc.com/images/e-zine/flow-through-elbow.jpg
http://www.cc.utsunomiya-u.ac.jp/~sugiyama/avs14/swirl/W.gif

Any pics of that y pipe they made for you ?

MR2
12-12-2011, 05:46 PM
check this shit

The useless banter thread - Page 69 - 3000GT/Stealth International Message Center (http://www.3si.org/forum/f155/useless-banter-thread-495646/index69.html#post6327073)

this one you've seen before...
http://3000gt.com.au/car/Misc/Pulse/DSC00760.JPG
how it looks on the inside more or less.
http://3000gt.com.au/car/Misc/Pulse/ypipeinternal.JPG

showing the actual inside so you can see the joins
90 degree lol
http://3000gt.com.au/car/Misc/Pulse/shitpipe%20(3).JPG
sharp edge
http://3000gt.com.au/car/Misc/Pulse/shitpipe%20(6).JPG
the crapped up Y pipe join, you can see the sensor in the pipe...straight in front of the worst possible spot.
http://3000gt.com.au/car/Misc/Pulse/shitpipe%20(7).JPG
http://3000gt.com.au/car/Misc/Pulse/shitpipe%20(8).JPG



keep in mind I have no battery beside the 90 degree here as well, so they could have done whatever smooth elbow they wanted :|

Chris@Rvengeperformance
12-14-2011, 11:06 AM
that bit of weld penetrating through the inside of the pipe isn't going to cost you any real power.

Ray has a baseball size dent in his intercooler pipes and doesn't even bother to fix it with over 1000AWHP.

I've experienced great driveability with a 3.75" LS6 maf.

MR2
12-14-2011, 03:25 PM
It's not the power I'm worried about, this thing is stock...

it's the fact that all these protrusions and irregularities will completely screw up the airflow just when it needs to be laminar to provide the ECU with a clean signal showing how much air the car is injesting

Chris@Rvengeperformance
12-14-2011, 05:53 PM
Maf's aren't designed to run in blow through anyway. If you are really worried about "laminar" flow then use a draw through setup with the stock intake bubble.

MR2
12-14-2011, 06:21 PM
or I could just get it right :) loads of people are running MAF blow through.