P J Moran

There are six, in my case, and each is apparently just an extension of the previous in terms of counts and flow. The first count value on the second table is the same as the last count value on the first table, and so forth.

Assuming they just effectively broke up one long table into six, shouldn't I expect the flow values to be the same in each table for those count values that overlap (are the same)? They're close, but not quite the same.

------------------
'87 IROC 5.7l TPI - original owner!

[This message has been edited by P J Moran (edited January 24, 2001).]

88GTASTK

Also on the MAF tables. I downloaded a couple from the BIN Archive. The values went from medium at the top of the table to low at the middle to high at the bottom. Is this bin not right or is this how they should be ?

I'm also guessing that I should read the table as, the MAF tells the ECM it picked up so many counts, the ECM looks at the table and sees that many counts equals the grams/sec at that spot on the table.
Or do I have this backwards.

Thnaks
Paul

doc

I've seen three .bins for my '87 MAF car.

All have 6 MAF tables where there are two columns of data, counts and gm/sec.

How does the ECM use these tables?

What is the significance of these tables?



------------------
'87 L98 TPI IROCZ, AFR190 heads, 3.70 gears, ZZ9 cam, 2400 Art Carr TC, SLP 1 3/4" headers, SLP cat-back, no cat, no AC, no MAF screens, Accel manifold base & SuperRam, Edelbrock double roller timing chain, MSD ext coil & distributor, trans cooler, 52mm TB with airfoil, TB coolant bypass, AFPR(58psi), K&N filters, Hotchkis lowering springs, Car-Pro custom chip, Lay Ind. ram air kit, 265/45R16 Kumho V700 tires.
ET 13.39sec @ 107.21mph
'90 Eagle Talon AWD, no rust thru 9 winters
'99 Camaro SS, red, 6-spd, T-tops, Mcleod clutch, Pro 5.0 shifter, MAF processor, Direct Flow airlid, K&N filter.
313.7Hp & 320.6ft-lbf, ET 13.55sec @ 105.1mph
next mod: 4.10 rear gears

Scott 88 GTA

I'm just getting into this myself, but this is how I understand it so far:

The counts in these tables are the actual signal that the ECM gets from the MAF sensor. The ECM then takes this value and goes to the MAF table to find out how many grams/sec of air are flowing into the engine. I guess you could call these tables the calibration tables for the MAF. Based on this fact, the gm/sec should increase progressively with the # of counts, if it does not, something is wrong. Also, on the overlapping values, the ones in the higher MAF tables do not match in the ARAP bin, and none of them match in my stock bin. Why, I do not know. Anyone???

Once the ECM has the value for the amount of airflow into the engine, it then uses this to determine several other things, such as injector pulse width and spark advance. Any table or array that uses the "load" on the engine is based on this. The amount of air entering the engine is considered the load on the engine.

The thing that I find interesting is that no where in the ARAP bin is there a table that calculates the actual pulse width time of the injectors, only tables that modify the A/F ratio by percentage. Is the PW calculated by an equation not seen or something? And modifying the A/F ratio through theses other tables the only way to richen or lean the mixture? Does anyone have some knowledge of this?

------------------
Black 88 GTA L98
261 RWHP, 345 RWTQ
ZZ4 bottom end, Edelbrock 6085 heads, LT4 HOT cam, GMPP 1.6 RR's, ported stock TPI, SLP 1 3/4" headers, no cat, Dynomax cat-back, Stock PROM
E.T.F.A Member #11

Guest

I dont know much about the ARAP bin. I havent really spent any time reading the information on it to try and understand what they did. Ive also been around the earlier MAF cars so much I know how they act, and gained a pretty good idea of how it operated without knowing anything from a numbers and bin standpoint.

As for the MAF tables... I see someone finally has it right. Ive been saying this for a while. Its a calibration table to convert the signal from the MAF (whatever that is) to a useable number for the ECM to perform calculations from. If you were to, say, take the sensing element out of the MAF and stick it in a larger tube, the actual airflow would be the same but the ECM would interpret it as less since there is not as much air passing the hot wire to cool it off. So all you have to do then is change the calibration table.
Why am I so convinced? There is no other table or spot I can find in the bin that attempts to convert the MAF information to the gm/sec that is used elsewhere in the bin for calculations.

Code 33 is also a dead issue, there is a table in the bin that has basically base settings table for what it *thinks* the MAF is supposed to be reading (in gm/sec I might add) and when you add a cam that drops vacuum at idle and low rpm, it doesnt match the parameter and variance established by the base settings table, and throws a pretty orange light in your face.

Its funny to me how much people run from the MAF to SD, but I think they just dont understand it and dont want to.

P.S.: Know about the 255 gm/sec max reading? Almost every table has a multiplier somewhere. 510 looks like a nice number to me.

[This message has been edited by madmax (edited February 12, 2001).]

87Z-ya

iTrader: (1)

Hey get a fire going on the maf ecm's. Maybe they will start to work on it some more.

Grim Reaper

Max, I think the major reason that people convert from MAF to SD may be because, until now, there hasn't been enough guys willing to devote the time into it to learn it. I wish I had access to a MAF car locally, so I too could dive into it and see "what makes it tick" and offer more assistance.

But I think you guys are getting closer. I also recommend looking at TunerCat's TDF Editor if you are using TunerCat. Even though the 8D TDF is quite well defined, it still does not have "everything" in it, as I just discovered with the "MAT Delta Multiplier Lookup Inverse Table". I found this table is key, if you want to use a relocated MAT and have the ecm calculate the proper injector pulse width.

I suspect there are a lot of "hidden tables" that are not yet defined for the MAF cars, that are waiting to be discovered and tested. Don't give up.

Guest

I completely skipped that. I have the hack file and have been looking through that. Maybe someday I'll toy around with the tunercat TDF stuff. As for now I just edit the bin file with the editor that came with my burner. I have a pretty good idea whats in there anyway. GM wasnt real fancy with making the system work.

Grim Reaper

Yeah, GM went for "reliability with minimum warranty claims"; definitely NOT "maximum performance". Guess if I was in there shoes, I'd probably do the same.

F22Raptor

Glenn, move to San Antonio texas....My car is all yours for research purposes.
Rick

------------------
'88 GTA 350
Headers, catback, free mods, Xtreme energy cam, intake pieces and 2100 RPM TQ

13.39@102.50 MPH
1.90 Sec ------- Best 60FT
104.23 MPH -- Best Trap Speed
AIM HIGH!

88IROCs

Been studying the $32 hack for about three month's, so there's a few things I can add about the MAF tables.

1. The ECM reads an output voltage from the MAF sensor, in the range of 0.18 - 5.12 VDC. It looks up this voltage in the MAF tables, finds the corresponding FCB value and divides that value by the table scalar value to determine airflow(gms/sec). Unfortunately, table six cannot have a table scalar of less than 1(it could but doing so would decrease resolution. this would mean at high airflow rates there would be increased distance between steps -table rows-, which would lead to leaning of the mixture between steps and deviations to the BLM's and INT's). Still, you might think that having a minimum table scalar of 1 is not a limitation. It is! Because the ECM uses 8-bit addressing, it can only recognize values in the range 0 - 255. This means the maximum allowable FCB value is 255, and because the table six scalar is 1, the maximum accepted airflow value is 255 gms/sec. 255 gms/sec equates to roughly 506 cfm, and when you think about it, this amount of flow would require WOT on a stock TPI engine, and at WOT the ECM no longer uses the MAF tables.

If the ECM had used 16-bit addressing, it would be able to recognize the MAF's theoretical max ouput of 5.12 VDC. This would yield a FCB and gms/sec value of 277. Which equates to roughly 549 cfm. Not a significant gain(only an 8% increase), and again unnecessary at WOT.

2. The six tables are seperate entities, they are not one long table broken into six parts. The tables decrease in resolution as airflow increases, this is so that the ECM can make fine adjustments at low airflow readings(nice stable idle, good response to changing airflow, good economy) as it tries to maintain it's desired air/fuel ratio. The reason the first airflow value in a table is the same as the last value in the preceding table is: to prevent unnecessary stepping between tables, and to let the ECM know exactly where to look in the next table as the MAF's voltage moves outside of the range of it's current table. Keep in mind, the MAF TPI ECM was designed in 1983 - 84, and that typical CPU and memory speeds were not as quick as today's hardware(in 1984 if you had a CPU that ran at 8 MHz and it used 250 ns memory, you were considered to be 'bleeding edge'). Add to that the fact Motorola processors were known more for their durability than their speed, and you get an idea of why you would want to limit the number of 'decisions' the ECM had to make in a given time period.

Even though the VDC will remain constant in adjacent rows of neighbouring tables, the FCB and table scalar values must change to allow increasing airflow in ascending tables. The important thing is for the ECM to see the VDC identical value as it moves from one table to the next. Changing the scalar and FCB values makes it nearly impossible to achieve the same flow(gms/sec) for identical VDC. In reality, the differences are less than 1 gm/sec, which is an insignificant amount for the purpose of the ECM's calculations.

3. The ECM uses a base injector pulsewidth based on the known flow rate of the injectors(19 lb/hr for LB9's, 22 lb/hr for L98's) to calculate fuel flow. This base pulsewidth can then be modified to achieve the desired air/fuel(0₂ sensor) ratio based on the measured airflow(MAF). Additionally, other inputs(TPS-accel enrich, CTS-cold start and engine warm-up, engine speed, etc.,...) can change the final injector open time. However all of this is based on having an accurate value of fuel flow. Increasing fuel pressure or using 'bigger' injectors will throw-off the fuel flow calc's and cause the ECM to make corrections for the unpredicted enrichment as it tries to achieve it's desired air/fuel ratio. This should cause changes in the BLM and INT, as the ECM corrects(INT) and stores(BLM) the corrections.

Finally(whew ), I wanted to post the MAF tables, to make some of what I said clearer. Unfortunately, to conserve space I made them into a 6 on 1 nested table. However, the last time I tried to post the nested tables, the thread crashed(would no longer load). If anyone knows how to post nested tables in a UBB thread(without crashing the thread), please let me know!

[This message has been edited by 88IROCs (edited February 13, 2001).]

88GTASTK

That's the answer that I've been waiting for. Thanks for the details 88IROC. My one simple question is what does FCB stand for.

Paul

Grim Reaper

If I recall correctly, FCB = Fixed Constant Byte.

Grumpy

On the MAF to SD:
The MAP system allows for a MAP AE, which means lot during trasistions. The MAP is an instantaneous read out of what the engine is actually doing.

The MAFs take time for the element to change temp.. The newer ones are light years ahead of the 3rd gen MAFs.

If you pegging the MAF (not too hard to do), then you've run out of resolution. You can cruthch things from there, but why when for 50-60 bucks you can get everything right the SD.

Guest

Yea, what Grumpy said about the MAF wire is right... old technology. The newer Wells design eliminates that, but I dont know about trying to take the electronics package out of that and have it still work, and then you are still limited without trying to change the resolution. I know there is a way, the Ford setup is the same and someone found a way to double the scale. It looks feasible if you find the right spot/spots to change.

88IROCs

I don't think you can change the recognized scale without changing the computed values to something greater than grams/sec. And of course, because there is a finite number of data points, going to larger values would decrease resolution(however I have some interesting ideas brewing on that subj). However, I think grams/sec is programmed into the microcode, and changing to a larger value would require a new .ecu program.

However, there is enough room in the MAF tables to recalibrate for increased resolution within specific airflow ranges. I am playing with a multiple .bin idea, and this would allow me to calibrate the MAF table resolutions to account for different operating conditions(i.e., highway driving equals low rpm amd mid airflow, city driving equals mid rpm and low airflow, etc.,...) From that perpective, re-ordering the MAF tables might be useful). It would be nice to have a device that could automatically switch between .bins based on rpm and vehicle speed.

Grumpy,

I am slowly coming around to your way of thinking, in that SD is an easier system to work with. However, as I still have an investment in MAF(three vehicles with '88 TPI's), I plan on learning(dealing with) this system for now(not to mention I can be quite pig-headed when it comes to making a change for the better). Plus, because I am less interested in performance gains than I am interested in improving all-around drivability and efficiency, sticking with MAF is an acceptable method - for now.

Guest

Switching bins is an interesting thought... could be pretty complicated to set that up though.
Ill have to dig some more I guess.

88IROCs

kind of an interesting development,...

got to page 243 of the $32 hack(bua_hac.bin), and according to the info there, the ~165 ecm was designed to recognize the output of Bosch FM digital MAF's. Now this probably refers to the old very restrictive 'flapper' MAF's. But the fact that it can accept freq's from 32 Hz to 65 536 Hz(64K), kinda makes me wonder if a newer digital MAF could be adapted to work with this system?

this is getting interesting,...

branz28

Yea, keep going, heh, this is kinda like a movie at the suspense scene...(sorry it's late saturday night)..

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

Grumpy

Trust me when I say I'm not biased toward one or the other, my GN is MAF, and I spent several hours as of last night tinkering with a MAF system

A friend has a 88 Bonnie and we changed it to a 148 and used the GN MAF and are trying to make a serious N/A machine. Ya kinda odd but that's what we do around here

I still can't make sense of how the darn tables are laid out.

For instantace moved the whole scale down by one entry and the car wouldn't run. Thinkering with the WOT AFR and injector size at least got it runing rather well, but still not right.

I've read numerous explainations of how the caler works, but as of yet, none pass the well it works and the car acts right.

FWIW

Grumpy

One other thought is according to Ludis the 747 has a MAF input.

Maybe someone could follow that up with him, and apply the 747 to a MAF system. Now that might be interesting.

branz28

Geez, why can't ya'll live in my neighborhood...I wouldn't mind being a lab rat, heh... The MAF is my concern, i'd like to keep it but with a more....i guess the proper word would be efficient? ECM.

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

GTARob

I am in awe...........

I don't understand most of it, but I am just in awe..............(jaw drops)

------------------
1989 GTA TRANS AM
Was a TPI 305 & t-5 car.
Is now 383 stroker w/ the same T-5
World products Sportsman II heads. Lunati pistons and custom ground camshaft. Will dyno after the break in but is estimated at 450 hp and 520 ftlbs. The T-5 is on a suicide mission.

Grumpy

I *think* an option for using the frquency Bosch units like the GNs, I haven't seen any flapper MAFs in the GM world.

As far as the Ford sensors go there are a real nightmare.

I've been having a number of discussions with guys trying to run the Fords MAFs on other then Ford Applications, and to say the least they are total nightmares.

They use a small sampling area, and any turbulence throws them way outta calibration.

The only saving grace about them is the learning ability of the Ford ecm. The instant you surpass that your in dodo.

As, an example on Porsche guy lost 130 HP going to the Ford setup from a non adjustable porsche unit. An this is with bank rolling alot of dyno time. The MAF would just go screwy at the higher flow rates. Not to say it can't be done, cause there are plenty of fast mustangs, just it may take a ton of work to get there.

If anyone really cares try looking at the guts of a late LT1/LS1, Ford, FBod, late S/C MAF and you can get a real grasp on the stratigies being used

Grumpy

After spending some time last night looking at some stuff with Tunercat it's easy to read what the Gram/sec, count rates are.
Still seems awkward, but sure demystifies things.

Still amazing what ya see when ya look LOL

CanadianBeast

88 iroc you say that the values in the maf tables should be the same, ie the last in table #1 the same as the first in table #2. But why then are they all different in the bins I've been looking at that I downloaded from diy-efi???
eg) table#1 volts=1.46, 249grams/s
table#2 volts=1.46, 114grams/s
it doesn't make sense for a value to be different in two tables at the same maf voltage.

88IROCs

The last row of values in a table should be equal to the first row of values in the next table. The only value that will change is FCB, as this value modified by the table scalar tells the ECM how much airflow is being detected.

Which .bin are you looking at? 249 gms/s is too high a value for the first table, which should max out at around 22.0 gms/s. If you went to a max value of 249 gms/s in the first table, you would have no need of the other five tables(Remember the max value in table six cannot exceed 255 gms/s).

You need to look at the .bin again, or find a good .bin to look at.

88_vette

O.K guys,

This is my first post, and I plan on burning chips in the future. I have a 88 Corvette with a MAF (which has just gone bad) we've all been there. So, about the MAF table

As I understand it the MAF has a voltage sent to it to keep the wire heated to a set temp? The more air that passes through the MAF the cooler this wire will get, thus causing the ECM to supply more voltage to heat the wire. This is how the ECM knows how much air is being pulled in by the engine? If this is the case than why would you want to change the MAF tables to start with? If the stock setting is already telling the MAF how much air is moving through the MAF than why change? You can mod the MAF as I'm sure most of you know to flow more air, but you can not mod the chip too allow the MAF to flow more than it will allow. The ECM also uses this DATA, and other sensors to figure out fuel delivery. Couldn't you just change your injector open time some how during the time the MAF reads a set number of GPS.

Example. If the stock MAF values tell the ECM the engine is flowing 300GPS then that is what the engine is flowing. In order to flow this the throttle would have to be open at X amount, which puts the TPS%/ postion at X. This would be a constent in the terms that everytime you open the throttle to the same X the TPS% would be @ the same X as before. Meaning the MAF is flowing the same amount of air as before +/- a few GPS. So, can't the injector open time be set based off RPM, TPS%, or MAF flow?

So, am I anywhere close to being right? If not I guess I best hit the books again. Thanks for the input guys.

------------------
88 Corvette
383 Waitting to dyno

CanadianBeast

Im looking at the Arap.bin from the diy-efi bin library. I also have a few other bins from that same library and they look the same. I don't understand it. What youre saying must be right but I wonder why they look like that???

branz28

I am under the same impression that 88_vette is. Someone want to elaborate more? I'm just really interested in what those tables do..Since i'll soon be getting into prom burning...

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

88GTASTK

If I get what you're saying there I don't think you're looking at engine air requirements quite right. I know that's there's someone else out there that can explain this better but I'll try. The air flow is going to vary with the load that is placed on the engine at a given TPS%. From what you've got posted above cruise control would be just keeping the throttle in the same place. You're thinking in a 2 dimensional plane rather than a three. If I'm out in left field someone please correct me.

Also on the MAF tables looking screwy - I just downloaded a '86 305 bin with my Pocketprogrammer and looked at it with Tunercat and a $32 tdf and everything looks correct. I'm wondering if one of the $32 with the extensions are needed for the newer year MAF bins. ie. only the '86 can be read correctly with the $32.

Paul

Red Devil

Now, I'm not the brightest bulb in the pack, but does this mean I could, with some experimenting, burn a chip that would recognise a newer digital MAF from a 4th gen using this ECM?



------------------
"A lawyer with his briefcase can steal more than a hundred men with guns."

-Mario Puzo

branz28

Gosh, now you got me wanting to do some research when i've done to much already (at least tonight).

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

MadMax350

I have done nothing but play with the 7165 system. The MAF tables are nothing more than your basic fuel map for the engine (simply put). Of coarse, you are correct in that once the ecm reaches the 255 g/sec boundry in terms of airflow, it will no longer register more airflow. However, something that needs to be considered here is this: At the point the engine reaches 500cfm or so of flow is the point in which volumetric efficiency is probably starting to drop off. Now, I know that the engine will probably take in more air past this point, but with increased RPMs comes increased injector pulses. While running my engine on the dyno, I experienced virtually no "leaning" out of the A/F mixture past the point in which I reached the 255 barrier (which I think was around 4300 RPM).

Getting to the late model MAF sensor issue: While going thru the hack for the 7165 ecm, I found a switch that selected between analog and digital maf sensor in use. I added this switch to my .tdf file in tunercat and tried to run an LT1 maf on my car. With the key on the airflow started out at 255, and once the car was started, dropped to about 181 (gm/sec) at idle. I went through the MAF tables and reversed all values and basically started in MAF table 5 where the 181 value was and changed it to 10 (gm/sec) and went backwards from there. I actually got this new style MAF to work with my car until it hit about 40 gm/sec airflow and the ecm wouldn't read it any farther. This may be due to the fact that I didn't change the scalor values in the maf tables or just the fact that the MAF sensor was transmitting a frequency to the ecm that it could no longer interperate. I will be attempting to make this thing work without a MAF translator and will keep all of you posted on my results.

------------------
1987 Trans Am GTA WS6
-5.7L SuperRam
-TFS Twisted Wedge Heads
-224/230 Xtreme Roller
-LT1 PM Rods
-4L60-E Transmission
-LS1 Aluminum Driveshaft
-3.73 SRD

1988 Pontiac Grand Am
-3800 SFI V6
-125-C Trans
-Stock emissions equipment
Domestic Terrors Car Club Online

branz28

Woah, now that's what i was anxious to hear. Keep us updated on what you find, MadMax350.

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Gutted cat
-!AIR

Grumpy

Bosch frequency range 34-130 Hz
LT1 LS1 1K to 12 **KHZ**
The older ecm can't even see things in the new range electronically. Meaning you have to make a hard wire redesign to change ranges, ie you'll have to run some sort of translator.

MadMax350

Out of curiosity, what did the old AC Delco 3 and 5 wire digital MAFs transmit signal wise? Perhaps I could construct an inexpensive MAF translator if needs be. However, the Bosch sensor without screens flows at 658 CFM, so I really see no need to put anything bigger on if the cost is going to be great compaired to the 5 or so HP gain I might see with a bigger MAF.

Red Devil

MM350, have you tried one of the older GM MAFS? If that works, I assume you can use all THE GN guys stuff with the newer MAFS. I think alot of us are awaiting results to start playing with the MAF cars. This would be great!

branz28

I think Red summed up my anxiousness to hear the results. I really wish i had all the stuff here so i could jump in but nothing has gotten here yet

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Gutted cat
-!AIR

88TPI406GTA

Just curious as to the outcome of this thread and see if any new info on this. Any new options here?

I am currently exploring the option of running two MAF sensors (one used and one dead for the same restriction) and then scaling the tables in the Bin to accomodate 2x fuel for the signal (because you are actually reading only 1/2 the actual air flow...

junkcltr

iTrader: (1)

I don't think you can take the outputs from 2 MAFs and hook them together properly. Wouldn't you need an adder circuit. Something to think about.........it could get costly if you start smoking 2 MAFs at a wack.
J

88TPI406GTA

Actually, check out this thread to see what I was referring to...

http://209.185.240.250/cgi-bin/linkr...goto%3dnewpost

I think this would be an option, however, I think the easier way to go it is to just add more fuel via the PE tables at higher RPMs. It may take some time to redo the tables to make it work, but I think that it should theoretically work.

I just like the idea of being able to ACTUALLY measure the airflow instead of going SD...but I probably will do that in the future if I can't figure out a fix.

88TPI406GTA

Here is the text of my idea from the other post (I can't get the link to work right now)

"My suggestion is exactly what Smokin87 suggested. My father is an electrical engineer and works with sensor inputs all the time.

Here is what it would be. Simply have an air filter, then branch the air inlet into two tubes with two MAF sensors...one is the active one and the other is not used (electronically) but is merely there to have ensure the same airflow as the functional one and then join the tubes back together again before the TB.

Then go in and modify the MAF scalar tables so that 255 gm/sec is now 128 gm/sec...effectively halve the signal from the MAF that is plugged in. So what would happen is that you would never have to worry about exceeding the 255 gm/sec limit of the MAF sensor...now when it was reading 255, the airflow would be double that or 510 gm/sec. There would be no circuit needed to mod the signal from the MAF sensor as the interpretation of the sensor counts would be done in the MAF table. It still would be reading the airflow, but only half of it...that is the reason for having another dummy sensor, to ensure it pulled EXACTLY the same amount of air through the other tube.

The downside to this is that the "resolution" of the sensor would be a bit grainier and each "step" in the sensor counts would be not as fine...but it still should be able to run effectively. One is effictively scaling the input from the sensor. This is what the stock tables do anyway, so this concept isn't anything too radical.

I hope this is making sense to others as well. I am sure that one would have to also tweak the fuel tables and the LV8 tables as well to compensate for twice the air flow.

If anyone has a reason why this WOULDN"T work, please state it...This is theory, of course, but it could work if you adjusted the tables accordingly. The downside is that one would need to tweak each table accordingly and that would take some time. I am not ready to go MAP yet "

junkcltr

iTrader: (1)

Yeah, that would work. It just isn't as accurate as using the sum/2 of the two MAFs. You loose a factor of 2 in resolution. Also, you are assuming that the air splits evenly between the two...in real life it doesn't. That is what the MAF fudge factors(scale tables) are for. They fix up the slight errors. Well, that is my interpretation of them. But yes, what you stated would work.
J

88TPI406GTA

My thought on ensuring the same airflow would be to utilize the same restriction through each...That would be the concern. Why again would it not pull air through each evenly if the "restriction" was the same and they are both hooked up to the same single inlet?

The real key would be to ensure that the incoming air supply is equal to both and they the airstream hooks back together equally. Hypothetically, the pull through each sensor would be the same if the airstream is the same.

Grumpy

For openers your ignoring all sorts of stuff.
the bondary layer changes as function of air flow, the design of the whole intact tract has peaks and vallies for when air flow is optimim. then there is reversion, PCV air flow, and how that changes as a function of engine load.

The actual output of the MAF per grams per sec on a ecm, verus is a car are very different. It's really a huge fudge factor. Saying a MAF directly measures air flow is wrong. The ecm calculates the air flow from the MAFs output. All the factors I mention above plus even more engine dynamics just throw a curve into it all.

For the work you mention of trying to patch a MAF system together, you could switch to a MAP system and be up and running in just a fraction of the time.

Plus MAP sensors are a whole lot cheaper to replace when they fail.

junkcltr

iTrader: (1)

That sounds OK. I guess just try to keep the MAFs away from sharp bends. Air does funny things at bends

89gta383

iTrader: (5)

Excellent topic. As an owner of a MAF car, I'd like to get started on some alternatives also.

But someone tell me this, why are we talking about the 255 airflow limit if at WOT the ECM stops reading the MAF and reverts to lookup tables?

Am I totally and completely wrong here? I understand part throttle operation, but when you hit a certain %throttle, doesn't the ECM read the PE vs A/F vs (can't remember) table?

88TPI406GTA

The ECM never stops reading the MAF sensor...the issue is that the limit of the MAF readings appear to be 255 gm/sec. However, with some other research I have found (along with Grumpy's accurate statements) the issue is that the MAF really doesn't measure true airflow as far as the ECM is concerned. The ECM does use the MAF signal to calculate the LV8 (load) variable, however, apparently there is a fudge factor (roughly 1.05) thrown into the calculation as well.

The real issue is when the airflow maxes out the MAF and the readings stay at a constant 255 gm/sec at higher RPMS. In other words, if you max out your MAF at 5,000 and yet you are making power to 6000 rpm, then you have a problem.

Glenn found that by increasing the fuel amount in the "PE % at WOT vs RPM table (not exact name) then you can manually add fuel at the higher RPM's and basically get it to work still. That is an option as well.

I guess I am a bit disheartened as my research has indicated that the ECM uses the MAF signal as part of a biased calculation rather than just using the pure signal. In that case, there really isn't true airflow measurement done, although it is still more accurate than a MAP setup (for actual airflow measurement), because a MAP setup doesn't measure airflow at all.

Given the nature of time vs money vs PITA factor, I am going to start researching the $8D (MAP) setup in the next few months. I still need to drive my car for a while with the MAF setup and see if I am maxing it out yet. That is the real key here.

formula5

I think that madmax is on point..

Quote from MadMax350 ,,"I have done nothing but play with the 7165 system. The MAF tables are nothing more than your basic fuel map for the engine (simply put). Of coarse, you are correct in that once the ecm reaches the 255 g/sec boundry in terms of airflow, it will no longer register more airflow. However, something that needs to be considered here is this: At the point the engine reaches 500cfm or so of flow is the point in which volumetric efficiency is probably starting to drop off. Now, I know that the engine will probably take in more air past this point, but with increased RPMs comes increased injector pulses. While running my engine on the dyno, I experienced virtually no "leaning" out of the A/F mixture past the point in which I reached the 255 barrier (which I think was around 4300 RPM)."

That sounds like hard core experience to me.But to dig a little deeper.The reason there is virually no leaning is the injectors double fire.Kind of like a mutiple spark ignition.Tunercat told me that the maf tables are calibrated for the stock maf.They should not be touched under most conditions.But if this is true then my screens out,conical filter and custom ducting will effect the sensor.not to mention my ported intake. All I know is I upped all 6 tables a bit and could feel my engine wake up under all loads.It seems like there is a time multiplier to the maf tables.Maybe like how many sec it takes to hit 255.Spread out over 6 tables.
I can say that each table should start at the end of the last table and so forth.I have tried them off a bit and my car hardly ran.