Grumpy

Both accomplice the same thing.
There are some wifes's tales that continue to be repeated, hopefully some will die after this.

MAF:
Uses a heated wire, or foil that a is maintained at a constant temp., the amount of current used is an indicator of how much air is flowing past it.

MAP:
Uses a diapham, mounted as a strain guage in a Whetstone bridge, to generate a signal referenced to a vacuum. So any variance in pressure generates a variable voltage, as referenced to an absolute.

Both, with only compensate to limited degee for any programming errors. There is no better of the two for those errors, they both have a fixed range of BLMs.

The MAF system *tends* to be a little more forgiving in some mild applications because the sensor is remotely mounted, and hence mechanically dampened. Whereas the MAP is plumbed right to the plenum, so it sees any modifications with less dampening. The lack of dampening also means, more responsive, you'll never get the absolute throttle response of a MAP system with a MAF.

There will ALLOWS be some lag time when using a MAF sensor, because of the thermal lag in the actual sensing wire/foil. You can shrink the size of the sensor, like in the newer ones like the LT1 / LS1 but you can't eliminate it.

In a steady state condition the MAF will develope a better average signal then MAP. Again, it's about mounting, and dampening.

Since the MAF is pre TB, air filter ducting changes effect how it responds. Also, if major can throw off the MAF tables. removing the screens, does nothing but induce errors into the system. Sometimes they might help, sometimes might hurt. Sometimes help in one place and hurt in another. Best way to eliminate guessing what is going on is to leave them alone. If you look into a LT1/LS1 MAF you'll note two (well actually 3) sensors. 2 are mounted on opposite sides of an airflow, to help eliminate any sensing errors, these two are averaged. Now these are the state of the art ones, and GM has taken the time to make them this complex, might think about how much money that cost, all in an effort to get the sensing right.

MAF Math:
Grams/sec=CFM*.5663
1.3 CFM per 1 HP
.75 Grams/sec = 1 SAE HP

The stock MAF is rated at 255 grams/sec. If your engine consumes more air then that, your out of calibration resolution, not a good thing.
What you and others can do, is set the fuel for WOT, and then just run rich till you get to peak airflow. As a last resort, it's even a poor idea. I will grant that there are 9 sec GNs doing that, but I still don't see why folks do that when there are options available.

The ecm only knows what you tell it, and then uses that info to generate answers. You have to figure out the MAF scalers, or figure out the VE table. Personally the VE style of calibrations are a ton easier for me. The accuracy of any tune is how the fuel and timing are done in these areas. Both take time to get right.

Accleration enrichments.
Depends on calibration code, but the one huge difference is that the MAP system has a MAP correction for Accleration, meaning the code handles AE, and for the MAP systems corrects for actual LOAD. You just can't do that with MAF since it has that lag from the sensor. They both have a coolant temp, and TPS correction.

MAF sensors are always alot more expensive then MAP ones

Both can be fooled into errored readings from reversion. At lower throttle openings the MAF is just a tad better for not being in error as the MAPs.

Remember 20% throttle opening is exposing 70% of the air flow thru the TB. This explains why large Butterflies seem to be better, by the buttometer, and can actually cost performance. It's just at the lower openings they are flowing more air at the same opening so they seem **peppier**

Using the LT1 VE tables are a good starting point for many MAP systems, YMMV.

The only way to find out what the engine wants is testing, testing, testing testing testing. Information about what is really going on and how it applies take hours of work. Sometimes it's easier to build an ecm bench to get a grasp of what's going on.

Which is better?.
It all depends.
For response and WOT performance the MAP will win, when done right. If for no other reason then the restriction of the MAF sensor itself. MAP systems don't care about engine flow, so aren't displacement sensitive.
Just to confuse things thou, some newer systems use both, which can give the best of both worlds.
Final Answer
Final Answer

Bobalos

Man, Grumpy, you are on a roll. thanks for the info. we look forward to the next lesson, teach

BW

------------------
Bobalos
aka Bob W.
www.r71camaro.homestead.com
r71chevy@earthlink.net
<><

JPrevost

I don't care what political party you belong to...you've got my vote for president .

branz28

Very nice... I definitly understand the MAF and MAP systems alot better now.

------------------
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
-Gutted Air Boxes

Grim Reaper

All great points Bruce. There is another difference, which is important if you are into modifying the actual code of the eproms.

With MAF, the BIN is basically full. There is hardly a free byte available. If you REALLY need to add some code or extend a table, you may find yourself having to yank out code that you either don't want, or are willing to sacrifice.

With MAP, the BIN is only 58% full, so you have plenty of room to add extra code or extend tables.

Of course, very few on this board are into Assembly Language Programming. But for us few **** guys that are, this is an important point.

Grumpy

You forgot to mention, that you're only taking about the 165 MAFs. There are other MAF programs that have empty areas.

If you want room to work then you'd want to use the 808 stuff. MAP, 165.

Also, as far as patches go, means being real clever, if you research all the various patches that can be added to the 148, it's amazing, anti-theft, stutter launching, recal for different MAFs, 128 locks at WOT. The 747 ones are limited but interesting.

But, the topic was comparing systems, not rewritting code.

For all the mention of rewritting code, as far as the 165s/730s there aren't any patches that have been completed or shared with the world in general so it's kinda mute anyway.

Bobalos

Glenn,

I was thinin' about this last night as I was taking my new proms out to the work bench in the garage. Here is what i was thinking. I have just bought these bitchen 256K EE's, but have to trick it to load the 128K SyTy prom into it. but when i stick it into the ECM it runs no problem (well that is the plan anyway). if they use the same processore & the same prom. if i am into messin with the assembly language anyway, why not stick it into a 256K prom?

as you know i could not find my bit counter in the dark with both hands so, if I am out to lunch feel free to tell me to go back to my desk & leave the rest of the class alone.

BW

------------------
Bobalos
aka Bob W.
www.r71camaro.homestead.com
r71chevy@earthlink.net
<><

Grumpy

Just load the 128 image in twice. Lots of GMs even come that way.

Grim Reaper

If the ECM is expecting a 128, it expects to access the code from Hex address C000-FFFF. I know the 165 works if you load Hex 00s from Address 8000-BFFF and the actual BIN from C000-FFFF. Some have tried "doubling the bin" onto both halves and claim that the 165 ecm does not like it, but work fine if you hav Hex 00s in the bottom half.

I didn't have enough time with my buddy's MAF 165 car to experiment with "doubling up" the bins, but I cannot see why "doubling up" should not work the same as Hex 00s. Others say they just "offset" the 16K Bin to start loading at Hex addresss 4000 at it works fine for them.

Logically, the MAF 165 only wants the top half; it shouldn't give a hippo's hynie about what is on the bottom half. Yet some people are saying it does.

My comment is, if you can get the AT29C256 to work by WHATEVER METHOD, be happy. I know that for whatever reason, my 7730 ecm wants the AT29C256 to be pre-initialized with Hex FFs before loading the actual BIN I want, while others say they don't have such a problem. If you find you are having problems with one method, try the other. Logically, it should not matter what is in the lower half of the AT29C256 Flash Prom when using it in place of the 27C128, as the 165 ECM only accesses the top half anyways. But, some are finding that it does matter.

Go figure.

Grumpy

The problem, is that the code is a series of patchs. The drivibility folks start with a given code, and then go out driving, as they encounter a problem, they call home for a correction. If you read enough different programs, you can kinda get feel for what's going on. Anyway, mix that with non perfect voltage filtering, and some times things hiccup when running. So by have everything *addressed* and occupied, there is a much better chance for an instant recovery.

I've had stuff run perfect on the ecm bench, but come time to actually run it in a car it wouldn't act guite as expected. Now a days I generally have the battery charger running, just as a lets be sure item.

There are lots of things that get to well this should, shouldn't work, but do. The voltage spiking etc., are some serious issues

Bobalos

[QUOTE]Originally posted by Glenn91L98GTA:
With MAF, the BIN is basically full. There is hardly a free byte available. If you REALLY need to add some code or extend a table, you may find yourself having to yank out code that you either don't want, or are willing to sacrifice.

I think that you missed my point. if the MAF prom (128K prom) has "hardly a free byte available" make the assy language changes that you need, stuff it into a 256K prom & put it into a 7730 ecm. I am doing just this but not making any changes to the code.

am I again trying to make this too simple?

BW

------------------
Bobalos
aka Bob W.
www.r71camaro.homestead.com
r71chevy@earthlink.net
<><

Grim Reaper

Bob, there will be major changes needed to the 7730 ecm (hardware drivers) to make it function as a "super 165" ecm for MAF. Interesting idea though. I am not much of an "electronics/hardware guy", software if my forte. If someone gets the proper hardware setup; and I can make it bark like a dog and rollover & play dead. (Or so I could before I killed all my braincells with my turbulent youth...but I did save a braincell or two).

But, we really should continue this discussion on a different post. It kind of "takes away" from Bruce's post/article on a tangent.

I would prefer to let Bruce's post/article remain "pure" as we are talking about a different aspect IMO; though an interesting discussion.

[This message has been edited by Glenn91L98GTA (edited May 27, 2001).]

Greenshamrock77

Is there any way to increase the ecm's 255g/sec limit. I was messing around with a table and the top few values were 255 and you could not increase the number. Is there a "max value" setting somewhere that you can change to get this? Is the ecm just not able to comprehend anything above 255, or is it just programming? How much can you hack this thing. I am very very new to programming, so be nice if any of my ?'s are off key.

------------------
82' Z-28. 327" w/ flat tops. 216/228/112LS. Pocket ported 041 heads 1.94/1.50 SS valves. World Class T-5 from a 91'. Best run? Runs with 2001 vette to 150mph!

Greenshamrock77

Is there any way to increase the ecm's 255g/sec limit. I was messing around with a table and the top few values were 255 and you could not increase the number. Is there a "max value" setting somewhere that you can change to get this? Is the ecm just not able to comprehend anything above 255, or is it just programming? How much can you hack this thing. I am very very new to programming, so be nice if any of my ?'s are off key.

------------------
82' Z-28. 327" w/ flat tops. 216/228/112LS. Pocket ported 041 heads 1.94/1.50 SS valves. World Class T-5 from a 91'. Best run? Runs with 2001 vette to 150mph!

Bobalos

No problemo
BW

Grumpy

The problem is the sensor.
Then once you change sensors, you have to translate whatever it's output is to something the ecm understands.

**ALL** you have to do is take like a LT1 or LS1 and make a translator so that the 165 ecm can use it.

I run a LS1 sesnor in a LT1 housing on my 87 GN by using a Translator. I can read to a tad over 500 grms/sec.. The translator is a very trick little unit.

The later MAFs are all relatively high freq., it's alot more complicated then what meets the eye.

junkcltr

iTrader: (1)

Seems like it is not a very fruitful effort. The 730 is far superior to the 165 or 808. Why not just remap the outputs (in the code) to what the 165 wire harness is looking for. Or just make a jumper connector to go from the 730 ecm to the 165 harness.

I just don't see the point of using the 808 code. Please explain why the 808 is thought to be better than the 730??

J

hectorsn

Not better but present. The 165 and 808 are the same pcb but the 165 has 8192 comm. What's being talked about is using the 808 code which is MAP and running it in a 165 ecm with a MAP sensor. This way, minimal rewiring and hardware changes and no need for a new ECM.

junkcltr

iTrader: (1)

OK, Yeah it is simpler.

J

drive it

iTrader: (1)

Hmmm, OK, I know this has got to be wishfull thinking but I have to ask..... When I scan my 165 with diacom, or any scan tool, it gives me a map volts reading-it doesn't change much, but why is it there? Is there actually map buried somewhere in the 165 code???? Or is it just an "quirk" of the scan programs?

kvu

I have map reading on my maf car too.I think its for the egr.

Grumpy

LV8 or VE, both are interchangable depending on the math you want to do.

The ecm needs to figure VE to work out the EGR stuff (I think that's a universally true statement).

Some code has the programming in them for using either sensor.

kvu

[quote] LV8 or VE, both are interchangable depending on the math you want to do.

The ecm needs to figure VE to work out the EGR stuff (I think that's a universally true statement).

Some code has the programming in them for using either sensor. [quote/]


Ok understood,but I have a map sensor readout on my scan tool.The 85 ecm had the same thing.I have used 87,88 and now the 6E mask.But both maf ecms have a map sensor readout.I do know that when I go up or down a hill the volts will change.But they are always steady.Unlike an sd map which changes w/ throttle and uses k/pa(?) not volts.I think there is two sensors that go to vac,the inline egr vac control thing and sensor at the charcoal canister.I have scanned both the 870 and 165 ecms for hours and found the volts change when I'm going up or down hills and such.This "map sensor volts" readout is on my scan tool.That I'm sure of but I am wondering what the ecm uses this for??

kvu

?

kvu

Ok I did some more scanning and the map did'nt move at all today.So I guess nobody knows anything about this???