Is any body making large MAF sensors for our cars? When updating your air induction system this is the most restrictive point. Is there any way to overcome this other than going to speed density?
85-87

 

The airflow limitation in the ecm is the major problem, not the MAF. It can only see up to 255 g/sec. An LS1 sensor can be made to work with a little tinkering and a converter, but youll need to sift through the calibration and make it so the airflow represented there in the tables is actually 0-510 g/sec or whatever you want rather then 0-255. Youll take a hit in the resolution department as not only are you doubling the flow, but the LS1 sensor doesnt have the same output curve that the bosch one has. Other then that, it would work.

 

The other way around it is with source coding so you can minimize the error in the tables. Much harder if youve never done it before. This also means the use of 16 bit table(s) rather then the 8-bit ones. I can tell you from experience that a full 16 bit MAF table and accompanying 16/24/32 bit algorithms is kinda tricky to implement and set up, though. All in all, the sky would be the limit after that. You could even use a dual MAF system to really increase the power you could support. Like 1500+ HP or something crazy like that.

 

I'm pretty sure on a 7730, that F10 (VATS) is a counter port.
It's label MAF on ~Ludis.
Is there a BCC for a 730 that used a MAF?
Never mind, Yet again I find something I'm not looking for....
Vats input

dimented24x7- I thought you used an 8 bit ADC port for your maf input?

 

Yes, I also used a frequency converter since the MAF is set up for a pulse accumulator (the output apprears to be a series of short pulses). The MAF is an uneven signal as its a series of pulses so maybe the number of counts you get over a period of time could be used to infer the frequency if you precicely knew what the waveform looked like. I gave up on it as I didnt have a scope and just used the converter to feed it into one of the ADCs.

 

Now that I revisit it in my memory, the above method could potentially be very accurate as you could use the frequency counter as a pulse width accumulator or to figure out what the duty cycle is using the pulse train from the MAF sensor. Basically just get the additional accumulated counts over a time interval and do the math. only problem is that an o-scope is a must for this. Would be hard to do without knowing what the waveform looks like. I may yet try this as with the 8 bit ADC the idle can be a little jittery from the relatively large step size in fueling.

 

And I'm actually a bit surprised to read that the MAF is the most restricive element in the induction system. If you do a little measuring, you might find that the cross sectional area of the throttle body is much smaller than that of the MAF.

 

In my case I'm working with a 383, heads, cam, 58mm TB, etc. when updating the induction system the MAF is like a brick wall. We can tune all we want but still gas and air is king.

 

Sure is. That 255 g/sec limit is set in stone. People have run 10's with the stock maf, but youll basically have to fudge it to get the AFRs you want.

 

just get a WBO2 and tune the upper range with PE vs RPM. if youre flowing over 255g/sec then its pretty certain your in PE mode anyway and with the WB you can get your AFRs damn near perfect where you want them. so theres really no brick wall with maf like people say, the only restriction is how much air will flow through the maf which depending on if you mod it or not can be up around 750cfm which will support lots of power.

 

Yes, my biggest problem is the air restriction, my AFR is running around 12.5 at WOT but at high RPM I noticed like the car stop pulling and I know I have more power in the engine. To ne able to make more power I need more air.

 

You can tell how much its holding you back by hooking up a vacuum gauge and seeing how much vac it pulls at WOT. Id say anything over 2-3 inHg and its time to address the restriction.

 

yes

i'd venture to guess that unless youve done one hell of a buildup, your bottleneck will not be the maf it will be somewhere else in your induction, like runners, cam or heads. also, alot of people scream that you shouldnt descreen your maf. i say bs. as long as you can tune your own chip, do it. you can get an extra 150-200cfm by modding it and by recalibrating the maf tables, with time, you can get it tuned better than it was when it was stock. i am running maf on a heavily modded 383 SR and it pulls great to 6500rpm.

 

There's more to life then WOT, IMO. If you're looking to make the car *fun* to drive, then you'll eventually switch over to SD. A MAF by it's very design can't offer the transistional response of a MAP system (as implimented in any 3rd Gen based Code).

 

Grumpy I certainly respect your opinion but if maf id so bad why did gm decide to go with it instead of sd? it wasnt because it is cheaper, mafs cost more. also, i dont see why a sd car is more fun to drive than a maf car. the benefits i see are possibly slightly better fuel economy and perhaps the ability to eliminate some intake restriction resulting in somewhat better topend. aside from that, i dont see why a well tuned maf car wouldnt be just as driveable and "fun" as as sd.

 

GM didn't switch back to MAF, gm added a maf sensor along with the speed density, to calculate airflow for EGR utilization. (emissions), and a few other things where the airflow calc was less accurate guessing by manifold pressure. (plus in some masks, GM got lazy when it came to calculating true VE, if you look at some of the 305 $8D bins you'll see what I mean)

Even grumpy uses a MAF on his GN for <B>datalogging</B> purposes.

If you dissagree with Bruce on the "fun to drive" aspect of speed density, than you simply just havn't gotten to that level of tuning yet. If your willing to "try new things" (like various ecm + mask combinations as I have over the years), you'll see exactly what he means. But it doesn't mean switching and back and forth in a day, it means switching, spending a few months tuning, etc.

-- Joe

 

I can't tell you why Gm makes the desicsions they do.

All GM type MAF systems have an inherent flaw, in misreporting sudden changes in airflow. The engine doesn't use all the air that is reported by the MAF, during a throttle snap, on accleration.

So in transitional situations the MAP based systems will be more accurate. Well, when tuned correctly.

I've run both MAF, and SD, within the same code, and in back to back testing, there's just no comparison, when you're really being picky. For transistonal situations, the MAP is superior, IMO.

Now, the MAF systems have an advantage in say WOT, when there are no sudden changes in airflow (ie more then just a gear change type situation), IMO.

In the latest OBII systems as I understand them they use the MAP for transistional/ low load conditions, and then the MAF for WOT, which just so happens lines up with what I've said in the above.

While some have *said* the long runners, small plenum of the TPI was for torque, it might as well be argued, that it was to min the MAF misreporting at transistional areas. You'll note that in all the later SD systems, they went to a more *conventional* runner/ plenum, and picked up more HP, and Torque with no hit in emissions.

BTW, the stock MAFs have been in the 9s in the TRs. But, then again, that's a car that stages at +4K, and shifts at 7K. So again, that's all in the area where the MAF shortcomings don't really hurt it. But, that's not typically the kind of engine/tune you run on the street. And with that lose of converter, going to a cruise in or for a trailer queen, you can fudge past having what I'd call a *nice driver*.

The 255 limit, really isn't a limit, when dealt with correctly. You'll note thou, that the newer ones, are calibrated so the resolution is better at the higher air flows.

The other point, in none N/A appls., the MAFs don't allow for running different AFRs at different RPM/airflows, which the MAPs do allow you to do (well as what they could be remapped to do). In the none N/A appls., you need to be able to adjust the fuel load to also aid in in-cylinder cooling.

So while you might get a MAF to perform, and drive well enough to keep you happy, you're still leaving some items on the table. IMO, if I'm going to go to all the effort of trying to get something correct, then why not just go all the way. While some settle for what they do, I tend to needle them along to be the best they can be.

 

this is exactly why i always respect grumpys opinion, never a hollow, pat answer or lame attempt at a putdown. he puts forth an opinion obviously well steeped in experience and knowledge. this was a far better answer than i could have ever hoped for, explaining facts about the benefits of both maf and sd. i wish everyone here could take a lesson from him on communicating. thanks, bruce.

 

I'm going to need some information on that. The only change I've known of was the removal of the cold start equipment in 89, a year before SD.

 

Lt1's

Grumpy, it's hard to say that they didn't take a hit on torque. The Lt1's don't make as much mid range torque (and horsepower) as the TPI's, even with different heads. Also, the Lt1's combustion chamber was fast-burn and had excellent quench resulting in lower emissions. The addition of the opti which had more accurate spark control also lowered the emissions at idle and transitioning. I'm inclined to believe MAP is better because of where it's located. Having a MAF ahead of everything is going to have those transmission delays just like having a wideband 10' downstream from the engine.
I was helping a friend yesterday with his blown cammed ported and stroked 93 Lt1 when I noticed the 2 bar map was reporting no boost when his gauge was showing 2-3psi. It dawned on me that he's telling the truth but having only datalogged "40kpa" (2-bar is 72kpa) I had to ask about where he's got his boost gauge hooked up. Told me it was after the intercooler . Which is fine when you're trying to measure WOT boost but it isn't reporting accurately what the engine is seeing under part throttle.
In other words it's just like the whole MAF pre blower/turbo and post or even post throttle body. What is the MAF going to report when it's installed just after the air cleaner on a blown engine when you lift on the throttle? Even without a blow off valve venting to atm (which would really screw up your fueling) it's still not reporting the air flow into the engine accurately (based on time). Eventually all that air will go in but not when the MAF reported it! Hence the newer stuff that has the MAF post throttle body. It just seems pointless to run MAF on a 3rd gen f-body when so many have had great success with SD. It's easier to install and plumb and all of the ecm's support it in one form or another . People shouldn't limit themselves to not trying SD with the excuse of "it's not what's on the car right now" or "I don't want to do a simple repinning." Those are stupid reasons. Just say you don't have the time to futs around with the tuning because if you did, you'd have the time to at least try SD.
Reasons I like MAF- I don't have to spend much time with the fueling so long as the MAF it's installed properly

 

Completly different intake and heads, I might add . Lets not forget that they reverted back to eventually including a MAF in the fueling mix in addition to MAP based fueling. The engine doesnt really care one way or the other so long as it gets the right ammount of fuel.

 

... when it needs it! MAF = poor in about 75% of the car's I see modified on the street/show/drags. Sure you can tune it to go down the 1/4 mile but open loop is crap with those cars and they try and say, "oh, it only takes 30 seconds or so before it clears up." My reply is, "it doesn't take me 30 seconds from when I start my car to when it drives like it should... and that's still in open loop "

 

I may not be a prom guy...

... but I do know that a MAF is the best way of KNOWING how much air is coming in. MAP's are estimating. I guess there are just too sides to the topic, but I can repin - I just choose not too handicap my engine with my potential inability to guess properly.

 

My car runs better in open loop with the MAF then it did with the SD. The MAFs readings are largely independant of temperature, which is a huge bonus. Same results could have been obtained with SD, but I got tired of tuning every time I changed something. Most of my remaining problems arnt related at all to the MAF, but other things like the stock PID and AE. With the PID once it hits closed loop its a bucking bronco at low throttle. Not going to bother with teh gains just yet.

One big problem with the MAF ecms is that they lack resolution due to only having 8 bit tables. All the fueling hinges on those tables and if the resolution is lacking, then so will the drivability. Increasing to one 16 bit table with interpolating 16 bit lookup routine is a huge improvement.

So far, the only one to address all these problems is me, and I dont even have TPI

Someone needs to do some coding to the 6E, or even a meld using the $8D to bring it out of the 80'sand make it into a real induction system with a respectable MAF and not that sorry excuse thing bosch made. Its got potential, but like so much factory hardware, it was sort of junk out the door.

 

really, i think the smoothest set-up for using a MAF and MAP on anything , would be for someone to run a '8051 LTx ecm, OBD1, and code it to understand either the opti, or a standard reluctor type igntion. You could run that on any engine, that could use a GM igntion module, and you get the benefits of the larger MAF, etc. Heck you could code it to run sequential or batch, MAF only, MAP only, etc. It seems to be a sweet hardware platform, if only it could read more traditional ignition.

of course, you could use the delteq stuff with that ecm too, but thats not as much DIY.

 

I don't think MAF is the best way. I think the best way is plotting out a VE table based on wideband data. Theoretical air flow vs theoretical actual air flow isn't any different. They're both theoretical. Like I said, the MAF is more forgiving but it isn't reporting actual air flow. Calibrating a MAF on a bench doesn't mean it's accurate when you change the ducting work on the car! Look at the differences between the vette and f-body MAF tables and you'll see what I'm talking about. The number of cylinders and the cam profiles change the "reported" air flow by an error. And again, there's a timing issue. MAF reports air flow potential, ecm/pcm adds fuel but did it add it in time, not in time, etc. These are small things but they add up. The MAP sensor uses an ideal gas law which with some math, and calibration (engine dyno) gives the same result of a MAF under stead state.
Want to get even more intense with these error's as proof that the MAF isn't any better than MAP at reporting actual air flow; both have an error in the injector pulse width calculations. Every injector is different than the next when fired in a pulse. 2 different injector designs that flow 24lb/hr open but when fired at a low dc at idle and cruising the same pulse width could account for a relatively large discrepancy between them. Increasing the fuel pressure changes the closing rate of the injectors so they "hang" open for just an itsi bit longer. These things aren't important. Having a MAF report the EXACT air flow isn't important when you've got the ability to change the calibrations (like we do with prom tuning). For another example of having the MAF report exact air flow; who cares? The ecm/pcm still needs to know how much fuel the engine wants. The MAF has an advantage of having better control over the commanded AFRs... but who's aiming for a specific AFR? I'm aiming for whatever AFR makes my engine (car, and me) happy. So what if I have a commanded AFR of 12:1 and it's actually 12.4:1 when a MAF could have a commanded AFR of 12:1 and be 12.1:1 ... let's just say that engine likes 12.8:1 at the RPM/load... I'm closer . And for startups you already have to be tweaking the open loop AFR tables... only if you're limited to crappy resolution (like in dim and most TBI cases) would one system shine over another. In the end though the engine with the best calibration is going to do very well... and having the MAP in the intake gives SD an edge in the transitioning over MAF. I could write a damn thesis on the differences and why they ultimatly mean nothing to a good tuner. But I'm not going to, instead we should post up proof and theory with sound reasoning at the differences. Maybe I should put together a tech article . If you want to debate my points please back it up with some physics and reasoning. Just telling me so and so tried it and so and so has a fast car makes baby Jesus cry

 

IMO, MAF is king of steady state fueling. Having the actual airflow takes alot of the empirical nature out of tuning. This greatly reduces, and in some cases, nearly eliminates tuning needed after mods. Most of the tuning then comes down to AE and timing. Even through just SOTP tuning, I was able to get the WOT AFR to within around .3-.5. Im no tuning ***... This is mostly due to the fact that there is only one variable involved, the airflow table. As far as tuning the MAF, the WB is the only way to go IMO. Actual bench flow data has little meaning in the car. I even went as far as to eliminate all intake ducting and just have the MAF in front of the tbi to improve performace.

I still believe the ultimate system would be one of the ecms or pcms with all the smog fat trimmed out of it and the two systems melded together with everything in simple, easy to use terms. The stock tpi maf system is just another ***-tastic design by EPA-whipped engineers that has to sacrifice performance for processor time so there will be enough left over for all the SMOG and other somewhat usless things these ecms came with. I personally like just having the airflow table, 2D open loop actual AFR and not multipliers table, PE table, AE tables, and choke being the main players in the fueling. Expends more cpu power to use human friendly units but it condenses everythign down into bite sized chunks. Even better would be to have the SD handle idle and low load, with the MAF handling the rest. Each system has its weaknesses, but together, they make for a better system overall.

 

Both require tables, and calibration data, or both are subject to the same errors.

Not to mention in transistional AE, the MAF can overshoot the actual *used* airflow.

 

Thats a bonus for me! With the MAF parked in front, it sees some of the manifold filling and corrects with extra fuel. Bad for port injection, somewhat good for tbi.

 

BTW - one solution for fitting a larger (i.e. newer) MAF on our cars will be this:

http://www.maftpro.com/

Haven't tried it personally yet, but apparently his existing stuff (for other applications) is pretty good and functional.

This unit is apparently completely configurable. Oh, and it will be configurable within TunerPro (he asked me to build in some functionality for the hardware and I've done so, but I have no other affiliation). ;-)

M

 

Nothing like using one error to cover for something else.

 

This may not be a bad as it seems. As a TBI setup fills the plenum it also needs to provide the fuel. This is a requirement of TBI. In this case very little AE is then required.

This is unlike a port MAF system where the fuel isn't required until the plenum filling is available to the head runner ports. Then the fuel is required to be added.

RBob.

 

But, without some code corrections, it seems that you'd always be chasing a somewhat random event, since you could never tune for the error, under *all* (heck without the code) any conditions.

 

Really its not error. The MAF is seeing the airflow surge as the plenum fills. W/o this, and/or AE, to correct for it, a large surge of air would travel by but have no additional fueling mixed in with it. To take full advantage of this, though, the MAF had to be relocated to right in front of the tbi so the extra fuel and manifold filling happen at the same time. With duct work, there would be a short delay that would cause the motor to crap out when I punched it. I also had standing wave problems with the long duct as well. Didnt effect the AFRs that much on the WB but it caused a rough, chugging acceleration under WOT. The airflow would surge up and down adn the MAP readings would oscillate from about 95 kPa to the max that the MAP could register.

 

Yes GRUMPY, they both have tables.

MAF - Table says X amount of O2 = Y amount of Fuel.

Map - Table says "What we think is" X amount of air = y amount of Fuel.

Much harder to determine how much load the engine is under with a MAP as well. MAP's messure vacume, MAF's messure airflow, ego's aside. Simply put, how can you better determain how much air and engine is intaking other than by putting a device for messuring that air in the stream.

 

The MAP is good for load. Just look at the MAP reading and theres the load. However, the fueling has to be inferred from the MAP readings rather then directly measuring the airflow. With the airflow known, the fueling is just based on: PW = (total injector firing time) x (MAF airflow/Total available injector fuel flow) x (1/AFR). One unknown variable, one relatively simple equation.

 

Its worth noting that with MPFI, the SD works better then on TBI as there arnt as much fueling dynamics at work. If I had MPFI, Id probably just rock out the SD. There was a really good article that touched on the fueling aspects of tbi and mpfi. I have to see if I can dig it back up. The main things they noted is that SD is better for MPFI while MAF is batter for tbi.

 

MAF translator Pro(http://www.maftpro.com/) is not for sale yet it looks like it will do the Job,

 

Bob Bailey (it's creator), had several at the Buick Nats last month. It's a nice unit, as per all his stuff.

Menu driven, and rather intuitive.

 

It's not an error, just something you have to reconfigure the engine to work around?.......

The airflow is used to develope an LV8, which is a load calulation. If you have an error in reporting the airflow, then the LV8 calc is going to be wrong.

 

I'd be interested in seeing if in that article, they even discuss the reporting error, I've mentioned.

 

Not really, the main fuel table is VE, which is actually just about cylinder filling, not vacuum.

As I said, in transistional areas, the MAF mesures the airflow, NOT what they engine is actually using. As the throttle is opened the vacuum level in the manifold changes, and the MAF over reads the airflow the engine is actually using, since some of this air is being used to just change the manifold vacuum level.

When not in transistional modes, where the manifold vacuum isn't changing much, the MAF is fine. But, the MAP is fine in both situations.

 

And what about AE and transitioning? All of a sudden MAP is easier to tune because there is less math . It's all relative, but when it comes to big cams, MAF is easier to tune on the street because it compensates for the idle rather well and stead state fueling (cruising) is very close. MAF is definatly easier to get close but from all of the tunes I've seen done I'd have to say the best have been SD. They respond to the blips of the throttle, they don't backfire through the exhaust nearly as much on decel. They have excellent control over the timing. That's my perspective. Want to get the tune done fast and close enough to make some 1/4 mile passes, MAF is a no brainer. Want to work with it to get it perfect, SD has no compromises when enough good effort is invested.

 

For me there is NO LV8. All load sensitive items like AE, timing, open loop AFR are handled by the MAP sensor. I wanted to make a system that works, not works like **** LV8 is just an aproximation. I wanted to avoid having to compromise by making approximations

And, Id like to say again that its NOT an error. Thats actual airflow into the intake. With the 4 or 5 feet of ductwork, there where compressibility as well as extreme resonance effects at work. Not somthing youd want to have on any motor.

With tbi, that 'error' must be accounted for, or there will be air sneaking in without any fuel added into it. This is either done via the MAP AE, or leaving the MAF to its own devices. With port injecton itll just show up as pressure at the intake ports, but with the injectors up top, thats now an additional flow of air that needs to have the correct ammount of fuel added into it. And, if you dont like it or have a port fuel system, you can always provide additional filtering during transition to revert back to using a MAP for the transitional fueling.

The only real drawback is that there is only an 8 bit A/D input to handle the MAF input, which equates to low airflow errors. I dont think we should be so closed minded. If improved input could be obtained, then a really good DIY efi system could be put together that has elements of both MAP and MAF.

 

I'm in over my head...

Anyone tried using a VAF on our cars?

 

Nice seeing you sand bag the conversation with what you're actually doing.

The Board is for the discussing what *people* are involved with, while it's neat your into doing your own one of kind code, it's just that, *your's* which doesn't seem to be for public consumption.

And while you might want to be in denile about it being an error, even if not used in AE (which would min it), it still is an error. If your trying to read the airflow thur the engine, the amount wasted in just changing the vac level in the manifold is still an error.

 

Everyone, I do hope we can keep this thread on track and cooled down.

Now before the next post everybody relax for a day.

Thanks,

RBob.

 

My mistake. I saw you quote me when you made that statement so I thought it was pertaining to what I had posted and not things in general.

 

Actually, an early version (with a few errors in it) is available on moates.net file section, under the name '8063_MAF.ASM'. I posted it there a while back.

 

On a note more related to the post. ..
What about the Granatelli adjustable unit?
Supposedly flows 30% more than the OEM.

http://www.thunderracing.com/catalog...vid=6&pcid=179

Could this be used in addition to adjusting the PROM, or is there no point to it?