determine engine load (MAF or MAP)?.( build fuel map?? )

It's all about the code!
Post Reply
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

hello forum. i have been the procces, of understanding what the Engine load, actually is, and how its connected to the outpouts, of MAF and MAP sensors...
1)
i know , that with the transfer function of the MAF, you can have the air grams/Second , going into the engine... but how can you use it to know the actuall engine load?
and that the MAP , Gives the vacum of the manifold, but again after that...

2)After you have the engine load. what can you do with it?? :P i mean, you will have a table Engine load / Fuel or Engine load /RPM / fuel ???


my concern is this.. considering you do have a wide band oxygen sensor, and a completely empty ecu. if you had like 10, engine load sites (10,20, 30 ) , you could , manually, begin with a semi-random map, filled with the same fuel on each Engine Load site, and then, during the tuning procces, holding the engine load stedy, add fuel via the software, untill you see stoichometric on the wideband... and then move to the next engine laod site.. but this, wouldent account for the rpm... correct??

so my question is this: how do you determine engine load? :D
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by AndreyB »

You have too many different questions in the same post :)

My own understanding is that "engine load" (I am putting this in quotes for the lack of better term) is just some value related to... engine load :). It can or it can not have any physical meaning - it's a question of opinion and there are multiple approaches. But what most implementations agree that there

Code: Select all

fuelAmount = f(rpm, engineLoad)
and that a table of that f(rpm, engineLoad) is a good enough start.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

yes, but how do you know, when the engine is at laod 10%, or load 100% ? ...
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by AndreyB »

settra wrote:yes, but how do you know, when the engine is at laod 10%, or load 100% ? ...
It depends on the approach you choose. For instance, in Alpha-N "engine load" would the throttle position, 0-100%. You DEFINE what you use AS engine load, and you start from there.

See rusefi_enums.h
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

i see. but still, since the engine load function, with MAF, arledy contains the RPM, why is the fuel table 3d?? (has a spot for RPM as well)
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by AndreyB »

settra wrote:i see. but still, since the engine load function, with MAF, arledy contains the RPM, why is the fuel table 3d?? (has a spot for RPM as well)
MAF is the amount of air per second. I do not see where it would it contain RPM.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

i thought that, engine load = MAF*2*60 / RPM ???
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by AndreyB »

settra wrote:i thought that, engine load = MAF*2*60 / RPM ???
Once again. Engine load is what YOU DECIDE TO CALL ENGINE LOAD.

So, if you decide to say "engine load = MAF", than "engine load = MAF". And not "engine load = MAF*2*60 / RPM"
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

ok:) i think i understand!!
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by puff »

To my understanding, engine load is the force which slowdowns the engine (e.g.something on the crank). You may run 3000 rpm with half throttle easily, continuing to ride at the same speed. And you may run the same 3000 rpm with full throttle and the car slowing down climbing on a steep slope. In the first case engine load would be small (with corresponding MAP, MAF, or TPS sensor readings), while in the last case, engine load will be regarded as high (the sensor readings will also be different).
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

so, i was thinking...
with the transfer function of the MAF, you can know grams/s.
for the sake of this post, i will refer to rpm , as CAM revs. so , if you know your rpm, you know how many revs the cam, does per second, and hence, how many grams of air, get in each cylinder.

grams/cylinder = MAF/ (rpm/60*4 ) cause u have 4 cylinders, or whatever, something similar. in that case, you know exactly how many grams of fuel you need to inject, and hence you know the timing...

(insert here some sort of correction based on the wideband oxygen sensor)

in that case, why is engine load OR any sort of fuel tables, neccecery ??
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

Generally, if you are going from idle RPM with WOT to full speed RPM, you are at 100% load, and you have a variety of RPM's that need a fuel pulse that will deliver the desired amount of fuel. However when you go from full speed RPM then snap the throttle shut, you are at 0% load and you'll be undergoing a variety of RPM's. Then of course there are a variety of situation in between. Hence the common load and RPM tables to get your fuel pulse. If you use a linear equation, it's like picking a one dimension table. This means you are either tuning for something like a constant 50% load and you'll either sacrificial power or efficiency. The table approach allows non linear adjustments. Also remember that load is very close to the MAF signal. So you equation could replace MAF with load with the way that many consider load.
Welcome to the friendlier side of internet crazy :)
User avatar
Sergey89
contributor
contributor
Posts: 839
Joined: Wed Sep 25, 2013 5:30 pm
Location: Russia, Velikiy Novgorod

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by Sergey89 »

settra wrote:grams/cylinder = MAF/ (rpm/60*4 ) cause u have 4 cylinders, or whatever, something similar. in that case, you know exactly how many grams of fuel you need to inject, and hence you know the timing...
You're right, but we need to know not just the voltage from the sensor. We need to know the value of the air flow. For this purpose we need to use the nonlinear transfer characteristic of the sensor, which is not always known. But for engine tuning we are still need a fuel table that will determine the air fuel ratio in the different engine operation points.
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

kb1gtt wrote: If you use a linear equation, it's like picking a one dimension table.
if the equetion is Grams = MAF/RPM and maf,rpm are both valuables. then you have a 3D equation, and if you plot it, it is a 3d map... (like a 3d table). besides, that equation takes on account, different rpms, on different airflows...
no??

when i say MAF, i mean air grams/s. i consider, i will have the transfer function of the maf sensor, to convert voltage to maf.......

(i think that the same EXACT equations would work for the MAP sensor as well, but you would need to do soem more equations, to go from pascal, to grams/rev )

EDIT: i dont get it. even if you decide to go the "Engine load table" way. IF i say that engine load = MAFvoltage/rpm . then why does the table need to have a seperate ROW, with rpm??? wouldent it be the same as having an : engine load / injection time one??
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

Some other issues to consider, most low cost pressure sensors have a 1mS delay, this causes a bunch of non-linear warping of the signal, which means the voltage reading doesn't correlate to air in the cyl. If you have a hot wire or flappy paddle MAF, it also doesn't directly correlate to the number of O2 molecules in the cyl. There's also quite a bit of signal warping caused by electrical and physical parameters of the engine. While you can de-warp (or pre-warp) signals like a MAP or MAF, there are to many to do reasonably well. Even if you use a very fast DSP, there will also be the issue of signal jitter added by variations in the software execution time. The simple approach for dealing with a very complicated signal chain has traditionally been to use a lookup table. It's easy to implement, it works reasonably well, and it keeps the software inducted jitter constant.

You are correct, I wasn't clear when I called it 2D, your equation is technically 3D, but it's a flat 3D. If you created a 3D graph of your equation, then compared it against a normal 3D table, you'll find spikes and valley's that are not accounted for by the equation. This flatness of the equation approach would result in not preforming to peak performance, or dumping extra fuel that will add to the stress of the system.

With more work, a matching equation could be generated that would generate the same or nearly the same parameters as the lookup table, A benefit of creating such an equation is that one could plug in parameter like the Volumetric Efficiency (VE) of the engine setup, fuel injection size and pressure, ect. Then it would just work and go. Also, if you are concerned with jitter, such an equation could be used to generate a lookup table, allowing the equation to be pre-processed, avoiding that computational processing in the MCU.
Welcome to the friendlier side of internet crazy :)
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

okey, so what else should the equation take acount for??? considering i target stoichometric, for economy??
becouse even with the tables, i still have a 2D table. i cant understand where the 3D comes from.
if EL =maf/rpm , then there is no reason to have another row with RPM. becouse when rpm changes, the EL will anyway change???

(unless of course, i should consider something else as EL.)

are you refering to a table like that ? :

http://www.stealth316.com/images/fuelmap-95spydervr4.gif

maybe i have it all wrong, becouse that table, has target A/F , and not injector open time ... :P

as for the load, i found this :

Load = mass_{ingested}/{mass_{standard}

mass_{ingested}[g] ={MAF[g/sec] x 60[sec/min]} /{RPMcam[1/min]}

mass_{standard}[g] = displacement[cc]x0.00122521[g/cc]
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

3D tables like this.
Image
dimension 1 is RPM
dimension 2 is load
dimension 3 is desired fuel

Often desired fuel can be a base fuel pulse width, which then gets modified by things like battery voltage, or dead time compensation and such. So these tables often aim to deliver grams of fuel, or a base injector pulse width. Then the result obtained from the table often is manipulated to compensate for variations based on battery voltage, or injector dead times.

Some people consider load to be the power that's desired from the engine, others consider load to be the fuel you deliver, others consider the intake air to be the load, and of course others have other theories. Load is well, a load.

Stoichlic is almost never the proper fuel ratio. The above noted equation is using one fuel ratio. You need to change the fuel ratio as you change RPM and load.
Welcome to the friendlier side of internet crazy :)
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

Here's a picture showing how the AFR varies based on power or efficiency.
Image
found here http://en.wikipedia.org/wiki/Air%E2%80%93fuel_ratio

DIY folks often push for the peak power all the time. However many OEM's will determine when you are in cruise mode and will then look for the lean condition to save fuel. Basically if you change your TPS, you go straight for power, then if your foot is steady for a period of time, or if you are idle, then switch to an efficency AFR. The efficient point is hunted down by your O2 sensor, which typically has a very slow reaction time.

If you have a drive by wire system, your foot commands desired power, then the ECU can vary the throttle body change the map such that you can get the power demanded while keeping at the efficient fuel ratio.
Welcome to the friendlier side of internet crazy :)
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

coool. i see now.
yes, my equation had a fixed fuel ratio, becouse i though, that this was the ideal. But i see now.
and how can you dteremine the best AFR for each spot?? i soppuse on a Dyno?? keep in mind that i mostly care about fuel economy, not so much power (i wouldent care if the car had half the Hp, as long as it had more fuel economy :P )
(btw , thank you very much!! your replies have helped me come long way into understanding what i look for!! )

so, i will anyway need to know the MAF transfer function.. it will be hard enough, since i can hardly find any docomunation at all :/
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

No problem and glad to help.

A couple cautions about going lean. You can go too lean such that the fuel charge doesn't pull away the heat, which can cause the "boundary layer" to fail, which then erodes pistons, heads, ect. It's one reason why DIY'ers go for the power as it's running rich and is less likely to have a heat build up issues. Also when you go to lean, you can blow out the flame front and you end up dumping un-burnt fuel. For fuel economy, what you really care about is max delivered power for a set volume of fuel. This is likely not the leanest or richest condition. It's likely somewhere between, but closer to the lean condition.

As for how to know your optimal lean condition, that would be with an O2 sensor. On a dyno is a common approach if you have access to one. Another is to use the butt dyno. Find a long up hill area and do a set of test runs, changing the fuel for a fixed RPM and load, finding as many points on the graph as you can. If you find a handful of points, you can often fudge most of the rest reasonably well. Seems most maps are very similar in the general shape. I seem to recall some some tuning software will also allow you to plug in a couple points, then it will fudge you a starting point. After that you really need to tune each point to verify it's good.

If you are looking for peak fuel economy, I'd suggest Ultra-Lean with FSI systems. http://en.wikipedia.org/wiki/Gasoline_direct_injection Oddly enough, you can also get significant increases in horse power with a FSI system, basically use a turbo and crank the boost way up. No worries of detonation as you can slowly inject fuel controlling the combustion pressures.
Welcome to the friendlier side of internet crazy :)
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

cool! when you say :
As for how to know your optimal lean condition, that would be with an O2 sensor. On a dyno is a common approach if you have access to one
what sould i look for in the O2? :P


i should change the AFR ratios, untill the dyno gives max torque for each spot??
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

Lean O2 is a bit of an unknown for me. I guess if you have a dyno, keep it at a fixed RPM and MAP which should result in a fixed amount of torque at the wheels. Then decrease the fuel pulse until the RPM can't be sustained. That would be the absolute min fuel pulse. However I don't know what AFR or might melt the engine, so this min may be an unsafe for long term use. Remember that flaming out, will also build carbon deposits.

A common approach if you have a dyno is keep a fixed RPM, and MAP for that constant torque, then increase the pulse until you start to loose power. You bounce between spark and fuel tables, tuning each until they are at the peak, Then you apply some black magic, as there are lots of things like to much fuel can heat your exhaust and cause other stresses and such.
Welcome to the friendlier side of internet crazy :)
settra
Posts: 24
Joined: Thu Jan 23, 2014 4:20 pm

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by settra »

well, O2 seemed usless on the above :P

ok, so there are two different sessions.

if you tune for best power, you have fixed RPM and MAP, and add more fuel, untill you no longer gain any power??

for safe economic tune: running too lean is indeed very dangerous for the engine... i would think that, you could start by having fixed RPM and MAP, and adjust fuel, untill you see stoichometric on the O2? much like what the closed loop tries to do??
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

NO2 usually looks for the crossing points of .7V, not the .7V itself. Basically it's a PWM signal. If 80% of the time it's above .7V, then it's either rich or lean. Then if it's 20% above .7V, it's the opposite. The OEM has determines where the optimal % above or below is and tries to tune to that %. Stoichometric would be 50% above .7V. Also stoichometric may be safe or unsafe, I don't know how to determine how lean is to lean. I also don't know how much of a variation you'll get. We may be talking fraction's of a % of gallons per mile, or we might be talking about an additional mile per gallon. I doubt it's more than a mile per gallon though.

O2 is about worthless for tuning power, it's all for emissions, and long term diagnostics.

Correct you tune each cell in the table for peak power at that load and RPM.
Welcome to the friendlier side of internet crazy :)
User avatar
abecedarian
Posts: 386
Joined: Fri Nov 15, 2013 10:49 am

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by abecedarian »

@kb1gtt
Not true... if it's above 0.7-0.8v the mixture is considered rich and below 0.3-0.2v the mixture is considered lean. The problem with NBO2 is that outside of stoichiometric, and actually rather close to that as well, the O2 sensor is not sufficiently accurate for any real use. Additionally, the fuels blended with ethanol can also throw their readings out of 'calibration' as they weren't designed for such use, which affects fuel mixtures, as well.

Nothing about a NB02 is like a PWM signal- it produces a constant voltage output given any particular O2 content; if the mixture doesn't change, the voltage output doesn't. What may make it seem PWM-like is that it is part of a feedback loop where the ECU enriches the mixture when the O2 shows 'lean' and leans the mixture when the O2 shows 'rich'; that is all.

My Toyota pickup (1991 22RE) ECU, when in 'closed loop', oscillates between slightly rich and slightly lean, sampling the O2 sensor 8 times per second. It tends to run well in the winter when the 'winter blend' fuel is at the pumps, but I have to adjust things when the weather warms to compensate for the ethanol blend California mandates. Over the past few years, I've got the adjustments down to a minimum but it's nothing like it was 10 years ago when I didn't have to adjust anything.

Side note: running slightly rich at times has some benefit: it heats the catalysts up and causes the sulfur deposits on the catalyst to release their bond and react with various hydrogen species present in the exhaust stream. I know you've smelled that rotten egg odor after hitting the throttle. ;)
You can lead the horticulture but you can't make them think.
blundar
contributor
contributor
Posts: 141
Joined: Tue Jan 07, 2014 4:38 am
Location: Cincinnati, Ohio
Github Username: blundar
Slack: Dave B.
Contact:

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by blundar »

Not to be too terribly rude here, but you guys really don't have any idea how effective engine management works. :)

The whole MAF load thing is a big question. You might know how much air you have with a hotwire sensor but coming up with a load value is more tricky.
-Ford normalizes the value to "standard" conditions and displacement. A "Load" value of 1.0 indicates that the engine is ingesting exactly the same amount of air each rev as would be present in the volume of one cylinder at sea level at 25degC air temps. (I think) This means that virtually no naturally aspirated engine will see a load value equal to or greater than one. This basically involves dividing absolute airflow (in g/s) by RPM, Displacement
-GM, Subaru and many others leave out the displacement factor. You end up with g/cyl as the final measurement
-Nissan and a lot of early Bosch systems use (essentially) injector pulsewidth as a load index. When you think about it, this actually works moderately well especially if you somehow factor out target AF

Many MAP-sensor based systems simply use a RPM vs pressure table. Crude but effective.

There is a trend towards interconnected models in most of the better engine management offerings available. The big idea here is to be able to change one piece of the puzzle and not have to re-tune the the whole system. I *really* think that it would be foolish to not implement something along these lines. It's not that hard to do when you're building something but adding this degree of modularity down the road... This basically requires thinking about engine management as a series of interconnected systems.

What I mean:

-You have a fueling model. This model takes as an input the amount of air present (in grams/s or grams/cyl), a target lambda value, fuel stoichiometry (either specified or measured from flex fuel sensor) and spits out the amount of fuel required in g/s.

-You have a fuel delivery model. This model takes as an input the amount of fuel required in g/s, the number of injectors present (so you can do things like staged injection) and figures out how much fuel to try to deliver with each injector.

-You have an injector model. This model takes the amount of fuel to deliver with an injector (in g/s), injector latency data (presumably measured in lab), injector flow rate data (presumably measured in lab), fuel density (specified / measured) and comes up with an actual injector pulsewidth to fire the injector.

Or on the airflow side:

-You have a speed-density model that takes a pressure reading (directly from a MAP sensor or from a MAP sensor and then averaged or ...), a configured value for engine displacement and then an estimated charge temp along with a configured Volumetric Efficiency table which then spits out airflow in g/s (i.e. classic GM, Megasquirt, older Motec, Haltech, BigStuff3, Fast XFI, etc.) Main weakness IMHO: density correction algorithms (i.e. charge temp prediction or figuring out how hot the air is going into the motor. Weather sensitivity)
-You can have different charge air temp algorithms that use a mixture of IAT, ECT, RPM, etc. to alter the way air temps are predicted

-You have a Alpha-N model that takes a throttle input and RPM and spits out an expected airflow value.

-You can combine Alpha-N and speed-density, making the (mostly correct) assumption that the movement of the throttle plate is going to be one of the better predictors of how engine efficiency changes. The setup for this is much like a "traditional" Alpha-N but the main tunable table contains efficiency values which are then used to scale a RPM x pressure volumetric efficiency table.

-You can use a classic hotwire MAF algorithm with a voltage-output style MAF, a MAF transfer function and spit out g/s.
-You can use a classic hotwire MAF algorithm set up to use a frequency-output style MAF like newer GM, Ford, Mitsubishi use and a MAF transfer function and spit out g/s.
-You can take a classic MAF algorithm and attempt to provide more robust correction but taking advantage of the integrated airflow sensor that is almost always present along with King's law equations
-You can do MAF anticipation using the present reading, past readings, rates of change in order to try and anticipate MAF values in situations where the MAF signal changes too rapidly to be entirely accurate.
-You can take an average of the MAF input over a longer period of time in order to smooth out uneven pulses, i.e. large camshaft at low RPM idle.

-You can have a transient airflow model that attempts to predict how much airflow is present during sudden RPM or throttle changes when all of your primary metering systems are going to fail.

...

So what's the big advantage here? Why are more and more OEM and aftermarket systems heading this direction?

One big reason: failures and error handling.

What happens on an OEM vehicle if you walk up to it while it is running and unplug the MAF? Does it shut down? In most cases, the ECU will throw the check-engine-light on to indicate that a fault has occurred and then it SWITCHES TO A DIFFERENT MODEL FOR AIRFLOW. Ok the MAF has failed and we can't figure out how much air we have from that? Substitute Alpha-N from a backup system. Structure your systems in such a way that the Alpha-N metering system and MAF metering system have compatible outputs for airflow and BAM the car continues to run.

GM has done this on nearly every one of their vehicles since the late 90s: vehicles come with both a MAP and MAF sensor. The MAP sensor is primarily used close to idle and lower RPMs. The system switches over to being much more of a pure-MAF system at higher RPMs with more laminar flow. Transient fueling (even at high RPMs) switches back to using Alpha-N and the MAP sensor as inputs.

Ford's factory supercharged vehicles are equipped with supercharger inlet pressure sensors that are used by a transient model during sudden changes in throttle until the MAF can stabilize.

This rant is long enough already...

p.s. I just bought a dyno. Woohoo!
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by AndreyB »

blundar wrote:p.s. I just bought a dyno. Woohoo!
I hate you.

PS: You are only 5 hours away...
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
skeeters_keeper
Posts: 44
Joined: Fri Nov 08, 2013 3:18 am

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by skeeters_keeper »

Excellent post!

I've been reading "Designing and tuning High-Performance FI systems" by Greg Banish... but I knew enough to know I don't know enough to participate in this thread yet, even though I saw a bunch of contradictory things being posted. Thanks for your "expert" opinion... and congrats on the dyno!
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by puff »

seems you are sarcastic Ж-)
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: determine engine load (MAF or MAP)?.( build fuel map?? )

Post by kb1gtt »

I'm one of those that doesn't know what I'm talking about, but will generally try to take a stab at a discussion. I use my hobby projects as a place for me to get over learning curve issues. Thanks for the post, there are certainly things to learn in there. Also congrats on the DYNO, I hope to see you post some things about how you suggest tuning a vehicle. Just like engine load, I've seen a massive number of ways that people claim to tune a vehicle.
Welcome to the friendlier side of internet crazy :)
Post Reply