MAF wish list

Testing and development of Megasquirt 3

Moderators: jsmcortina, muythaibxr

techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

muythaibxr wrote:Techsalvager:

You can do more or less exactly what you want by making your secondary load MAFMAP and tuning using VE table 2.

Ken
I'm not interested in mafmap and using a ve table, The whole point of using maf was because all you had to tune was the transfer function and you're done, set your commanded lambda and away you go. And it has worked very well like this for my different setups.
I tell it to target lambda 1 it targets lambda 1 reguardless of airflow. Like I've already stated I think trying to use a ve table to do corrections isn't the way for it to be done. That means something else is wrong in your setup.

More then likely before people that have had need to tune the low airflow region is because of improper injector settings. Fuel injector settings have to be correct for maf fueling to work right. This is how I found out my rx7 460cc injector dead time isn't 1.2ms like shown in diyautotune base map, (which I appericate they had a base map for my car) but is very close to .75 ms. With this fixed my setup fell right into place. Also my Speed Density setup also became easier to tune in properly and I can stick it around stoich with a nearly flat VE table.

I think the big issue is that people don't realize is maf and mafmap are different in the way they act on the fueling alogthrims. If you want to use mafmap fine its there use it, but I don't have a need for it and don't care for it.
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
elaw
Super MS/Extra'er
Posts: 2926
Joined: Fri Oct 16, 2009 6:20 am
Location: Wilmington, MA

Re: MAF wish list

Post by elaw »

techsalvager wrote:Upper and Lower bounds for the maf sensor.
Under or Over this much for certain time ecu knows the sensor has gone bad and can swap to speed density or alpha N
This I wholeheartedly agree with... although I suspect implementing it is not trivial!

I've actually thought about switching back to SD on my car because with MAF sensing on a forced-induction car there are two potential issues. First is MAF sensor failure - not that common, but much more frequent than MAP sensor failure.

The second possiblity is something many of us with forced induction have experienced - a hose blowing off or some similar failure under boost (exploding intercoolers, anyone?). With speed-density control in that situation, power is of course reduced but the car's still drivable. Using a MAF sensor the car is no longer drivable.

Actually for my purposes, rather than have the ECU try to decide when the MAF sensor has failed, I'd be just as happy being able to use an external input. Besides, being able to select MAF or SD operation with a switch would be kinda cool...
Eric Law
1990 Audi 80 quattro with AAN turbo engine: happily running on MS3+MS3X
2012 Audi A4 quattro, desperately in need of tweaking

Be alert! America needs more lerts.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

techsalvager wrote:I think the big issue is that people don't realize is maf and mafmap are different in the way they act on the fueling alogthrims. If you want to use mafmap fine its there use it, but I don't have a need for it and don't care for it.
But instead you say that you have butchered your AFR table. That doesn't sound like a good approach either.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

elaw wrote:
techsalvager wrote:Upper and Lower bounds for the maf sensor.
Under or Over this much for certain time ecu knows the sensor has gone bad and can swap to speed density or alpha N
This I wholeheartedly agree with... although I suspect implementing it is not trivial!

I've actually thought about switching back to SD on my car because with MAF sensing on a forced-induction car there are two potential issues. First is MAF sensor failure - not that common, but much more frequent than MAP sensor failure.

The second possiblity is something many of us with forced induction have experienced - a hose blowing off or some similar failure under boost (exploding intercoolers, anyone?). With speed-density control in that situation, power is of course reduced but the car's still drivable. Using a MAF sensor the car is no longer drivable.

Actually for my purposes, rather than have the ECU try to decide when the MAF sensor has failed, I'd be just as happy being able to use an external input. Besides, being able to select MAF or SD operation with a switch would be kinda cool...
I actually think all the sensors should have upper and lower bounds settable by the user with some type of fail mode value settable by user of course incase the sensor dies.
jsmcortina wrote:
techsalvager wrote:I think the big issue is that people don't realize is maf and mafmap are different in the way they act on the fueling alogthrims. If you want to use mafmap fine its there use it, but I don't have a need for it and don't care for it.
But instead you say that you have butchered your AFR table. That doesn't sound like a good approach either.

James
Can you explain what you mean there so I can explain back exactly what I did. I'm confused what you mean there.
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

What I mean is that your AFR table should be a real AFR table. Bodging the AFR to unreal values isn't the "right" way to tweak MAF fuelling, you should use VE2 like Ken said.
However, I also recognise that this is not yet documented in the manual.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

jsmcortina wrote:What I mean is that your AFR table should be a real AFR table. Bodging the AFR to unreal values isn't the "right" way to tweak MAF fuelling, you should use VE2 like Ken said.
However, I also recognise that this is not yet documented in the manual.

James
I'm still confused.
I have not bodged any AFR values.
In cruise, low to mild acceleration I have stoich lambda 1, in higher acceleration I got .95 and it morphs into .78 going wot lambda. I'm confused as why you think I have botched up my afr table.

Where did I mention anything about changing AFR values to unrealistic values?
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
elaw
Super MS/Extra'er
Posts: 2926
Joined: Fri Oct 16, 2009 6:20 am
Location: Wilmington, MA

Re: MAF wish list

Post by elaw »

James,

I think part of what Techsalvager may be talking about is tableswitching? If the AFR (actual and target) were able to be controlled by editing the VE tables, then you could do tableswitching based on TPS.

I've implemented that a different and rather kludgy way... I use AFR tableswitching linked to an input pin I've got hardwired to a programmable output that toggles based on TPS value. It sure would be nice just to be able to do AFR tableswitching based on TPS (hint, hint)! :lol:
Eric Law
1990 Audi 80 quattro with AAN turbo engine: happily running on MS3+MS3X
2012 Audi A4 quattro, desperately in need of tweaking

Be alert! America needs more lerts.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

techsalvager wrote:Where did I mention anything about changing AFR values to unrealistic values?
Just the other day when you were saying that you couldn't use closed loop EGO because you had tuned your engine by altering the AFR table away from the real values - or was that some other user?

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MAF wish list

Post by muythaibxr »

In any case, if you use VE2 with MAF, it's no longer really a "VE" table... it becomes a method of trimming the MAF with a 3D table. 100% becomes "no correction," lower than that leans it out, higher than that makes it richer.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
ashford
Super MS/Extra'er
Posts: 1603
Joined: Sun Apr 27, 2008 4:29 pm

Re: MAF wish list

Post by ashford »

techsalvager wrote:
muythaibxr wrote:Techsalvager:

You can do more or less exactly what you want by making your secondary load MAFMAP and tuning using VE table 2.

Ken
I'm not interested in mafmap and using a ve table, The whole point of using maf was because all you had to tune was the transfer function and you're done, set your commanded lambda and away you go. And it has worked very well like this for my different setups.
I tell it to target lambda 1 it targets lambda 1 reguardless of airflow. Like I've already stated I think trying to use a ve table to do corrections isn't the way for it to be done. That means something else is wrong in your setup.

More then likely before people that have had need to tune the low airflow region is because of improper injector settings. Fuel injector settings have to be correct for maf fueling to work right. This is how I found out my rx7 460cc injector dead time isn't 1.2ms like shown in diyautotune base map, (which I appericate they had a base map for my car) but is very close to .75 ms. With this fixed my setup fell right into place. Also my Speed Density setup also became easier to tune in properly and I can stick it around stoich with a nearly flat VE table.

I think the big issue is that people don't realize is maf and mafmap are different in the way they act on the fueling alogthrims. If you want to use mafmap fine its there use it, but I don't have a need for it and don't care for it.
there are several thing i don't think you are taking into consideration which may not have affected your installs.

1. bigger cams. i have a lumpy idle i haves used ve table 2 simce my first install of the ms3(i got one of the first batch sold) to account for this. with a lumpy cam genneraly it acts like an egr valve and get a puff of exhaust gasses into the intake during overlap at idlecausing a back and forth pull in the intake which skews actual reading of air that is ingested into the motor.

2. pcv. this alone if the clean air pipe is hooked to the intake post maf has no effect. but also introduces blowby to the intake which is unmetered. after i hooked my fresh air supply post maf it helped out alot.

3. intake pulses. i have scoped my maf to try and set up maf sampling idle had some pulsing midrange was fine and top end looked like a saw wave. ultimatly i inserted a small capacitor on the signal line to smothen it out. the take the lowest of this window under the sampling did not work well in my case what i actuall needed was the average of the signal. i have tried several intakes and one in particular would not run on a maf and had to go back to speed density.

4.meter resolution. a meter that is good for 250hp will give finer resolution than one that can read 500hp. my particular meter does not read lower than 5g/s and pegs out around 400(have the calibration info) my idle is around 12-15 g/s and cruise is about 20. very near the bottom of what it can read.
shauer
Experienced MS/Extra'er
Posts: 364
Joined: Wed Oct 05, 2005 10:39 am
Location: Rochester, NY USA
Contact:

Re: MAF wish list

Post by shauer »

I personally like the current VE table correction. My one question is if we could simply use MAF (g/s) as the load variable and get rid of MAFMAP all together? I don't understand the technical need to create a calculated load value when the MAF reading is right there.

I too have a NA engine that has a lumpy cam with high overlap, plus an intake design that has strong resonance and a little inversion at a few different RPM/Load points. The VE correction table does a nice job of taking these other variables into account. I have my transfer curve dialed in to where the VE table only needs about +/-5% correction to account for these other factors. I have found that VEAL set on "very hard" for auto-correction does a nice job with the current VE correction table.

On a completely different note, I would like to see TS add a MAF transfer curve tuning function similar to VEAL. If you are going to get rid of the separately downloaded calibration curve and instead implement a single 32 (or more) point transfer curve, I think TS could do a nice job of auto-correcting this curve. It would need to collect EGO error values for each point in the curve and adjust the data points based on the error. Very similar to the VE auto-tuning with VEAL.
Steve Hauer
http://www.77e21.info
1977 BMW 320i with MAF, COP, IAC, Sequential fuel and spark, MS3 knock sensing
MS3 / MS3X / V3 mainboard, MS3 V1.5 a4
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MAF wish list

Post by muythaibxr »

I don't think just doing grams per second vs RPM will work as well since the range in grams/sec you can see at various RPMs is likely to be drastically different. Even if MAFMAP isn't accurately representative of real MAP, it at least normalizes the "load" across RPM so that having tables of the size we use is feasible.

I feel like if we use g/sec vs RPM, we'll end up using the lower few g/sec bins for all loads at low RPM, then the middle few in the middle RPM ranges, etc... so you'll lose resolution, where MAFMAP should allow you to use most of the table as you would with normal speed density.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

I agree with Ken on this, even though it might sound good at first you would end up with a very narrow diagonal region of interest.

So even though "MAFMAP" might sound bogus, would it should give you is a repeatable means of apply a table based adjustment. Just tune where the dot is.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
shauer
Experienced MS/Extra'er
Posts: 364
Joined: Wed Oct 05, 2005 10:39 am
Location: Rochester, NY USA
Contact:

Re: MAF wish list

Post by shauer »

jsmcortina wrote:I agree with Ken on this, even though it might sound good at first you would end up with a very narrow diagonal region of interest.

So even though "MAFMAP" might sound bogus, would it should give you is a repeatable means of apply a table based adjustment. Just tune where the dot is.

James
Ok, I can see the logic in that. Never thought of it from that point of view.

If MAFMAP is necessary for the reasons stated above, then I would just ask that it is used everywhere that load is required. Right now I suspect that MAP is still used in a few areas where you do not get to specify primary load.
Steve Hauer
http://www.77e21.info
1977 BMW 320i with MAF, COP, IAC, Sequential fuel and spark, MS3 knock sensing
MS3 / MS3X / V3 mainboard, MS3 V1.5 a4
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MAF wish list

Post by muythaibxr »

shauer wrote:
jsmcortina wrote:I agree with Ken on this, even though it might sound good at first you would end up with a very narrow diagonal region of interest.

So even though "MAFMAP" might sound bogus, would it should give you is a repeatable means of apply a table based adjustment. Just tune where the dot is.

James
Ok, I can see the logic in that. Never thought of it from that point of view.

If MAFMAP is necessary for the reasons stated above, then I would just ask that it is used everywhere that load is required. Right now I suspect that MAP is still used in a few areas where you do not get to specify primary load.
No, if MAF is selected for fuelload, then anything that keys on fuelload uses MAFMAP automatically.

For anything else (primary ignition load, EAE load, AFR load, etc...) you have to select MAFMAP explicitly.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

Devs and\or maintainers awnser me these questions

Why is it called mafmap?
Is the mafmap variable a hold over from b&g firmware?
Has any of the devs and\or maintainers ever used the MAF implementation?
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

techsalvager wrote:Devs and\or maintainers awnser me these question
Please?
Why is it called mafmap?
It is a simulated MAP calculated from MAF.

Code: Select all

outpc.mafmap = ((MAFCoef / outpc.aircor) * outpc.maf) / outpc.rpm;
Is the mafmap variable a hold over from b&g firmware?
Per B&G request the whole implementation is taken from the B&G firmware.
Has any of the devs and\or maintainers ever used the MAF implementation?
I do not have a MAF.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

It is a simulated MAP calculated from MAF.

Code: Select all

outpc.mafmap = ((MAFCoef / outpc.aircor) * outpc.maf) / outpc.rpm;
Is it not even close to a simulated MAP, it acts exactly like a calculated load and follows exactly the torque output of the motor.
The higher the torque the higher the calculated load, and yes I can easily change peak calculated load output with my turbo setup.
Infact depending on gear and rpm and boost control and timing, calculated load peak can be in different places.

Per B&G request the whole implementation is taken from the B&G firmware.
Have you checked out the newer code as it has changed compared to the older version.

We can see that b&g states abs load in their new mafmap guide.

http://www.microsquirt.info/mafmap.htm



"Improved range of MAF airflow calculation and corrected calculation of injector pulse width:

Old Equation:
PW = Topen + MAP x VE x ReqFuel

New Equation (model-based):
PW(sec) = Topen + ((MAF(g/s) x (120 / Rpm) / No_cyl) / AFRTGT) / InjFlowRate(g/s)

This is important because now AFRTGT has a big effect on fueling and ReqFuel is used only with MAP."

^

http://www.useasydocs.com/details/usv3code.htm


Thanks you for answering the questions james.
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: MAF wish list

Post by jsmcortina »

techsalvager wrote:Have you checked out the newer code as it has changed compared to the older version.
The source is unavailable so no I haven't.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
techsalvager
Master MS/Extra'er
Posts: 443
Joined: Sat Jun 28, 2008 3:47 am

Re: MAF wish list

Post by techsalvager »

jsmcortina wrote:
techsalvager wrote:Where did I mention anything about changing AFR values to unrealistic values?
Just the other day when you were saying that you couldn't use closed loop EGO because you had tuned your engine by altering the AFR table away from the real values - or was that some other user?

James
You indeed were reading what I was posting, but I don't believe you were understanding what I was trying to convey.

As I have stated before the commanded fuel table and the closed loop fuel table are the same table called the AFRtarget table.
You can not have wideband controlled closed loop fueling and your main commanded fuel tables different as they are the same one.
See the issue.

like I mentioned I want to keep stoich going into boost 2-3-4 psi, but when I want to go I want it to drop fuel in there. thats why you need seprate maps.
I've already mentioned in a previous post about one possible way to do maf tuning.
MS2extra
Mass Air Flow fuel and spark
Miata 1.6 turbo
TechSalvager is banned for ignoring admin warnings not to link to his childish videos.
Post Reply