Title: Torque model on MAFLESS files Post by: orienz on October 23, 2011, 04:29:23 AM Hey guys,
I am trying to modify torque control maps after I upgraded to 80mm HEMI throttle body. I am using Eurodyne Maestro editor and under torque control we have 4 maps available. - "Accelerator pedal map" (determines how much torque is requested at what pedal %, x,y axis are RPM and Pedal %, values are in Torque %). - "Throttle angle vs airflow" map (marks the mass airflow through TB at certain angle. x,y axis are RPM and Airflow (kg/hr), values are in TPS). This maps also reflects AlphaN map that we have under "Fueling". - "Target filling" map (determines load at requested torque % @ certain RPM ?, x,y axis are Torque %, RPM, values are in Cylinder filling % (i think this is load %)). - "Optimum engine torque" map (No idea what this map does, it seem to be reflecting Target filling map, x,y axis are % Cylinder filling, RPM...values are in Torque %). So those are the maps we can modify to dial in larger throttle body since we do not have MAF on these files. Right now fueling is all over the place of course due to more air entering the engine under same TB angle. How would you approach this problem? I found most of these maps in RS4 stock file (which I believe has 75mm throttle body). Could I use this somehow? Or different characteristics of the engines will throw these off totally? Thanks for any help. Title: Re: Torque model on MAFLESS files Post by: nyet on October 23, 2011, 02:15:12 PM Stupid question:
Why not start with the throttle angle vs airflow map? Or am I missing something? Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 23, 2011, 02:30:21 PM Going mafless IMO is a big no no with ME7... it's always best not to trick it.
That said, I had the chance to use Maestro yesterday and wow what a POS software that is. Title: Re: Torque model on MAFLESS files Post by: nyet on October 23, 2011, 02:37:18 PM I agree
Title: Re: Torque model on MAFLESS files Post by: orienz on October 23, 2011, 03:22:18 PM Yea I agree that going MAFLESS is not the best thing to do but well...it's the only option right now until I recirculate the blow off valve.
nyet, is there ANY way to calculate this map? I wouldn't know what values to start with to be honest. Shall I just...guess? Title: Re: Torque model on MAFLESS files Post by: nyet on October 23, 2011, 04:47:25 PM Dunno. Honestly, you're looking for a world of hurt. The amount of knowledge/effort to get ME7 mafless is likely to far exceed just tuning it properly.
Title: Re: Torque model on MAFLESS files Post by: orienz on October 24, 2011, 02:50:51 AM This same mafless file has been running flawless on stock throttle body, it's upgrading to 80mm TB that threw everything off :-/
Title: Re: Torque model on MAFLESS files Post by: ejg3855 on October 24, 2011, 05:29:55 AM Yea I agree that going MAFLESS is not the best thing to do but well...it's the only option right now until I recirculate the blow off valve. Blow Thru :) Title: Re: Torque model on MAFLESS files Post by: orienz on October 24, 2011, 10:12:28 AM What's that?
Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 24, 2011, 10:20:44 AM It's when the MAF is installed in the charge air path
Title: Re: Torque model on MAFLESS files Post by: orienz on October 24, 2011, 10:49:30 AM Isn't that a bit unreliable? and won't MAF get wrong readings like that? since air enters under pressure there...
Title: Re: Torque model on MAFLESS files Post by: s5fourdoor on October 24, 2011, 11:00:43 AM man, you should just look into running a pro-m maf. way more upside in that than eliminating the MAF logic...
Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 24, 2011, 11:02:45 AM Isn't that a bit unreliable? and won't MAF get wrong readings like that? since air enters under pressure there... I'm sorry but don't you find it funny you're worried about wrong readings when you have DELETED your MAF on a system that relies on a MAF sensor as a PRIMARY load input and also was never designed to run MAFless? Either way, no it won't get wrong readings if you've properly scaled your MAF for that setup. Title: Re: Torque model on MAFLESS files Post by: s5fourdoor on October 24, 2011, 11:08:46 AM call pro-m racing, they could solve this problem using a common mustang maf. you'd know your exact kg/hr - buy the extra 30point calibration sheet for $75. all you need is an interpolation to go from 30point to 512 point used in ME7. i'm going to do this. in fact, i bought 4-pin connectors so that i can hot-swap the bosch maf, hitachi maf, and eventual pro-m maf.
no brainer if you ask me. run a stock airbox / maf in the winter for mountain-ish, switch back to the high-flow maf / open-filter during warmer weather. not much different than swapping winter tires to gumballs+wheels.... Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 24, 2011, 11:19:00 AM sounds complicated.. why not just run the pro-m all year round?
Title: Re: Torque model on MAFLESS files Post by: s5fourdoor on October 24, 2011, 11:27:49 AM sounds complicated.. why not just run the pro-m all year round? you could... either way its better than no MAF. LOL... Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 24, 2011, 11:29:48 AM agreed
Title: Re: Torque model on MAFLESS files Post by: Gonzo on October 24, 2011, 11:32:44 AM MAF-less on ME7.1 is a big no no but on ME7.5 its doable.
Its still a headache even with ME7.5 though. Title: Re: Torque model on MAFLESS files Post by: orienz on October 24, 2011, 01:58:46 PM Ok I got it dialed in pretty good now I think. Bucking is completely gone. Here is a log file of a WOT run. O2 correction is pulling lots of fuel...I am running rich :-/
Also I have NO idea why it isn't correcting in the begining, it tays 0.75 for a long time. Log: https://docs.google.com/spreadsheet/ccc?key=0AmwyVuNKgXtjdDV4eTVPLUViX1k1N0VmSUtBRUxKbWc&hl=en_US (https://docs.google.com/spreadsheet/ccc?key=0AmwyVuNKgXtjdDV4eTVPLUViX1k1N0VmSUtBRUxKbWc&hl=en_US) Here is another log of part throttle, just cruising around: https://docs.google.com/spreadsheet/ccc?key=0AmwyVuNKgXtjdHprdzNpNU5EZ2NKUkJaNTZQNElraXc&hl=en_US (https://docs.google.com/spreadsheet/ccc?key=0AmwyVuNKgXtjdHprdzNpNU5EZ2NKUkJaNTZQNElraXc&hl=en_US) Some input would be helpful :-) I am just getting started with this stuff. ps: NOTORIOUS VR, I know that MAF delete is a wrong move but it's the only move right now. Gonzo, can you please remove that picture from your signature? Very disturbing and annoying (way too big). Thank you. Title: Re: Torque model on MAFLESS files Post by: nyet on October 24, 2011, 02:08:14 PM Also I have NO idea why it isn't correcting in the begining, it tays 0.75 for a long time. If req. AFR isn't lambda 1, you are open loop Quote {links to google docs} Pretty much worthless. Post the original CSVs, and/or graphs from ECUxPlot or xls templates (there are several floating around) Quote Some input would be helpful :-) I am just getting started with this stuff. ps: NOTORIOUS VR, I know that MAF delete is a wrong move but it's the only move right now. "right now". Then wait till you can make the right move :P Trying to get this working is pointless unless you aren't a noob. The only reason ME7 works at all MAFless is so the car is basically drivable till you can get it to a shop to fix the MAF. Title: Re: Torque model on MAFLESS files Post by: orienz on October 25, 2011, 02:18:57 AM nyet, you realize that this is ME7.5 with wideband O2 right? And you realize that there are several large tuning companies selling mafless files for this ECU? So it's totally doable.
Requested lambda wasn't 1 further down either and O2 is correcting. So it is closed loop all the time... Not trying to sound bad but you people on this forum are a bit too stuck on S4 cars and ME7.1 ECU. Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 25, 2011, 07:06:16 AM nyet, you realize that this is ME7.5 with wideband O2 right? And you realize that there are several large tuning companies selling mafless files for this ECU? So it's totally doable. So what? You're still tricking the ECU... The ECU still wasn't meant to run without a MAF. Just because something is doable, doesn't mean it's the right way to do it. Quote Not trying to sound bad but you people on this forum are a bit too stuck on S4 cars and ME7.1 ECU. that has nothing to do with it. We're giving you this advise because you're trying to fool ME. Generally that is not a good thing to do when you're talking about deleting the primary load input. Title: Re: Torque model on MAFLESS files Post by: nyet on October 25, 2011, 09:33:23 AM nyet, you realize that this is ME7.5 with wideband O2 right? And you realize that there are several large tuning companies selling mafless files for this ECU? So it's totally doable. Requested lambda wasn't 1 further down either and O2 is correcting. So it is closed loop all the time... Not trying to sound bad but you people on this forum are a bit too stuck on S4 cars and ME7.1 ECU. Fair enough. I didn't know it was a wideband project. Title: Re: Torque model on MAFLESS files Post by: vwaudiguy on October 25, 2011, 10:04:37 PM Fool ME? You act like it's alive. ;D You're creating a target afr and based on the wideband it follows it. Now you're relying on the 02..Maf's fail/become inaccurate over time too. I have a buddy running a large company's mafless file that drives 95% identical to when he had a maf. And the other 5% that is slightly different is very low load/rpm which he's almost never at. I personally can't feel the difference, but he says it's very slight but there. Well worth the lack of intake restriction and bay cleanliness of having a filter mounted on the turbo.
Title: Re: Torque model on MAFLESS files Post by: NOTORIOUS VR on October 26, 2011, 05:35:36 AM It's much easier for an O2 sensor to go bad or read wrong undetected then it is a MAF sensor.
And yes you're fooling ME... Say what you want but it was never intended to be run Alpha-N style. It's primary load input is the MAF sensor, everything else is secondary. Trusting your motor to ONLY sensors which are only able to read AFTER the combustion cycle has already happened is well stupid IMO. You're at this point relying heavily on your knock sensors to save your engines butt if the O2 sensors are wrong. Title: Re: Torque model on MAFLESS files Post by: vwaudiguy on October 26, 2011, 09:23:31 AM Actually on the 1.8t's I see MAF's go left and right, and that's on stock cars not big turbo where there's a lot more airflow... and rarely primary 02's. Not trying to argue just wanted to say I know of people running happily without a MAF for years with no issues and good gas mileage/driveability. I agree with you that you are going against the architecture of Motronic working Alpha-N but it seems to work out in the end. And it's not my car anyway! haha
Title: Re: Torque model on MAFLESS files Post by: Giannis on October 26, 2011, 10:13:37 AM i am one of those people for a while. :)
Title: Re: Torque model on MAFLESS files Post by: hammersword on December 02, 2011, 06:18:58 PM MAFless can be done with the same way and same results at ME7.1 and ME7.5
If you have no idea about injectors calibration or injection calibration then yes ME7.5 will do its best to save the engine or try to follow the target because it has wideband! If you can calibrate fine the injectors and tune well then injection calculation disabling the fuel trims then you can have an excellent MAFless with NO issues. But this need experience on ME7 Title: Re: Torque model on MAFLESS files Post by: Giannis on December 02, 2011, 11:07:22 PM Or a lot of testing i may add.
Title: Re: Torque model on MAFLESS files Post by: aef on August 22, 2012, 01:35:39 AM Hi,
can someone please give a statement regarding the pros and cons of mafless? And i dont mean clean engine bay ::) When is it really necessary to go mafless? Cheers Title: Re: Torque model on MAFLESS files Post by: prj on August 22, 2012, 03:06:51 AM Title: Re: Torque model on MAFLESS files Post by: nyet on August 22, 2012, 09:38:13 AM When is it really necessary to go mafless? When you don't have the equipment or knowledge to tune a MAF based ECU. Often people unwilling to learn Motronic opt for standalone. There are also race (or high boost) applications where Motronic will not cut it, and the only reasonable (or affordable) way to do it is with a standalone MAFless ECU. There is definitely NO reason to use Motronic MAFless, IMO. Title: Re: Torque model on MAFLESS files Post by: yosh on September 01, 2012, 07:14:54 AM MAFless can be done with the same way and same results at ME7.1 and ME7.5 If you have no idea about injectors calibration or injection calibration then yes ME7.5 will do its best to save the engine or try to follow the target because it has wideband! If you can calibrate fine the injectors and tune well then injection calculation disabling the fuel trims then you can have an excellent MAFless with NO issues. But this need experience on ME7 +1 ^ Title: Re: Torque model on MAFLESS files Post by: jdm ej1 95 on September 01, 2012, 02:19:04 PM there are plenty of cars out there using factory MAF ecus running mafless... i dont know how MAF based they are.. but GTRs come into mind.. evos.. sti's most end up going speed density on factory ecus without issue. its kind of commonplace
Title: Re: Torque model on MAFLESS files Post by: prj on September 01, 2012, 03:14:34 PM there are plenty of cars out there using factory MAF ecus running mafless... i dont know how MAF based they are.. but GTRs come into mind.. evos.. sti's most end up going speed density on factory ecus without issue. its kind of commonplace Those ECU's have speed-density as a control strategy built in. On ME7 it is still limp mode. |