M42club.com - Home of the BMW E30/E36 318i/iS
DISCUSSION => General Topics => Topic started by: bmwman91 on January 21, 2011, 08:59:58 PM
-
Some of you know that I have been tinkering around with a MAF conversion for a few years now. Initially I wanted to develop it into a product, but that idea died when it became obvious that a MAF won't really do anything for the M42's power output. Well, I have been expanding some of my converter's functionality & decided to see if the current idea was popular.
At this point it might be a desirable replacement option for 20+ year old AFMs being that they can cost anywhere from $160 to $700 to replace (or $40 at a local wrecker, but you never know what you'll get). This may change, but I am looking at a ~$200US price point for the converter only (subject to change, up or down). The MAF sensor, & hardware to make it fit on the car, are not things I am planning to manufacture or sell at this point in time.
Here are the features I would start out with:
No-nonsense MAF conversion for M42 cars
- Some cutting of factory AFM harness
- Extra connector parts so that stock AFM could be plugged back in with minimal effort
- Overall design such that MAF / AFM could be interchanged in ~5 minutes with no need for tools
- About 20ms latency with filtered output
Selectable output variable
- Puts out either mass flow rate, or volumetric flow (AFM-emulation) rate
- Choice of filtered or unfiltered output (filtered is a good idea unless you know a LOT about the ECU you are working with, and is a must with Motronic)
Tunable output parameters
- With volumetric output (AFM-emulation), choice of stock-style logarithmic output or linear output
- - Tunable logarithmic & linear outputs
- With mass flow rate output, linear output
- - Tunable linear output
PC Graphical Interface
- Used to select output mode and tune output curve parameters
- Can log & plot filtered+unfiltered air flow and intake air temp in real-time
- - Thinking about having analog input for wide band & narrow band O2 sensors - not 100% sure yet
- Used to apply firmware updates (not 100% sure about this yet)
The current sensor can handle flows up to ~810 kg/hr (according to the data sheet), but I think that somewhere in the 500-600 kg/hr range is more reasonable. A stock M42 will generally flow ~350kg/hr at most.
When I say tunable output curves I mean:
- For linear output, Vout = m * FLOW + b with m & b being adjustable
- It's a similar deal for the log output - only 2 parameters to adjust
The purpose of the linear outputs would be to make tuning with things like the WAR chip / TunerPro and running MS easier.
The timeline for this would be a few months. I have a day job & I don't plan on quitting any time soon!
-
I would be interested in at least one...maybe more. If it will allow for more flow and then to be used with the warchip and in turn a turbo or SC setup then I would be all over it. Let me know if you need someone to help test it.
I do m42 swaps into 2002s and soon an e21 and have been wanting to add a bit more power....
-
very interested. this can help alot with my turbo build.
-
Very interested! Subscribed.
-
I just want to make sure that people are aware that a MAF will not really get them power gains. The AFM is not restrictive, believe me. Multiple dyno pulls proved that when I was developing an earlier version, years ago. You DO get better idle stability & throttle response though, particularly when coming off of idle.
M20 guys claim gains from a MAF, but that is only because the stock AFM was too small to begin with. The power gains from a MAF on the M20 are the same as putting an M30 AFM on. A certain other MAF conversion that will remain nameless claims gains too...but a chip comes as part of the package. The gains come from the chip, and are remarkably similar to the power gains seen when one adds JUST a chip.
I just want to be 100% up-front & honest with this. I don't want to sink a bunch of my own money into developing a product & then have people pissed off because they expected something & didn't get it!
-
I just want to make sure that people are aware that a MAF will not really get them power gains. The AFM is not restrictive, believe me. Multiple dyno pulls proved that when I was developing an earlier version, years ago. You DO get better idle stability & throttle response though, particularly when coming off of idle.
M20 guys claim gains from a MAF, but that is only because the stock AFM was too small to begin with. The power gains from a MAF on the M20 are the same as putting an M30 AFM on. A certain other MAF conversion that will remain nameless claims gains too...but a chip comes as part of the package. The gains come from the chip, and are remarkably similar to the power gains seen when one adds JUST a chip.
I just want to be 100% up-front & honest with this. I don't want to sink a bunch of my own money into developing a product & then have people pissed off because they expected something & didn't get it!
But with boost the AFM does become restrictive, or so I heard.
-
But with boost the AFM does become restrictive, or so I heard.
With boost, it is another story entirely.
a) The M42 AFM maxes out at ~400 m3/hr (stock M42s pull up to 360 m3/hr). If you put the AFM behind the turbo, you can sort of cheat since the density is higher, but you need to tune the fuel properly to account for that.
b) Yes, it probably is a restriction when pulling more than ~400 m3/hr. 400 m3/hr equates to ~1.7 PSI on an internally stock M42 lol.
-
With boost, it is another story entirely.
a) The M42 AFM maxes out at ~400 m3/hr (stock M42s pull up to 360 m3/hr). If you put the AFM behind the turbo, you can sort of cheat since the density is higher, but you need to tune the fuel properly to account for that.
b) Yes, it probably is a restriction when pulling more than ~400 m3/hr. 400 m3/hr equates to ~1.7 PSI on an internally stock M42 lol.
So what would be the ideal Maf the replace the AFM with? The m44 unit? or the m3 unit?
-
not really looking for a m42 perfomance upgrade , but this sounds like i may want this for my project turbo m20.. put me down for interest there !!
-
So what would be the ideal Maf the replace the AFM with? The m44 unit? or the m3 unit?
The one I have been working with is a Bosch from a ~2003 Jetta Turbo. It is considerably cheaper than any of the ones used on BMWs (as far as I can tell). I was talking with another member about higher flow ones for FI use, and I may make it an option to use the $125 VW one (you could run ~9PSI boost at redline with it), OR the $260 one from something like a 540i (higher flow, not sure what the max is yet). Anyone who NEEDS the higher flow one has likely already dropped enough $ on their project that the extra $130 for the sensor probably won't be as much of an issue.
not really looking for a m42 perfomance upgrade , but this sounds like i may want this for my project turbo m20.. put me down for interest there !!
M20 stuff is definitely planned. Once I figure out exactly how many M20 AFM variants there were & get some data on them, it will be a relatively simple software update release to have this be PnP on M20 cars. Even before that, this thing will be tunable enough that someone with an M20 could make it work though. In linear output mode, you could basically use this on ANY car if you have a stand-alone or other ECU hacking/tuning hardware.
-
I think something that might work for everyone is the MAF from a 95 m3....which is the same as the 325/525/etc etc They can be bought used fairly inexpensively and brand new for right around $250. Used around $100 or less. There are some non Bosch units brand new for around $70 as well... Not sure what the flow rate is on them tho.
The other options would be the one from a 540 or something that even allows more flow is the 803 maf which is from a 993 porsche (about $250 brand new). The 803 will flow enough for over 400hp turboed. So it might be a bit excessive.
-
Thanks for the information. It sounds like having two MAF options might be pretty popular. I can find a way to allow the unit to be switched between them in the control software. That way everyone is happy. I will see what I can dig up on the 540 & 993 MAFs.
-
.
-
Maybe...but first you need to post up pictures!
-
..
-
..
-
Holy crap, that is seriously badass.
-
.
-
Damn, do you have a turbo AND SC M42? Those are some freaking sweet looking builds!
-
.
-
no the supercharger is my old motor and the turbo is on my new motor... going to buy a junker strip it and put my old set up in it...
Have you considered selling the supercharger?
-
soon ill have a dasc for sale in the near future fully rebuilt with powder coat will be asking $2400.
-
M20 guys claim gains from a MAF, but that is only because the stock AFM was too small to begin with. The power gains from a MAF on the M20 are the same as putting an M30 AFM on. A certain other MAF conversion that will remain nameless claims gains too...but a chip comes as part of the package. The gains come from the chip, and are remarkably similar to the power gains seen when one adds JUST a chip.
I'm really glad to see this happening. Coming from such a well-regarded member of the board I'm really looking forward to seeing this happen. I'd suggest that you don't forget the M30 guys (there are a surprising number still out there).
I had a similar idea a long time ago, but I was thinking along the lines of a Megasquirt/Ford Mustang MAF/integrated IACV to eliminate a few of the M42's problems. Your idea is far more economical and has many more advantages - much cheaper, stock hardware, no cutting the stock harness and it can pass enhanced inspections as it looks relatively stock.
-
soon ill have a dasc for sale in the near future fully rebuilt with powder coat will be asking $2400.
:O I'm getting my wallet ready.
-
I'm really glad to see this happening. Coming from such a well-regarded member of the board I'm really looking forward to seeing this happen. I'd suggest that you don't forget the M30 guys (there are a surprising number still out there).
I had a similar idea a long time ago, but I was thinking along the lines of a Megasquirt/Ford Mustang MAF/integrated IACV to eliminate a few of the M42's problems. Your idea is far more economical and has many more advantages - much cheaper, stock hardware, no cutting the stock harness and it can pass enhanced inspections as it looks relatively stock.
It should be fairly easy to get the M30 AFM working here too.
There is a small amount of cutting required on the stock AFM harness, but you are basically sticking GM Weather-pak connectors on & the kit will include an extra one that you attach to the original AFM connector so it can all be put back together.
-
I'm definitely interested. My m42 beater is almost ready and I would to switch over to MAF on that car.
-
Quick update...
I am almost done with the toughest part of the project: developing a firmware update system. I want to be able to release firmware updates just in case some sort of unforeseen bug appears, or if I want to add features later / etc. This is the most time consuming part though, because it has to work perfectly. I can't be sending out updates that turn the converter into a brick!
-
.
-
bmwman91 why not go threw the eprom code and adjust the constants for the AFM and scale for the MAF ?
I will gladly help out I just can't be testing such Ideas on customer cars due to time consuming issues, but if someone wants to be a Guinna pig I will help out the best way I can
To me is just sounds better if you stick the MAF chip in the DME and then plug the MAF in and GO!
Kind Regards|
-
.
-
well i got the miller war chip and its all bad i dont want to say anything bad just yet. but if you pm me ill tell you the whole drama im going thu right know ....
please enlighten us :)
-
so what's with the maf lets make this happen ill pay you if you can make it work.
Oh, I wasn't planning on giving them away. :p It'll be competitively priced with other products of this nature.
bmwman91 why not go threw the eprom code and adjust the constants for the AFM and scale for the MAF ?
I will gladly help out I just can't be testing such Ideas on customer cars due to time consuming issues, but if someone wants to be a Guinna pig I will help out the best way I can
To me is just sounds better if you stick the MAF chip in the DME and then plug the MAF in and GO!
Kind Regards|
Editing the Motronic code won't really work. It is set up to take in volumetric flow rate and air temperature, from which it computes mass flow rate. Editing the maps only lets you scale the behavior. What you would need to do to get the ECU to run a MAF properly would be edit the executable program code and remove the whole IAT / mass flow rate calculation section. THAT is something that COULD be done, but would be far more work than it is worth (unless someone is REALLY good friends with Jim Conforti & can get him to give out the annotated assembly code he came up with back when he was writing this stuff for Bosch). Chances are that he isn't about to do that though!
-
Oh, I wasn't planning on giving them away. :p It'll be competitively priced with other products of this nature.
Editing the Motronic code won't really work. It is set up to take in volumetric flow rate and air temperature, from which it computes mass flow rate. Editing the maps only lets you scale the behavior. What you would need to do to get the ECU to run a MAF properly would be edit the executable program code and remove the whole IAT / mass flow rate calculation section. THAT is something that COULD be done, but would be far more work than it is worth (unless someone is REALLY good friends with Jim Conforti & can get him to give out the annotated assembly code he came up with back when he was writing this stuff for Bosch). Chances are that he isn't about to do that though!
I know exactly how to do the above /\ and I am more then familiar with disassembley of the code... intel 8051 8bit processors are a breeze compared to the newer stuff... If ever you want to collaborate on something let me know and I can give you hand but otherwise cool idea and best of luck :)
-
I know exactly how to do the above /\ and I am more then familiar with disassembley of the code... intel 8051 8bit processors are a breeze compared to the newer stuff... If ever you want to collaborate on something let me know and I can give you hand but otherwise cool idea and best of luck :)
I tried disassembling the code a few years ago. I was a lot less savvy with microcontrollers then, but it looked like a nightmare to separate the LUTs & other constants stored in with the machine code instructions. When I am done rolling out the MAF conversion, I would be interested in seeing what can be done. I have always dreamed of completely re-writing the Motronic firmware to add some features & work with other types of sensors. I have the datasheet for the M175's uC, so it isn't out of the realm of possibility.
-
I tried disassembling the code a few years ago. I was a lot less savvy with microcontrollers then, but it looked like a nightmare to separate the LUTs & other constants stored in with the machine code instructions. When I am dont rolling out the MAF conversion, I would be interested in seeing what cab be done. I have always dreamed of completely re-writing the Motronic firmware to add some features & work with other types of sensors. I have the datasheet for the M175's uC, so it isn't out of the realm of possibility.
shoot me a email with that data sheet and I will send you back a pdf file you have never seen before ;)
-
Come on guys, share the information. I have been to many local meets with Jim C and he won't share any of the information either.
I'm interested in the MAF kit, but I wouldn't be able to get it for a couple months.
-
Come on guys, share the information. I have been to many local meets with Jim C and he won't share any of the information either.
I'm interested in the MAF kit, but I wouldn't be able to get it for a couple months.
If I share the information you will need to be busy for the next 5 years :)
-
..
-
You are the man thanks for helping me with the problem on my miller war chip ews decoding ....if it wasnt for you i wouldn't no know what to do !!!!
no problem glad to hear you got it working :)
-
If I share the information you will need to be busy for the next 5 years :)
I don't understand what you mean by this.
-
Sorry I haven't updated in a little while, my "real" job and life have been hectic. Work is still going on this, and I am hoping to finish testing all the boot-loader stuff this weekend. From there, I need to work out the tuning section. At this point, I WILL make it possible to use different MAFs. As I get parameters for new sensors, I will be able to release MAF data files that can be imported into the GUI and then loaded onto the converter.
Thereafter, I will finish up the firmware for the converter's controller. THat will probably take a week or two. Luckily, I already have the main portion of the code written. I just need to add in the interface stuff for tuning.
Finally, I will need to actually BUILD this. The hardest part will be sourcing molded cable strain reliefs and finding an appropriate enclosure for this. I have a few leads on companies that do this sort of thing for small-medium volume products. This is all being funded personally, which makes things a little trickier. If anyone knows of a good source of reasonably priced multi-conductor cable, strain reliefs & small water-tight enclosures let me know!
Thanks!
-
I would like to update you too, I will probably be your first customer, I have ran into some money and I really would like to fix that little 1-2 second delay in throttle response, causes me to slip my clutch a lot uphills.
-
Haha thanks!
If you have a 1-2 second delay in response though, I think there are other things awry with the car!
-
Haha thanks!
If you have a 1-2 second delay in response though, I think there are other things awry with the car!
I'm pretty sure it's my ugly AFM that I refuse to clean because it's a piece. The response is fine once I'm off idle.
-
Just a teaser...I have been making good progress this week.
(http://www.e30tuner.com/other/mafconint_00.png)
-
ahhhh SHIIEIEEtt
-
Very impressed! keep up the good work
-
I'm excited.
-
You know,,, this would ROCK for tuning your car to run on E10,E15 or even E85...
Dave
-
Yup, combining this with a tunable WB02 setup like the LC-1 would let you ru nwhatever fuel you wanted (that works properly in an Otto cycle motor).
-
Just a little more preview action. The windows software is basically done. Now I need to polish off the firmware for the actual controller so I can test the software with the "real deal!"
(http://www.e30tuner.com/other/mafconint_01.png)
(http://www.e30tuner.com/other/mafconint_02.png)
(http://www.e30tuner.com/other/mafconint_03.png)
-
very nice!!! keep it up
-
One additional update:
I think I am going to ditch the serial stuff in favor of USB. It'll add a little cost to the unit, but it'll certainly be less than the cost of a USB-serial adapter you'd get online or in a store. It would be a pain if I required users to have an available serial port, since I think some people might already be put-off by the fact that you need a computer to configure the thing. At least every laptop made in the last 15 years has USB ports!
-
SO how does it feel? pull any harder?
-
On a stock M42, there is unlikely to be any major gain in HP. The stock AFM is well-sized, and contrary to popular belief, the flapper door is not restrictive (there's maybe a 0.13 PSI drop across it...the TB & head lead to whole-digit PSI drops). The stock AFM is also running at 90%+ of its max capacity on a stock M42 (which is good design practice from BMW's end). That leaves very little room for built engines & forced induction.
However, I have found that idle stability improves, and the throttle response coming off of idle is noticeably better. All around, throttle response seems to benefit. This is likely because the MAF basically provides instantaneous output with no overshoot, versus the AFM that has a delay somewhere around 50ms and some major overshoot, so really it takes 50-100ms for the AFM to start giving proper output after a rapid throttle adjustment.
(note: if your O2 sensor is shot or you have vacuum leaks, your idle will still suck lol)
-
ahaha, Im assuming though if you have ITBS or cams that this will reap more benefits?
-
As far as power goes, the MAF conversion allows you to gain power by being able to meter more air. If you have a stock M42, dyno it, install a MAF (properly tuned) & dyno it again, I wouldn't expect there to be much of a difference.
With hotter cams & ITBs, you probably will stand to gain power with a MAF. Once you start flowing more than the stock design, the AFM can begin to be restrictive in a significant way since it was only designed to be non-restrictive in the stock application.
In short, yes a motor that has had intake system & internal mods will stand to gain power from the MAF conversion.
-
(http://i100.photobucket.com/albums/m11/tazy_01/IMG_3345-1.jpg)(http://i100.photobucket.com/albums/m11/tazy_01/IMG_3466.jpg)
-
any updates?
-
I'm still working on the conversion firmware. It's coming along pretty well.
What's going on in the pics above? Is that a flow calibration rig, or some kind of turbo inlet pipe?
-
As for the hardware, would you need something like this?
(http://i119.photobucket.com/albums/o143/clayweiland/IMG-20110331-00013.jpg)
I have one of Scottie Sharpe's pieces, but I'm not going to use it. See here:
http://www.m42club.com/forums/showthread.php?t=13328 (http://www.m42club.com/forums/showthread.php?t=13328)
-
Sorry about the huge gap in updates. A couple of engineers on my team (at my day-job) quit in the last couple of months, so things have been super busy while we try to replace them.
As for the MAF kit, I have been making slow but steady progress. All the software is complete, although it still needs a lot of testing. I also built a hardware prototype to test the software with in the actual environment. I haven't forgotten about this by any means, but my day-job still puts food on the table so I can't go neglecting that either!
-
Still interested, and hopefully I can convince my (very soon to be) wife that putting a MAF on Betty will be worth the moolah, so I can be a lab rat on this when it's ready for testing.
-
The MAF should be reasonably easy to justify. Try telling her you want a fully built Metric Mechanic Rally M42 w/ options for a cool $11k!
-
Would having this piece of CNC'd awesomeness help?
(http://i119.photobucket.com/albums/o143/clayweiland/IMG-20110331-00013.jpg)
I don't need it and it should really go to someone that can use it. I'll sell it for $150 or trade for an M42 COP kit.
clay dot weiland at gmail dot com
-
At the moment I might not be able to make use of it. I am looking at using inexpensive off-the-shelf Bosch MAF sensor units & finding the simplest way possible to mate them to the airbox (or just put an open element filter on them).
-
Well, I have made a lot more progress on this. Software testing is almost done, and I have ironed out a LOT of bugs. This thing is starting to be really solid. I have been delayed by nuisances like having to tear my M42 down thanks to a dead timing chain idler sprocket, troubleshooting a bad ground on my wide-band O2 sensor setup and trying to convince my fiancee that she is at least as important as my car projects.
Anyway, I expect to be designing & ordering some debug PCBs in the next couple weeks so I can really iron the last details out. From there, it is just working out the converter's packaging & cabling. I know this has been a long time coming, but it should be well-worth the wait.
-
Any updates?
-
Yup, I am working out the final elements of the PCB design and working out a deal with a contract manufacturer (in Canada) for assembly. I still need to source some cables with strain reliefs, but the bulk of the development work is done.
-
Now I’m doing MAF conversion on my M42 (E36)
I bought M52B28 MAF sensor and VAF-MAF converter from here
http://shop.pilotpowersupply.com
-
Hi , I am interested in the MAF conversion and would like to know if the completed units will be programmed with the stock M42 calibration . At the moment I am running a stock engine and needs to replace my AFM but not necessarily interested in tuning a MAF sensor. Also what is the projected cost per unit ?
Thanks
Fabian
-
I will send the units pre-programmed for whatever car they are to be used on.
Currently, I am looking at $200-$225 for the converter module & harness, depending on the connectors people want.
-
Great pricing :)
-
Good to know. As someone that wanted a MAF conversion for YEARS, I know what seems reasonable & what doesn't. The two companies we all know of that make MAF conversion products have pretty steep price points. It always put me off. For the most part, I am trying to target the price of a rebuilt M42 AFM.
I want to make sure I am up front about the total cost though. The converter is $200-$220. The sensor & "making it fit" are not part of the package at the moment. A new MAF is like $120, and a lot less used. "Making it fit", at its basic level, will take a cone air filter.
-
So the big question is "When can I make a Purchase? "
-
I plan to have the first production run in my possession by the first week of September. I am still working out some sourcing issues on automotive connectors & trying to find a wiring harness assembly house (that I can afford). Those are the last few logistical issues I need to deal with.
-
Status
-
just matter of asking why cant we use 0-5v output maf ?
-
How is this going?