M20B25 Motronic 1.3 MAP Sensor Conversion

[GeoffBob]

IIRC correctly E30 M3's had an altitude sensor built in, mounted on the bulkhead, they use roughly the same ECU.

I know absolutely nothing about the S14 engine, but if they were tuned alpha-N (which I am guessing they were so as to negate the necessity for an AFM) then they would have had a pressure sensor open to atmospheric pressure (to compensate for changes in pressure brought about by weather and altitude).

Wouldn't the intake air temp sensor compensate quite well for this, ie high intake air temp = lower air pressure?

'Fraid not. A high intake temp could just as easily indicate that it's a hot day. BTW, as a "general" rule, temperature falls with higher altitude. But then it's not really altitude that we need to compensate for, its air pressure (which also decreases with altitude). Regardless of altitude, however, both air temperature and air pressure change regularly with the weather.

[e301988325i]

IIRC correctly E30 M3's had an altitude sensor built in, mounted on the bulkhead, they use roughly the same ECU.

I know absolutely nothing about the S14 engine, but if they were tuned alpha-N (which I am guessing they were so as to negate the necessity for an AFM) then they would have had a pressure sensor open to atmospheric pressure (to compensate for changes in pressure brought about by weather and altitude).

This confirms my thinking E30 M3 has an AFM, copied this post found by searching 'M3 S14 AFM'

thats about 3 hours from me any idea how long a remap takes

Hi Terry,

I've replied to your pm.. I can just remap your eprom in the ECU to give good results, but for the best results I would look at removing the AFM and fitting a Rongineer air box, TPS and remapping it with the Unichip. This is by far the best way to go to unleash power from your S14.

Jon_BMW was with me while I converted an M3 and saw how the Unichip works.. he is about to put up a thread about it, if he decides to pull his finger out

If you do a search under my username and Unichip or Dastek, you'll see plenty of info here.

BTW, its about 3 hours from you and about 2 hours back

and this linky confirms the presence of an on board barometer

http://www.e30zone.net/modules.php?name ... ude+sensor

[DmcL]

you´d need to create a circuit that will change the TPS voltage based on MAP voltage. This needs to be done electrically , maybe with a gain transistor or something. It´s no easy task and not really within the capabilities of the Motronic code either.

The is nothing wrong with a map sensor signal on a stock M20, it can be very steady, say 45-46kpa at idle as an example.

Basically with a MAP sensor or Alpha-N conversion you are just feeding a 0-5v signal into the AFM signal line to the ecu, from there you need to readjust scales and tables and table values to make the input work properly to give out the right pw and spark angles.

Its not simple as changing a MAF , there is a lot more to it in terms of adjusting within the ecu.

yea i know but it should be by no means impossible. i know the ECU wont know whats going on so to speak but all its looking for is that 0-5v input from something along with an IAT sensor input. the 0-5v input references a 1x8 map which, combined with RPM signal tell the ECU where to look on the part throttle maps. in theory with a clean, non fluctuating 0-5v input (from a MAP sensor in this case) i should only really have to re-tune the 1x8 map and maybe make some small adjustments to a couple fuel maps. u could imagine the 8 points in the 1x8 map as being spread across the linear voltage rise of the stock AFM, to change for a curving output the points should be able to be adjusted to essentially match the curve, this is what ive done with a MAF already. should also be the case for running a variable TPS with MAP sensor for baro correction. motronic 1.3 wasnt designed for MAF/MAP/Alpha-N but the way it works means it should only be a matter some fine tuning in fuel maps and re-tuning of that 1x8 map which is basically like the translator between the rest of the load related maps and whatever airflow/pressure metering device is fitted.

It´s no easy task and not really within the capabilities of the Motronic code either.

But well within the capacilities of a good aftermarket ECU, which is what I think you should be looking towards DmcL. I know that these goodies come at a price, but they also usually come with instruction manuals and on-lone forums detailing how to set them up to work with MAF's, AFM's MAP's etc in all the various tuning modes. Megasquirt might be a good place for you to start since it comes in kit form and is great fun for those who want to experiment. Otherwise forget Alpha-N with your Motronic and stick with speed-density (MAP) until you get your tuning right. Hard as hell to tune without an 02 sensor though Bit like playing darts in the dark.

yea but im the kinda guy that gets something and never bothers looking at the instructions haha. motronic 1.3 is actually quite a good management system and can be tuned just as much as any aftermarket system. it has its limitations but they are more about adding new stuff like knock sensors or add-ons for wasted spark or sequential injection, etc. that stuff is cool and all but not really that necessary in the grand scheme of things unless ur building a seriously crazy or huge BHP engine. who said i havent got an O2 setup? AEM wideband on the car as well as fuel pressure, oil pressure and vacuum. trust me when i say i know most of the bells and whistles in motronic 1.3, my own personal tune puts some big names to shame. full throttle maps are more agressive and give more power than Dinan, Active Autowerke, Evolve and Jim Conforti to name a few and more agressive than all of those bar the Dinan one by a country mile too. the part throttle maps are smooth as a babies bottom also, none of this copying values from the unused 2nd set of part throttle maps like alot of brands/people do including most/all of the above mentioned bar Dinan who didnt even bother to do that, they just left the PT maps bog standard. i run a small business doing this stuff on these cars just havent hit the mainstream with it yet but ive sold chips far and wide around the world since i started up and im working on developing my own brand MAF conversion as well as possibly MAP and maybe even Alpha-N with baro correction. all things come about from my own experimenting/tuning, no copy and paste artists here.

ive picked up some 4.7k resistors and 100uF capacitors, going to see if one of each will smooth out the MAP sensors output signal into something more workable so might have an update on the MAP sensor conversion front soon.

[GeoffBob]

This confirms my thinking E30 M3 has an AFM, copied this post found by searching 'M3 S14 AFM'...

So the S14 had ITB's and an AFM. Sounds like a contradicition in terms IMO.

I found this on the DTMPower forum:

What ever you do if you stay with the S14, ditch the AFM. The S14's AFM is a major weak point and it's deletion is the most dramatic fairly straight forward mod you can make for the car.

I can well believe that deleting the AFM makes a difference.

[GeoffBob]

[e301988325i]

This confirms my thinking E30 M3 has an AFM, copied this post found by searching 'M3 S14 AFM'...

So the S14 had ITB's and an AFM. Sounds like a contradicition in terms IMO.

I found this on the DTMPower forum:

What ever you do if you stay with the S14, ditch the AFM. The S14's AFM is a major weak point and it's deletion is the most dramatic fairly straight forward mod you can make for the car.

I can well believe that deleting the AFM makes a difference.

Yes, AFM, plenum chamber, then ITB's. . .

See here for how the E30 M3 uses the barometric control (page 29) and FYI my facelift 1.3 '88 325i has no input to pin 30. . .

http://www.wedophones.com/Manuals/BMW/1 ... Manual.pdf

[Gunni]

i should only really have to re-tune the 1x8 map and maybe make some small adjustments to a couple fuel maps. .

this is NOT the case

cause 3v @ 2k and 5k used to mean the same volume and thus mass per second with a AFM

now 3v @ 2k or say 65kpa is entirely different then 65kpa @ 4k.

This is why just adjusting the 1x8 map will not do, even if your just using a 1bar absolute sensor.

[DmcL]

i should only really have to re-tune the 1x8 map and maybe make some small adjustments to a couple fuel maps. .

this is NOT the case

cause 3v @ 2k and 5k used to mean the same volume and thus mass per second with a AFM

now 3v @ 2k or say 65kpa is entirely different then 65kpa @ 4k.

This is why just adjusting the 1x8 map will not do, even if your just using a 1bar absolute sensor.

well i could start with the 1x8 map and then re-trim fuel maps to suit. im not totally convinced its as far off as what ur suggesting tho but i know it will take more than just tweaking the 1x8. when i tried the MAP sensor on the car it seemed to read the part throttle maps as if it was a MAF or standard AFM from what i saw, granted i didnt drive it or anything so might be different under load but basically the part throttle maps are RPM on one axis and load on the other so as load increases so does fuelling and as revs increase fuelling also generally increases. im sure there will be some tweaking to be done with fuel maps but i dont think it will be too much.

ive picked up some 4.7k resistors and 100uF capacitors

Nooooo, 100 nanofarad, not microfarad! 100 nanofarad = 0.1 microfarad. Only 1000 times smaller than the value you bought.

whoops haha, well they might come in handy for something or i could just stick them up on ebay. must grab some 100nF caps to play with.

EDIT: just had a look and seems most of the 100nF caps im seeing are for way above the voltage i need unless i were to get some of those super small ones for soldering directly onto a circuit board and maybe just heat shrink over it or something if i could manage to solder wire onto something that small.. any ideas? the 100uF caps i got are 10v rated which should have been ok. lowest voltage rating i can find for a 100nF cap is 25v, looks like anything lower is meant for direct soldering to a PCB and would probably be too small to solder between 2 peices of wire. ill pick up some of the 100nF-25v caps and see what happens..

[GeoffBob]

just had a look and seems most of the 100nF caps im seeing are for way above the voltage i need unless i were to get some of those super small ones for soldering directly onto a circuit board and maybe just heat shrink over it or something if i could manage to solder wire onto something that small.. any ideas?

Your typical 100nF ceramic or polyprop cap is rated at 63V, but don't let this bother you, it just means that its good for use to a higher voltage that you'll be using it. It doesn't mean it works any worse at only 5V. If you went looking for a 100nF cap that was only rated to 5V it would be about the size of the head of a pin (and a bitch to solder in place ).

[DmcL]

thats what i was thinking but wasnt sure as i thought there must be some sort of upper limit and even 25v seemed quite a bit above the 5v im dealing with.. googled it and counldnt really find much info on cap voltage rating vs circuit voltage. i knew resistors were like that as are alot of other electronic bits but wasnt sure about capacitors tbh. the resistors i got are a good job, big beefy 4.7k wire wound ones that are rated up to a silly 7w. what way should i wire up the resistor and cap to the MAP sensor tho?

i got some 100nF 25v caps so that should do the trick. trying to think of what else i could possibly use some of these caps (or the 100uF ones) for.. wonder what would be involved in some little dimmers for the interior lights or something silly like that.

[GeoffBob]

i should only really have to re-tune the 1x8 map and maybe make some small adjustments to a couple fuel maps. .

this is NOT the case

cause 3v @ 2k and 5k used to mean the same volume and thus mass per second with a AFM

now 3v @ 2k or say 65kpa is entirely different then 65kpa @ 4k.

This is why just adjusting the 1x8 map will not do, even if your just using a 1bar absolute sensor.

Oh dear

DmcL, I wasn’t aware that the Motronic worked this way, but what Gunni says makes perfect sense. The Motronic was designed (originally, as we know) to take its input from an AFM. The signal at the output of the AFM essentially reports back to the Motronic the air flow into the engine. All the Motronic need do in response is to then inject the right amount of fuel corresponding to the amount of air reported by the AFM. This is completely and absolutely regardless of RPM!

It seems to me that if you want to operate your Motronic from a MAP sensor you will need a 1x8 map for each and every rpm interval (say every 500rpm from idle up to red-line). I can assure you that the situation is exactly as Gunni says and that the amount of fuel that your ECU needs to inject is not purely a function of the signal from your MAP sensor (as was the case with the AFM). To determine how much fuel to inject your ECU absolutely must factor together the MAP signal, the IAT signal, and a measure of the engine RPM.

If I understand Gunni’s description of the Motronic correctly then I think this means it will not be possible to tune your Motronic to work with a MAP sensor. I think that there is still hope for the alpha-N idea though.

DmcL/Gunni, please comment if my understanding of the Motronic is incorrect. It sounds to me (from the above) that the Motronic1.3 has a 1x8 map to interpret the original AFM, followed by a number (three by the sounds of it) of fuel maps to enrich or enlean the fuel as a function of rpm. Which of the three fuel maps is selected depends upon the position of the throttle. I assume that one of the fuel maps is specifically for WOT.

[GeoffBob]

the resistors i got are a good job, big beefy 4.7k wire wound ones that are rated up to a silly 7w. what way should i wire up the resistor and cap to the MAP sensor tho?

A 100nF, 63V ceramic or polyprop cap, and a 4k7, 1/4 watt carbon resisitor would have been fine. But what you have is more than good enough to do the required job.

Cut the line from the MAP sensor to the Motronic and insert the 4k7 resitor into the line. Now connect the capacitor between ground (0V) and the motronic (not the MAP) side of the resistor.

TBH, from the above, I'm not sure that any of this will do any good. Unless you have a full 2D map with RPM on the x-axis and MAP signal on the Y-axis, I am not sure that you will be able to tune your Motronic to work with a MAP sensor? I could be wrong but that's how I interpret Gunni's post.

[DmcL]

im pretty sure its possible just not sure how much tuning will be required tbh.

i was originally looking at smaller resistors but then i saw these big beasts and thought they would be better suited to being under the bonnet, plus bigger = easier to work with. would there be any issue with just heat shrinking over this stuff and the wiring? was thinking the cap would go between ground and the signal wire but wasnt sure about the resistor.

dont really have any expectations on this, just fancy having a bash since ive already managed a MAF conversion without too much hassle. wanting to see how far i can push the envelope so to speak. also want to try a variable TPS and MAP sensor for baro correction but i havent a baldies what all id need to make circuit wise to get it to work as desired. may even be easier than tryign to use a MAP sensor as im assuming the TPS output would be linear. both setups are probably very do-able just may be a case of re-tuning after any breathing mods or changes but id be able to manage that on my own car so why not have a bash and see if theres any benefits over a MAF.

also planning to try using a different MAF and see how it compares to the one i was using. thinking one of the E36 2.8 or bigger 4 pin MAF's as wiring it up would be the same as the 4 pin MAF ive been using but with a bigger housing and maybe a smaller MAF sensor, the one in the MAF i was using is pretty big and ugly.. plus some of the 4 pin E36 MAF's have the same fitting on the housing as my current MAF so i could stick the current MAF into the larger housing if all else fails or i dont like the other MAF sensor. ultimately aiming for custom 3 inch one peice intake pipework with an integrated fitting for the MAF sensor (unless i end up going MAP or TPS/baro correction) with a 3 inch to 2.75 inch reducer from the end of the pipe to my 64mm TB. wanting to keep things as internally smooth and unrestrictive as possible and most likely with a velocity stack on the end to mount my large foam sock filter over.

[Gunni]

i should only really have to re-tune the 1x8 map and maybe make some small adjustments to a couple fuel maps. .

this is NOT the case

cause 3v @ 2k and 5k used to mean the same volume and thus mass per second with a AFM

now 3v @ 2k or say 65kpa is entirely different then 65kpa @ 4k.

This is why just adjusting the 1x8 map will not do, even if your just using a 1bar absolute sensor.

Oh dear

DmcL, I wasn’t aware that the Motronic worked this way, but what Gunni says makes perfect sense. The Motronic was designed (originally, as we know) to take its input from an AFM. The signal at the output of the AFM essentially reports back to the Motronic the air flow into the engine. All the Motronic need do in response is to then inject the right amount of fuel corresponding to the amount of air reported by the AFM. This is completely and absolutely regardless of RPM!

It seems to me that if you want to operate your Motronic from a MAP sensor you will need a 1x8 map for each and every rpm interval (say every 500rpm from idle up to red-line). I can assure you that the situation is exactly as Gunni says and that the amount of fuel that your ECU needs to inject is not purely a function of the signal from your MAP sensor (as was the case with the AFM). To determine how much fuel to inject your ECU absolutely must factor together the MAP signal, the IAT signal, and a measure of the engine RPM.

If I understand Gunni’s description of the Motronic correctly then I think this means it will not be possible to tune your Motronic to work with a MAP sensor. I think that there is still hope for the alpha-N idea though.

DmcL/Gunni, please comment if my understanding of the Motronic is incorrect. It sounds to me (from the above) that the Motronic1.3 has a 1x8 map to interpret the original AFM, followed by a number (three by the sounds of it) of fuel maps to enrich or enlean the fuel as a function of rpm. Which of the three fuel maps is selected depends upon the position of the throttle. I assume that one of the fuel maps is specifically for WOT.

Basically it goes like this.

AFM voltage comes in
AFM voltage is translated into volume (effectively it was pressure differential that caused various flapper angles)
Air temp used to find density of said volume
Thus mass / time known.
divide by engine speed to give per cycle mass

Then you have the injector constant,.
It basically tells the ecu what the load will represent as Ti or injection time , i.e the injector variable will give out a PW for lambda one, this is motronics "LOAD" (pulsewidth exactly)

then you have a fuel table

Fuel table has LOAD(Injection time) as one scale and rpm as the other, given the perfect fuel system and correctly flowing injector at the right fuel pressure without any vacuum leaks
you can fill the fuel tables with Lambda 1 target values or
128 (out of 255) and the results would be lambda 1 at all operating ranges, so Lambda 1 = 128 , a higher value will math out as

130/128 = 1.5% more fuel then "LOAD" says , the load again being the lambda one pulsewith for said mass of air per cycle

So if the ecu has figured out from the 1x8 map and AFM scalar + Injector constant the pulsewidth for 0.17kg/s is 13ms
and your fuel table says at 13ms open time and 5000rpm I want 1.5% more mixture then lambda one
Resulting in Lambda = 0.984

Then at 2000rpm it can be lambda 1 .



This is how the ecu thinks, it first mathematically figures out how much injection time is required to give lambda 1 based on the AFM signal and injectors installed, it then looks up the fuel table where it compares the resultant mixture change based on the pulsewidth(almost literally torque/cylinder pressure producable) and then at that proposed "LOAD" it will use the fuel table to adjust the richness of the mixture.

Now to alter the ecu for MAP you need to change the AFM scalar , 1x8 map and then retune the whole fuel and ignition maps.
It will work like a speed density system when executed right,
The fuel table will no longer be a lambda target adjustment map but a straight VE map like standalones use.

no need to change code or anything like that.

[DmcL]

thats more or less what i was trying to say but not in so much detail haha

this is why i think the 0-5v AFM input is a nice thing on on motronic because of how it calculates everything, it doesnt really matter what the source of the 0-5v is it just wants to see an input which then references the part throttle maps which can be tuned as necessary.

my sticking point so far has been getting a steady signal from a MAP sensor, with the capacitor and 4.7k resistor im hoping that will smooth the output enough to give me something to work with.

See here for how the E30 M3 uses the barometric control (page 29) and FYI my facelift 1.3 '88 325i has no input to pin 30. . .

http://www.wedophones.com/Manuals/BMW/1 ... Manual.pdf

so its pin 30 that the barometric sensor would have been wired up to then?

Motronic 1.3 does have a 1x3 barometric correction map but all values are set to 128 since there is no barometric compenation circuit/sensor fitted. buuuuuuuuut if the maps there and we know what pin on the ECU the sensors output runs to then we could give our cars baro correction for fun and giggles. or if using a variable TPS as the 0-5v input to the ECU from the AFM then maybe it would just be a case if fitting the variable TPS and sticking in an E30 M3's barometric sensor and tuning the 1x3 baro correction map.

20 odd year old engine management and were still figuring out new tricks

EDIT: been googling trying to find info on the baro sensor in the M3, looks like it wasnt just on the M3. must have been one of those "this country got it, that country didnt" kinda things.

its listed here as fitting basically all models of E30, also has a parts drawing showing what it looks like:
http://data.bmrparts.com/bmwparts/8260099664953U.html

[GeoffBob]

Basically it goes like this.

AFM voltage comes in
AFM voltage is translated into volume (effectively it was pressure differential that caused various flapper angles)
Air temp used to find density of said volume
Thus mass / time known.
divide by engine speed to give per cycle mass

Then you have the injector constant,.
It basically tells the ecu what the load will represent as Ti or injection time , i.e the injector variable will give out a PW for lambda one, this is motronics "LOAD" (pulsewidth exactly)

then you have a fuel table

Fuel table has LOAD(Injection time) as one scale and rpm as the other, given the perfect fuel system and correctly flowing injector at the right fuel pressure without any vacuum leaks
you can fill the fuel tables with Lambda 1 target values or
128 (out of 255) and the results would be lambda 1 at all operating ranges, so Lambda 1 = 128 , a higher value will math out as

130/128 = 1.5% more fuel then "LOAD" says , the load again being the lambda one pulsewith for said mass of air per cycle

So if the ecu has figured out from the 1x8 map and AFM scalar + Injector constant the pulsewidth for 0.17kg/s is 13ms
and your fuel table says at 13ms open time and 5000rpm I want 1.5% more mixture then lambda one
Resulting in Lambda = 0.984

Then at 2000rpm it can be lambda 1 .

This is how the ecu thinks, it first mathematically figures out how much injection time is required to give lambda 1 based on the AFM signal and injectors installed, it then looks up the fuel table where it compares the resultant mixture change based on the pulsewidth(almost literally torque/cylinder pressure producable) and then at that proposed "LOAD" it will use the fuel table to adjust the richness of the mixture.

Now to alter the ecu for MAP you need to change the AFM scalar , 1x8 map and then retune the whole fuel and ignition maps.
It will work like a speed density system when executed right,
The fuel table will no longer be a lambda target adjustment map but a straight VE map like standalones use.

no need to change code or anything like that.

Thanks Gunni . That all makes perfect sense to me.

I do however see a very definite problem with using the Motronic1.3 with a MAP sensor.

In the case of an AFM the input to the fuel table is effectively (as you said above) the injector time. The fuel table may then scale this value up or down (as a function of RPM) from 0/256 (1/2x the specified injector time) to 255/256 (2x the specified injector time), with 128/256 (1x the injector time) in the middle.

The fuel tables are there (in the case of the AFM) to enrich or enlean the calculated injector times.

In the case of the MAP sensor the input to the fuel table is now some value that does not directly relate to injector time. This is OK (in principle) since all we need is for the fuel table to output some sensible injector time as a function of this arbitrary input value.

The problem is, the fuel table (by your own description) works to scale the input value up or down by a value of a to 2 times the input. It’s not simply a 1-to-1 look-up table. It’s a table that scales the input value. And therein lies the problem because the input value does not relate directly to injector duration (as was the case with the AFM connected).

Sorry, I’m not deliberately trying to poke this idea in the eye. I only raise what I see as a potential problem in the hope that there is a solution for it. And the more problems we knock on the head early the easier you path, DmcL, to a finished item.

[GeoffBob]

Motronic 1.3 does have a 1x3 barometric correction map but all values are set to 128 since there is no barometric compensation circuit/sensor fitted. buuuuuuuuut if the maps there and we know what pin on the ECU the sensors output runs to then we could give our cars baro correction for fun and giggles. or if using a variable TPS as the 0-5v input to the ECU from the AFM then maybe it would just be a case if fitting the variable TPS and sticking in an E30 M3's barometric sensor and tuning the 1x3 baro correction map.

20 odd year old engine management and were still figuring out new tricks

This sounds promising DmcL. Sounds like you might be able to work a Barometric compensated Alpha-N setup without requiring an external circuit after all.

[Gunni]

Remember that you have some serious variance available

anything from

1/128 = 0.0078125 * PW (AFR = 188:1)

too

255/128 = 1.9921 * PW (AFR = 7.35:1)

So if the calculated PW is 10ms, then it can go as far down as to 0.07ms and upto 20ms.
That scale is plenty adjustment given the possible duty cycles of a running engine with a rpm limit of 7k. this also semi auto adjusts as the injectors will always be adjusted according to the power range you need.

This is where injector and scale adjustment comes in, i.e 300kpa sensor with a 0.5v-4.5v range will give close to 1.8v for 100kpa or so, if you have stock injectors then you´ll have a problem as you need 15-16ms at 6k to fulfill the fuel need on a stock engine as the ecu will only want to give you 4ms and even with 255 in the fuel table you only get 8ms.

This is where the injector constant comes in and "grows" the target PW´s to suit.

Ideally you grow it to suit the map sensor range and your maximum pulsewidth required based on the fuel injectors, fundamentally you need to be able to reach 20ms or so for absolute maximum torque possible at a given voltage (specific to your map sensor and the range of it you plan on using) with 255 as the PW multiplier from the fuel table

Say you have a 250kpa sensor, most likely plenty for everybody to not run more then 1.5bar on their M20. This would then mean adjusting the injector constant and fuel table to give 20ms or so at 4k at max voltage. Then if you do need 16ms at that voltage the fuel table value would be

20/16 = 255/X
X= 204 .

If by some strange occurrence you need 3ms at peak voltage (un godly large injectors)
then

20/3 = 255/X
X= 38

The injector constant directly effects the resolution you have of the 0-255 range to effect the calculated PW.

[GeoffBob]

Remember that you have some serious variance available

anything from

1/128 = 0.0078125 * PW (AFR = 188:1)

too

255/128 = 1.9921 * PW (AFR = 7.35:1)

I see!!! Makes perfect sence Gunni. My mistake was in my understanding of how the value in the table is used to scale the "load" value at the input to the table. Thanks for the explanation.

The injector constant directly effects the resolution you have of the 0-255 range to effect the calculated PW.

Fully agreed, and I fully appreciate how important it would be to optimise the resolution since the 20 year old Motronics are probably capable of 8-bit non floating-point arithmetic at best. Setting the injector constant so that the fuel table values lie in the range of, say, 1 to 32 (5-bit) out of the available 256 (8-bit) would not be ideal.

[DmcL]

yea the constants are a handy thing to use for scaling the overall amount of fuel delivered without touching the fuel maps other than for fine tuning (or maybe tuning for a MAP sensor or TPS input).

u can also use the baro correction map as basically a second injector constant as both maps are 1x3 and both are read by the ECU and at default value both are filled with 128.

it really is amazing how the older motronic system works, IMO it gives standalone management quite a run for its money. as long as ur not after sequential injection or wasted spark/coil on plug or something along those lines motronic will suffice quite well.

[Gunni]

Let´s put it this way, as long as you want fuel to inject based on your requirements and ignition to occur from a single coil then it can do the job,

a standalone is miles ahead in terms of throttle response and resolution though

[DmcL]

resolution yes, dunno about throttle response.. with a MAF response is pretty damn quick.

[GeoffBob]

it really is amazing how the older motronic system works, IMO it gives standalone management quite a run for its money. as long as ur not after sequential injection or wasted spark/coil on plug or something along those lines motronic will suffice quite well.

Fair enough, but I doubt you'd look back if you had a standalone to play with. The reason I know so little specifically about Motronic is because my first reaction is to throw them out. Don't get me wrong, I don't mean to knock Motronic, but I do share Gunni's feelings on the matter. The extra features that come on a good quality aftermarket ECU aren't just nice to have, they really do contribute (if used correctly) to a better tuned engine in the long run. I don't know where I'd be, for example, without my knock sensor (probably swapping pistons at this very moment).

But all this is just my own personal opinion, and by no means intended to influence yourself. At least I now understand a little better how the Motronic ECU works .

[DmcL]

oh yea not saying standalone isnt better on the whole but for what can be done with motronic its not really worth wasting extra money on standalone unless its a necessity.

EDIT: its also alot more difficult to work with motronic because generally people who know alot dont want to share what they know because theres alot still not figured out even nowadays. basically its cheaper but the trade off is u have to figure alot of it out urself whereas with MS u pay the price but theres 2489612398639 other people running MS on the same engine u have and loads of info and nothing thats unknown or undiscovered like there is with motronic. ive found loads of new to me maps and they arent in any of the definition files ive seen on the net so im assuming they are new to all of us motronic geeks, havent figured out what the maps are yet but i can tell which ones are used and which ones dont appear to be used by the ECU. ive also figured out what some previously unknown maps were so as u can guess i dont really need to go standalone with that sort of understanding of early motronic. any motronic E30 i should basically be able to just plug into the ECU and map it or burn a chip once ive finished playing with it. the tuning software is free so the only thing really needed for DIY tuning is the moates ostrich 2.0 emulator and a laptop.

[Mikey_Boy]

Worryingly, back in the day, I was exposed a lot to Motronic version 1.x when I was a young engineer..! Judging by the thickness of the strategy book (from memory!) it's capable of quite a bit! Basic issue of course is simple processor speed. Now I have decided on the spec of my new engine (it's going to be an M20 despite being tempted by M52s and S50s!) I will try Motronic first and see how it does on that - standalone after to compare and contrast...

Cheers,
Mike

[DmcL]

should also check out similar era porsche motronic, very similar in some aspects but very different in others. dunno if all had it but i know some had a fuel quality switch in the ECU which could be adjusted manually, not sure about other bells and whistles on the porsche motronic but its something i wouldnt mind looking into just out of curiosity. also been noseying at E36 files as well as other E30 and M30, M42, etc, etc.

[GeoffBob]

...the only thing really needed for DIY tuning is the moates ostrich 2.0 emulator and a laptop.

Good to know.

Purely out of interest, was the Motronic developed in-house by Bosch or did they contract their ECU development out?

[Gunni]

In-house, they however where the contractors to many other ecu manufacturers.

[DmcL]

so getting to know how motronic works can carry over into some other brands of management then.. cool..

[Mikey_Boy]

+1

Motronic really was the daddy of them all - lots of Bosch type methodology in many other ECU types...

[e301988325i]

thats more or less what i was trying to say but not in so much detail haha

this is why i think the 0-5v AFM input is a nice thing on on motronic because of how it calculates everything, it doesnt really matter what the source of the 0-5v is it just wants to see an input which then references the part throttle maps which can be tuned as necessary.

my sticking point so far has been getting a steady signal from a MAP sensor, with the capacitor and 4.7k resistor im hoping that will smooth the output enough to give me something to work with.

Just use the map sensor for baro correction and MAF/ TPS for air flow

See here for how the E30 M3 uses the barometric control (page 29) and FYI my facelift 1.3 '88 325i has no input to pin 30. . .

http://www.wedophones.com/Manuals/BMW/1 ... Manual.pdf

so its pin 30 that the barometric sensor would have been wired up to then?

Motronic 1.3 does have a 1x3 barometric correction map but all values are set to 128 since there is no barometric compenation circuit/sensor fitted. buuuuuuuuut if the maps there and we know what pin on the ECU the sensors output runs to then we could give our cars baro correction for fun and giggles. or if using a variable TPS as the 0-5v input to the ECU from the AFM then maybe it would just be a case if fitting the variable TPS and sticking in an E30 M3's barometric sensor and tuning the 1x3 baro correction map.

20 odd year old engine management and were still figuring out new tricks

EDIT: been googling trying to find info on the baro sensor in the M3, looks like it wasnt just on the M3. must have been one of those "this country got it, that country didnt" kinda things.

its listed here as fitting basically all models of E30, also has a parts drawing showing what it looks like:
http://data.bmrparts.com/bmwparts/8260099664953U.html

Interesting how the MAP sensor is powered from the AFM out?

It's good to learn motronic, some of us just don't need MS, nor want to pay that price for a what can be achieved with motronic, with the benefit of keeping the car largely stock.

[Gunni]

It would be powered by the common 5v output of the ecu that was used to power the AFM.

Need and not need is only relevant to what you want to accomplish, motronic is only good for two things, fuel and spark, batch injected, sloppy if any acceleration enrichment, single coil, distributor bits to replace,

[e301988325i]

It would be powered by the common 5v output of the ecu that was used to power the AFM.

I went back and checked the wiring diagram, as you suggest, it is of course powered by the feed from the ECU, there are current direction arrows on the schematic wiring diagram too, my bad.

[DmcL]

im not sure bit i think the female connection in the cars loom for the baro correction sensor may even be there tied up somewhere under the TB/inlet manifold. there is 1 connection i noticed tied up down there previously, never noticed any others tho but then i never looked hard either.

so far i know we didnt get the charcoal canister/EVAP system for the fuel tank which would have been situated under the inlet and connected up around there, we didnt get the stock lambda sensor which i think also would have connected in around there i think and now also a baro correction sensor which im also guessing would have been connected around there somewhere too. if the loom in all the cars are the same then we should have like 3 unused connectors around that area in the engine bay for those 3 systems/sensors.

i know the lambda sensor would be nice to have, it would give the ECU more control over air/fuel mixture which would be a good thing if running larger injectors or having modifications done to the engine but without having the ECU specifically tuned for the mods.. without it the ECU keeps working like normal and doesnt correct the mixture at all.

the charcoal can/EVAP system really intrigues me tho.. the control map for it is a decent size and has a fair amount of resolution, seems like the perfect map to tune and use to control an electric fan or water/meth injection system or something else along those lines. definately something i will be experimenting with in future.

regarding the MAP sensor conversion, i havent done anything with it since my last update a page or 2 back in the thread. between work and weather i havent really had time but ive got the whole thing sitting here beside me so ill wire up the 4.7k ohm resistor and 100nF capacitor into the loom i made for it and give it another shot when the weather gets a little warmer.

[e301988325i]

The connection under the manifold is for the charcoal cannister.

Under the battery tray in the engine bay is where you'll find the lambda socket, and yes I run with a lambda and yes it is usefull for running my 19lb injectors when on petrol.

Well we know the baro sensor was on the bulkhead and I don't know of any unused connectors there.