Jump to content


Welcome to Ultimate Subaru Message Board, my lurker friend!

Welcome to Ultimate Subaru Message Board, an unparalleled Subaru community full of the greatest Subaru gurus and modders on the planet! We offer technical information and discussion about all things Subaru, the best and most popular all wheel drive vehicles ever created.

We offer all this information for free to everyone, even lurkers like you! All we ask in return is that you sign up and give back some of what you get out - without our awesome registered users none of this would be possible! Plus, you get way more great stuff as a member! Lurk to lose, participate to WIN*!
  • Say hello and join the conversation
  • Subscribe to topics and forums to get automatic updates
  • Get your own profile and make new friends
  • Classifieds with all sorts of Subaru goodies
  • Photo hosting in our gallery
  • Meet other cool people with cool cars
Seriously, what are you waiting for? Make your life more fulfilling and join today! You and your Subaru won't regret it, we guarantee** it.

* The joy of participation and being generally awesome constitutes winning
** Not an actual guarantee, but seriously, you probably won't regret it!

Serving the Subaru Community since May 18th, 1998!

Guest Message by DevFuse
 

- - - - -

Updated oddcomp's MS n' EDIS fuel/DIS


  • Please log in to reply
108 replies to this topic

#1 Guest_shadow_*

Guest_shadow_*
  • Guests

Posted 19 July 2004 - 08:36 PM

MS, short for MegaSquirt is a build-it-yourself digital fuel injection controller. I have been running MS on my daily driven RX for almost 2 years as the fuel injection-only controller.

MS n' EDIS is an offshoot of this project. With minor modifcations to the assembly code and hardware of the MS controller, it is possible to use the ECU as both a fuel AND spark controller (MS is a fuel-only controller by design).

The Ford system is attractive for DIY'ers for several reasons. One of the main reasons is the the EDIS module controls both spark and dwell all on its own and can run in a 'limp home' mode without any ECU attachments whatsoever. In this mode, the timing is locked at 10 BTDC for saftey. Another key to DIY'ers is getting rid of the distributor/single coil arrangement. Multiple coils (coilpacks) have less chance of misfire under the same conditions; there are less connections to go bad or corrode on the high voltage side; and having multiple coils means that the car may still be driveable even if one coilpack fails, affording better overall reliability and saftey margin.

Beyond having an MS controller at your disposal to use as fuel control and to interace with the Ford spark system, you need to collect the following parts from a 4 cylinder Ford (EDIS-4) car. The donor car you are looking for to collect the necessary Ford parts is the 90's Ford Escort/Mecury Tracer w/ 1.9L, 4 cylinder SOHC engine. All of the Ford parts average a total cost of ~$30-$50 at salvage yards nationwide.

EDIS Control Unit - This is the Ford Brain box. On the donor car, these can be found on the front of the driver's side strut tower, behind the stock fuse box. See picture below. Make sure when taking this you also hack the connector with at least 6-12" of good wire for splicing.
EDIS Coil Pack - This is the coil unit that converts the low voltage to high voltage for use in the sparking system. The coil has a small connector for control and 4 prongs for spark plug wires. It is adviseable to buy a new coilpack for reliability sake. Make sure even if you don't take the coil pack, get the connector and as much of the harness as you can. Pictured below.
EDIS Trigger Wheel - This is a 36-1 toothed wheel that is found on the harmonic damper/crank pulley on the Ford car. It tells the EDIS control module the crank angle of the engine, much like the optical sensor works inside the EA82T distributor. You will need to grab the entire pulley, because the wheel is press-fit on and its best to take it home to work on removing it. These are much easier to get off with the engine pulled from the car, and often times with the sheer numbers of these in the yards, the engines can be found laying around because the tranny was needed. See picture below.
EDIS Trigger Wheel VR Sensor- This is a magnetic sensor that reads the teeth on the above wheel, and feeds that data to the EDIS control module. Make sure when you snag the sensor you get the connector as well. See pictures below.The EA82 crank pulley is a perfect candidate for surgery - it is a double V-belt pulley where on 90% or more of the roadgoing EA82's, only the front belt groove is used. On the pulley pictured below, the rearmost V-groove has been turned on a lathe to about 4" O.D.
The Subaru pulley above was not turned to an 'interference' or press fit dimension. Instead, it was left to be a snug fit, but loose enough to allow the trigger wheel to be indexed once the pulley was re-installed on the engine and the VR sensor location could be presisely determined. Below, you see the mockup of the trigger wheel mounted on the pulley:
With the trigger wheel in place (but not permanently attached), the crank pulley was re-installed to check for clearances. Here a problem was found - both the engine oil dipstick and the waterpump pulley's rear V-groove were dangerously close. After the test fit, it was found that the waterpump pulley was not going to fit in the same space with the trigger wheel. The solution is to remove the rear, unused V-belt groove. To do this on the test install, a bandsaw and grinder were used.
Once that was fixed, it was found that the dipstick will just clear, but it is a tight fit. The dipstick tube could be tweaked a bit to get a couple fractions of an inch to spare. Next, the location for the VR mount had to be found. With a lack of space around the crank pulley, with the waterpump to the right, dipstick to the left, plastic timing cover cases unsuitable for mounting all around, and a power steering and alternator and A/C brackets to the top, it is a tough game to find a location. After some testing, the location between the dipstick and power steering pump was chosen, using the dipstick bolt as a solid mounting point. It is imperative that there be no vibration in the mount - afterall your ignition timing is depending on it!
The mount is still under construction even as I type, it is being made from sturdy 1/4" thick plate steel. Using the mocked up position of the VR sensor, you can now index the missing tooth on the wheel and properly align the trigger wheel on the crank pulley and attach it. You could use evenly spaced bolts (the rotating assembly must be balanced) or weld the wheel in place. For speed and strength, the test system's wheel was welded.
After the VR mount is completed, the remaining steps are a matter of wiring and simple placement and hookup of the EDIS control module and coilpack. Once installed, and even before attached to the MS for timing control, the engine should be able to be started to test the system, with the EDIS operating in 'limp home' mode. Once attached to the MS, timing is load and RPM mapped in the software with an 8x8 table, with max. system advance of 57 degrees BTDC possible. The curve is fully tunable in the computer and optimal timing can then be achived. Using this combination of fuel and spark control will give you an affordable engine management solution that will rival and can even beat expensive aftermarket computer solutions - at a fraction of the price.
With a few more weeks of evening tinkering, the system should see its first ever Subaru engine alive and ticking with EDIS and MS.

#2 WJM

WJM

    SUBARU

  • Members
  • 7,828 posts

Posted 19 July 2004 - 09:20 PM

AWESOME. :banana:

#3 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 19 July 2004 - 09:45 PM

heh if anyone up i pdx has a lathe and better welding skills than me <still learning> and wants to help me redo the welds on the crank pulley and then clean it up on a lathe so its a bit better balanced
than just using a grinder on it to cleanup the welds let me know :)
might be nice to also clean up the backside of the waterpump pulley as well ...its really really ugly

so far the trigger wheel has with stood several sprints to 4500 rpm .. residential area after all...
i will see how well it does on a half hour steady state rpm i hope the thing is not to far out of balance and breaks loose
i hope to finish makeing some of the parts i need to mount the vr tonight at work where i have accsess to a drill press




oh and i am sure you all can guess why i asked shadow to help me write this up
since my typing skills are .. hehe nonexsistant

thanks shadow
and
thanks garner for the nifty pulley work !

#4 subaru420

subaru420

    USMB Regular

  • Members
  • PipPipPip
  • 81 posts
  • Seattle

Posted 19 July 2004 - 10:16 PM

That Rocks SO HARD! Keep us updated on it's performance :headbang:


That Cog looks HEAVY! Curious what kinda weight it adds to your allready meaty pully

#5 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 19 July 2004 - 10:20 PM

its actually really light
i would thing the amount of steel thats lost by machining the puley down for it to fit over just about comes out even.. i did forget to weigh the dam thing.. oops

#6 GeneralDisorder

GeneralDisorder

    Elite Master of the Subaru

  • Members
  • 20,280 posts
  • Portland

Posted 19 July 2004 - 11:18 PM

Hhhmmm - looks like this would be lots easier on an EA81 - no problems with space there..... Many other more difficult problems with fuel delivery tho.

GD

#7 baccaruda

baccaruda

    YOUR FAVORITE MOD

  • Moderator
  • 6,941 posts
  • SpoVegas, WA

Posted 19 July 2004 - 11:30 PM

that is extremely slick. i'm curious about having the trigger wheel mounted/welded to the front of the crank pulley to solve the clearance and water pump problem?

Also, if anyone's interested (heh heh) there's a JY in spokane called Bill's Fords Only Auto Dismantling. Who knows what he might have sitting on a shelf so you don't have to go dirty your tools in the domestic side of the JY :P
I'm sitting in the middle of a disorganized mess with no phone book in sight or I'd list the #, but it's somewhere on the interweb, i'm sure. good luck and again, this is extremely cool stuff you guys are doing!

#8 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 20 July 2004 - 09:48 AM

well mounting the ring to the front of the pulley would require a custom pulley made with extra space for the timing ring to mount on and that could be spendy
not to mention every time you change a alternator belt you gotta take off the v.r sensor to get the belt off and on
heh i thought about it

if i decide to redo this pulley later i will ask to have one made as a interferance fit just like the original ford application i will be makeing a jig that positions the timing ring correctly in relation to how i am making my vr mount

so far after 60 miles seems like things are good nothing came loose and i didn't notice any "extra" vibrations.... :rolleyes: so nothing added to the usuall one's is a good thing

made a few more parts at work last night so i can finish the sensor mount
when i get the chance prolly in a day or 2

you people who liek to drown your subies in creek beds and stuff.. hehe
may want to think about this since its not as prone to water problems as a regular distributor system
and for redundancy if the controller dies the edis module reverts back to a limp home mode where it will run at a fixed timing of 10 deg btdc
also at lower rpms say idle to 2500 or so i think .. it is after all programamble with the controller it functions as a multispark unit just like a msd box
after a preset rpm it is no longer multispark and at higher rpms ms is pretty much usless anyways

so far the only real downside
is that as far as i know no one has come up with a working knock sensor for this do it yerself project
on a normal subie engine thats not a huge problem
but on turbo's it can be but my saab apc turbo controller should take care of that or it should :)

with any luck by friday i should be idleing off this setup and hopefully into the ignition tuning phase of it
i will leave the stock subaru ecu to control fuel for now.. less head ache to tune 2 systems at the same time
then after i get the ignition semi tuned i will start running the megasquirt to also take care of fuel right now its just hooked up as a really fancy datalogger so i can see how badly the stock ecu manages fuel... and well lets just say.. yike's !

i think there is more i was gonna ramble on about but i forgot ..
anyways any question or whatnot you can get me on msn underthis email i use on the board look in my profile
or yahoo and aim its my usmb name

oh yeah
no i remember
another benefit is with out a disty the drivers side timing belt should last a bit longer i plan to leave my disty in place but not wired up for about a month or 2 i figure if after a month or 2 the system has not failed then i don't need the disty as a back up

but people who do alot of off road and stuff. might wanna leave theirs in just in case.. something fails edis module ,coil pack ect prolly won't happen but.. a failure in town is not as bad as one out in the boonies 30 miles from a parts store
:)

#9 garner

garner

    Brat Master Flash

  • Members
  • 689 posts
  • Vilas, NC

Posted 20 July 2004 - 01:35 PM

Cool!! I'm glad to see this coming together. I'd never heard of megajolt or megasquirt until oddcomp pm'd me a few months ago about making a pulley. Honestly, I thought he was nutz, i kind of still do;). Long story short, I've got al my megasquirt and megajolt stuff on it's way. I'm sold. When I was dealing with merging the wiring harnesses for my swap, oddcomp says "go megasquirt and megajolt, you'll be up and running sooner" He was right.

As for the pulley, I am turning an aluminum version like my ultralight crankpulley. On my set up, I'm using juts the inner belt track/grooves. I know clearance will be better and can live with pulling the sensor when changing belts... Since the pulley will be aluminum, I'll use a press fit and perhaps some roll pins to keep the trigger wheel from rotating. Should be dandy.

Anyway, keep us posted. Glad that someone will be the test pilot for this. Hope you'll share your maps and such when you get them close.

garner

#10 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 20 July 2004 - 01:41 PM

well mr garner
i am insane
thank you very very much :)
and yes i plan on posting my base timing maps and fuel maps
if i ever get bold enough to hook up the rest of the ms to run my injectors
heh i don't have the luxury of lots of money or many subies to choose from so i have to be gentle when i mod my drive to work and home again suby :)

#11 garner

garner

    Brat Master Flash

  • Members
  • 689 posts
  • Vilas, NC

Posted 20 July 2004 - 02:45 PM

i will start running the megasquirt to also take care of fuel right now its just hooked up as a really fancy datalogger so i can see how badly the stock ecu manages fuel... and well lets just say.. yike's !

:)



Tell me more about using MS as a fancy data logger, did you tie into all the sensors but not have the MS running the injectors. Data logging would help me tune my stock set up until I go fully sparked n' squirted.

garner

#12 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 20 July 2004 - 10:23 PM

well i have o2 and air temp and vacum/boost hooked up
and with my temporary tablet puter i am using until my screen shows up for the carputer plugged into the ms i can record the o2 voltage output dam it bounces around alot seems the stock ecu hates modded suby's...
intake air temp.. wow scary at 10-11 psi its hits about 210F yipe....
and i did have rpm for a while but then i started getting processor resets from ig nition noise so i have to redo the grounding for the rpm circuit
i don't have tps hooked up.. but you are suposed to be able to share that and some other stuff with a few changes.. i should look into it :)
water temp is not hooked up either same reason no tps don't wanna screw around with the stock ecu more than i have already
i do know that the ms verify that my junkyard dodge shadow boost gauge is dam near dead on :)
oh hey lookit the time i gotta head to work gotta make moeny so i can try and create more stuff err adapt :)

#13 rallyruss

rallyruss

    Subaru Master

  • Members
  • 1,653 posts
  • San Jose CA

Posted 21 July 2004 - 03:15 AM

odd comp- O2 is supposed to bounce. the more the better. the reason why you may ask? it makes the cat work the most efficently. a stock fuel system needs to bounce between rich and lean to make the cat work properly.

as for those intake temps? I cannot stress the importance of intercooling enough. my rather small VW unit drops aprox 100deg under full boost on a warm day.

edis is a decent system. at least as far as ford stuff goes.

keep it up. I have done a little custom stuff before but not on a sube. I used the SDS system on my toyota. never could get it to drive well. hauled arse under full throttle but could not tune it to run well at idle. I also was running turbo injectors. probably a big part of the problem. I saved the system and will probably put it on my air cooled VW when I get it up and runing.

#14 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 21 July 2004 - 09:26 AM

well i am still waiting to get a intercooler... ahem :)

with any luck i can zap together the vr mount when i get back from town depends if i fall asleep when i get back home or not really tired

yeah i can understand the bounce... but at steady cruise from almost .0 up to almost 5 volt its like watching a tennis match
give it a bit of gas and it jumps right up to where it.. should be

if your curious what i mean about it maybe being a bit excessive
i can send you a datalog of it versus rpm and boost and intake temp and the program to replay it

#15 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 21 July 2004 - 02:40 PM

well i now have a semi final vr mount... but

instead of being at the 10 degree btc that is supposed to be.. i am like at 9 grrr
got a tiny bit more tweaking to do with any luck by the end of this week i can at least run it for a bit in the "limp" mode just to make sure i have the base timing right

yawn
nap time

#16 rallyruss

rallyruss

    Subaru Master

  • Members
  • 1,653 posts
  • San Jose CA

Posted 21 July 2004 - 06:44 PM

yeah i can understand the bounce... but at steady cruise from almost .0 up to almost 5 volt its like watching a tennis match
give it a bit of gas and it jumps right up to where it.. should be

if your curious what i mean about it maybe being a bit excessive
i can send you a datalog of it versus rpm and boost and intake temp and the program to replay it


I would be interested to see that. o2 should not run above 1v. the bounce will ususally go between .2 and .8 volts. under throttle enrichment .8 to .9 steady is normal. there is a referance voltage from the ecm to the o2 for diagnostic purposes. I think thats 5volts? need to double check. you may have a crappy ground or somthing like that. you should not see the referance voltage after the o2 is hot and outputing its own voltage.

#17 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 21 July 2004 - 09:38 PM

it is actually a heated bosch sensor

but it did the same wthing with the old one just way worse
i will get another log here in a day or so

#18 WJM

WJM

    SUBARU

  • Members
  • 7,828 posts

Posted 21 July 2004 - 09:53 PM

3 or 4 wire sensor?

#19 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 22 July 2004 - 09:25 AM

3 wire eventually it will be replaced with a nice wideband unit but is alot more stable and does not drop out like the stock unit did and some of my burping and surging is gone now

but i am still running off the stock ecu for now till i am sure i got everything done up right

#20 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 25 July 2004 - 03:19 PM

well i have a small delay i gotta do a clutch on a rx to make some moer money to pay for stuff all i really need to do it bolt the sensor on and wire it up and see if it idles

then modify the megasquirt and reprogram it to run the edis

so darn close.. but i need money for wire/ metal/ wcss /junk :)

oh and i found out you do not have to be dead on the 10 deg btdc when setting the timing ring and vr sensor as you can account for it in the programming
the only bad thing is it lessens the total amount of advance or retard you can program into the system depending on how many degree's your off
it will also affect the "something broke" limp mode so as long as your not way far off you can still get home

like for instance i am off by 1 degree maybe 1.5 so i can only program in up to i think about 47.5 degree's of ignition advance .. but since have a turbo with 10 psi... i would be a bit stupid to be that far advanced anyways i think :)

when i pull teh crank pulley off for a inspection in a few weeks look for cracked welds and junk i am gonna put it on my carbed ea82 block and refine the sensor mount so its at 10 deg exactly .. i don't have to but.. with all this work i may as well do the best i can :)

#21 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 01 August 2004 - 11:23 AM

woohoo
right now i am posting on a small computer sitting on my firewall
i got things half arsed wired up and i think i got the firing order right
it will run for about 10 15 seconds then die
but i think thats cause i am still using the disty to trigger the ecu to inject fuel so the injection timing is off
i need to go and make the tach circuit adapter that will combine the 2 tach pulses from the 2 edis coils and try again

but for those curious.. it looks like this will work !!!!!!
i am gonna go run upstairs and grab my camera and take pic of the wiring mess i have
then disconnect things for today and go make the tach adapter and try again maybe tomorrow

once again yay!

#22 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 01 August 2004 - 01:31 PM

well here is the "to see if it will work" setup i had
working on the tach signal combiner right now
since each coil set fires alternatly to trigger off one side would only result in fuel being injected at half the required rate
so if you combine both sides it is injected at the correct rate in relation to the rpm
..
or so thats how its sposed to work .,...
but here is the mess!
Posted Image

it ran for about 10 seconds at a time after that i think it starved out cause of not enough fuel
since the injection was still trigered off the dizzy
but for the short time it did ran.. wow was it smooth

#23 WJM

WJM

    SUBARU

  • Members
  • 7,828 posts

Posted 01 August 2004 - 02:31 PM

Awesome.

#24 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 02 August 2004 - 09:45 AM

ok well now that i feel like the smartest idiot alive.. and after re reading some things...i know what i did wrong...
and i will in fact blame it on my dyslexia and my over eagerness to get this going
see... turns out i misread/misunderstood things .. i have it set for the base setting of 10 deg btdc....
when in fact it is suppsoed to be 90 deg from top dead center.... oops..
this is the page on the "official" instructions... perhaps you can see how i got confused...
eh maybe not but now i know why it would not run for more than a few seconds
now to go and fix it..
dam i feel stupid...
http://www.jsm-net.d...is/install.html

oh well thats why its called do it yourself :)

#25 oddcomp

oddcomp

    Subaru Master

  • Members
  • 1,763 posts
  • longview ,wa

Posted 06 August 2004 - 02:28 PM

well i may be dumb but i am not so dumb :)
if anyone is actually attempting to follow my progress with even the slightest bit of interest here is a update,...

after pulling the pulley off cutting all the welds out and re setting the ring according to the instructions on the previous post....
set motor to tdc
and then set ring ten teeth ahead of sensor... ...
now before i reveal my latest... ooops... or is it a DOH!
let me state that i did and does run fairly well even until i shut off the ignition :)
so ..
the fact that it ran and idled .. means i am sooooooo close
now here is the problem
if you go up and re-read where i said 10 teeth before the sensor.. at tdc
that would... give me the close enough to correct timing ...
now look at this pic i took
Posted Image

see the missing tooth is .. ten teeth before the sensor...
anyone guess the actuall problem yet?
..
,,
,,,
,,
when i set it.. i forgot the direction the subie motors spin
so instead of the before the sensor .. that i thought i had..i actually did 10 after the sensor... making it .. more or less 180 or so off.. wel i am prolly wrong on the amount its off
but...
it did run. ran pretty dam good in fact i had to move my advance dial on my timing light to 40 deg before i could see the 10 deg mark on the flexplate

so now since i have a night off
in the morning or maybe later today
i will go get some new grinding stones.. re-re-grind all the purty welds i did
re-re-set the ring so its correct "with tack welds only this time"
install it hook up the coil and ignition module and see if its close enough for government work
then yank it off again
and add more substantial welds to it and go from there


but it did/ does run as it is now so that means when its correct this is not a wasted effort :)

i dunno how many are actually interested in this
but at least iw ill be able to point and say ... see i did that :) pretty important for someone who does alot of dreaming but not alot of "action"

anyways ... it does prove i shoudl not do complex thinking when i am tired after work
:)




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users