COP 4 cylinder - 924s/944 & other content - Very Long Po

[Dean924s]

EDIT: 2008-12-07

For those that are looking at going down this road (and have never done it before) I strongly suggest installing a "standard" EDIS based system first. There are many possible things to try / change but first and foremost you want your car to be up and running. When I got started with this project I had huge dreams of re-inventing the wheel (and I still do) but I found that the basic things required to make it work at all take much more time than you think. The biggest is the fabrication required to mount the VR sensor and the trigger wheel.

Installation of a trigger wheel and a VR sensor sounds simple. It is! But it is also time consuming to get it correct. This is probably the most important part of the install. This is not rocket science but it does require some precision. The more time you spend making sure that things are aligned / centered and properly secured the better your installation will be. Taking time here will do two very important things.

1. When you go for first start up it works the first time. You want to eliminate as many variables from the equation as possible. That guy Murphy (the creator of "Murphy's Law") will trip you up if you give him a chance. When I did my first start up I had most of it wires with jumpers, gator clips, duct tape and crazy glue. BUT the trigger wheel and the sensor were rock solid. No mine did not start on the first try. I swapped the leads on the VR sensor and then it started (the second try)

2. If you are like me you will be driving your car a lot. On the road or on the track is a very harsh environment redundancy and overdoing things is not a bad thing. Making sure things are secured properly is key to long term success in any project. The simple rule is that if there is a better way that you are thinking of you probably should do it that way! If it decides to brake it will do it on a dark road late at night with that significant other you are trying to impress. Trust me this will happen! Plus it will be raining. Don't let this happen to you.

Lastly. As you read through all my posts you will see that there were several things i tried that never made it to the final installation. I don't like to call it trial and error or making mistakes. I call it trial and gaining wisdom. They were not mistakes! I did them on purpose and what I set out to do I did very successfully. They were just not good enough to be used in the final installation. Unless you have some one to copy from you will be forging a new path. Expect to go down some dead ends and have to backtrack. For me the process of designing the system, researching, fabricating and installing was fun. In fact I am sort of sad it is over for now. But on the same token the thrill of it firing up for the first time is fantastic. In fact you don't even need to have a Megajolt unit to get to this point. The EDIS system will run in limp mode by its self. So you could do the install get the car "running" and then order your MJ unit. My unit was at my door in less than 5 business days!

I guess what I am trying to say is keep it simple. Don't over think it (as I did). I am sure that I will eventually go down the road and have a COP system. But getting it running and now having my car run better than the stock set up is tremendously pleasing.

I put all this out here for anyone to use in any means that they see fit. I am not responsible if you get hurt, blow up your car, or anything else that may result from anything you take from what I post here.

So read on. I hope you learn from seeing what I did. I know I learned a lot and I hope it helps others realise that it is not some mysterious secret that only the select few can accomplish. Megajolt and EDIS is not some black art. It is a very simple electronic ignition that is very flexible and still reliable as a hammer.


Most importantly; I hope I make you smile while reading this. After all the whole point of this is to have fun!!!

=============================

I understand that many will want just the list with instructions/comments that are relevant to getting there car converted and up and running. I recommend reading the whole thread as there were many valuable contributions from others that helped make this possible. It will at least in part explain why I did what I did. Skip to page 7 if you want the list of things to build your own.

=============================

First off what got me started.

For those that know 924s/944's we are cursed with the Bosch motronic system that uses two VERY expensive speed and reference sensors that tend to die. There replacement cost is in the neighborhood of $150 each and two are required. So with that budget in mind and knowing how all of the other fords that I have had over the last 25 years or so have never had ignition problems I set out looking to swap a ford based ignition in my 924s. This is how I stumbled on Megajolt as I was already familiar with the MS systems.

Train of thought that I had (kind of).

1. Installing a 36-1wheel on the crank of a 944 is a bit of a nuisance due to how the timing belt is installed as well as how the power steering belt pulley is installed. Add to this how both the harmonic balancer and the two pulleys are sandwiched together and there relationship to and how they are secured to the crank and how this is also being intricate to the proper function of the oil pump make this a not so fun place to try to install the 36-1 wheel. Long story short I really don't want to mess with this unless I am doing my timing belt.

2. Along with the elimination of the speed and reference sensors the installation of an EDIS based system would also eliminate the stock cap and rotor that are mounted to the end of the camshaft at the front of the motor. Due to the problems with mounting the wheel noted in #1 above this in turn becomes an ideal physical location to mount the wheel and the pickup. However since the cam is only rotating at half the crank speed this means that stock 36-1 wheel will not play well with the EDIS unit due to it turning at half the crank speed. Or so I thought. It should be noted that I thought a possible solution to this was to make up a 72-2 wheel. This is still a possible way to go but for now this has been shelved and will be pulled out and dusted off and implemented with a standard edis-4 unit and a standard 4 cylinder coil pack.

3. A the same time I was mulling over #2 above I it occurred to me that if I wanted a true COP set up on my 4 cylinder I wold need the EDIS-8 unit as it has 4 coil drivers.

Then it hit me! If I install the 36-1 wheel on the cam and use an EDIS-8 unit I will end up with each of the coil drivers firing only once per 720 deg of crank rotation. This would make a true COP possible on a 4 cylinder motor. So next came the plan.

The plan:
Mount 36-1 gear to the cam. Because the cam is spinning at half the crank speed it would cause the EDIS-8 to only initiate a spark once for every coil trigger / connection for every 720 deg of crank rotation. This in effect eliminates the wast spark/allows for potentially true COS. This also allows for what looks to be a very easy installation of the wheel and sensor as it is at the top of the motor at the front of the engine bay and in the location where there is currently cap and rotor.

I have the 36-1 wheel and the pick up from a motor swap i did in the wife's mustang sitting on the shelf.

I would use COP coil units from a ford 4.6 V8. (I have 7-8 spares for my 2000 expedition I have accumulated over the years. Picked most of them used from ebay really cheep.

Spark control would be by megajolt v4 with the MAP sensor. I have a M62 based forced induction project all but completed for the same car.

I have an EDIS-8 unit form the junk yard.

I need to get wiring harness pig tails for the EDIS, the pickup and the coil packs

Things I don't know / Things that get weird: (the grey area)

RPM would be half if based on the wheel. An issue?

How many teeth off from the missing tooth would I mount the sensor? 5? in keeping with it is an edis 8? or would his mess up the default. After thinking about this I believe it to be 5.

Timing: Since the cam rotates half as much as the crank I would assume I would need only half the timing when tuning. IE: 10 deg btdc on the crank is only 5 deg on the cam. This would mean I could (in theory) put in or pull twice the crank timing as the EDIS timing normally would allow. Because of this how do I mount my sensor? still ten degrees off of TDC on the 36-1 or should I cut that in half as well. Thinking more about the I am convinced that I would want to mount the sensor at 5 deg before TDC on the cam giving me an effective 10 deg on the crank. This would allow the limp home function of the EDIS to would work.

Can I revamp the maps in megajolt to take into account the cam rpm is half the crank? (ie 3000 rpm on Th cam is 6000 rpm at the crank) Would my maps only go up to 3250 (the cam rpm redline of this motor)?

Same for the timing and the megajolt maps. Assuming that I will have to position the sensor at 5 deg BTDC on the cam to get an effective 10 deg BTDC on the crank. I assume that my timing inputs would be half that of a "normal" installation/setup?

More things that come to mind:

If the set up I am thinking of could be made to work it would allow for a really cheep COP set up for a motorcycle. Because you have cut the RPMS in half you could now use the EDIS system up to 18000 rpm (EDIS used normally can go to around 9000.)

At the RPMs that motorcycles turn COP is very necessary. This is mainly due to the time needed for the coils to recover. There has also been improvements in coil technology but that is another thread. One of the main reasons the wast spark and now COP was invented was for emissions. As RPMS go up the single coil systems have to fire more times a second reducing the time they have to recover between sparks. This is amplified by the fact that they are having to fire for each cylinder at every revolution. The coils will always fire but as the coil recovery time decreases so does the power of the spark. Eventually it does not fire we use to call this ignition breakdown in the old days. Since emissions were not a consideration and when they started to be tested they were only a sniff test at idle where the ignition system was most effective there was really no incentive for the car manufacturers to improve on it. But with the advent of two speed emission testing and then load testing or rolling road tests it became apparent that the single coil system needed improvement. The EDIS was a way to both simplify the mechanical side and make it even more reliable and also allow for much better and more reliable high RPM performance making huge improvements in emissions. Especially on V6 and V8 motors. This was also dictated by the move away from the low revving V8 to the higher revving 4 cylinder cars. COP was the just the next evolution of this and became especially important on the V8 and larger motors mainly due to there working range being moved to the upper rpm range. In short the more cylinders a motor has the more they benefited going to an EDIS system.

Another benefit of the EDIS and COP systems is that if a coil dies you can limp home. with the EDIS it can get a bit harry especially if you have complete coil failure as you loose 2 cylinders but on a 6 or 8 you can still drive home. On a 4 with COP if a coil fails getting home is much more plausible.

Then there is the Multi spark part of the edis. If I understand this correctly it is only usable up to about 1800rpm and then it is cut. With the cam turning at half the crank speed it would be possible to get the Multi spark effect up to 3500-3600 crank RPM. What effect will this have?

Anyway I really have really babbled on here.

After re reading this whole post I think I may need some herpy. I think I have really over thought this one. But dang it it is fun!!!! And I am going to try it "Just because"

Any thoughts? Will it work? What am i forgetting? Has anyone else done this? Am I nuts for even considering this?

At the moment it looks like all I need to purchase is the ignition controller (megajolt) and get back to the junk yard and get the the wiring harness pigtails.

[brentp]

Hello and welcome.

Yes, that was a long post! Here's hopefully a short answer.

See the following post on mounting a 72-2 tooth to the distributor shaft:
http://www.autosportlabs.org/viewtopic. ... x7&start=0

This technique lets you use a convenient and standard EDIS-4 setup.

Alternatively, you may still be able to sense from the crank if you consider other creative ways of getting the 36-1 pattern onto your crank pulley. You would also retain the accuracy of directly measuring the crank position, avoiding the errors induced by distributor gear slop.

See:
http://www.autosportlabs.net/Ford_EDIS_ ... el_Designs

[Dean924s]

Brent,

Thanks for the reply. I had read all of what you noted in your post. I have been lurking here for a very long time. I guess the really short question is would this work with the following set up.

EDIS8 with four coils one per plug on a 4 cylinder motor getting signal from a 36-1 on the cam by doing the following?

Manipulate the maps in the Megajolt so the BINS for rpm are half normal. (compensate for the half speed of the cam)
Keeping the MAP pressure BIN settings stock/normal
Use half of the "normal" timing values in the grid .... . This is the one I am not really sure of. I think halving the timing grid values it is the answer.

I think the system would run. The worst thing is I can try it and put light on i and see what the actual timing is and adjust it.

The ease of witch I can mount to the gear to the timing belt pulley where the cap and rotor are now and having one coil per cylinder that is not wast spark is enticing. Definitely not better than a crank fired wast spark system but it would be a leap forward form the stock set up and would get rid of the speed and reference sensors that are the bane of 944 most owners existence

Lastly if it does not work it is completely reversible in less than 10 minutes. I know that the answer is to just make the darn thing work at the crank but getting a true coil on plug working would be cool. I have most of the parts.. Actually I have all the parts less the Megajolt and some pigtails.

EDIT: I just ordered the Megajolt so now I just need the pigtails.

I think the real reason is because I don think anyone has done this and I really want to see if it works.

[brentp]

Hi Dean,

Agreed it would be a fairly straightforward process to find out if that arrangement would work. And you can certainly 'fake out' the settings by adjusting the RPM bins in the MJLJ to account for the setup you're running.

Since you've been doing your research you might've seen this installation with a "coil on plug" installation using dual EDIS-4's.
http://www.autosportlabs.org/viewtopic.php?t=1359

While this is still firing in wasted spark mode, it's reasonable to say there's a incremental improvement in performance compared to the conventional one coil firing a pair of plugs, in addition to the elimination of plug wires and a cleaner install.

Depending on your goals- going for absolute performance or just cool factor (or, of course, a combination of the two...) realize that with a wasted spark setup you're *still* getting approximately 4ms of potential dwell time at 7000RPMs, which brings us back to why the wasted spark setup offers such an improvement over a standard distributor setup.

To reinforce this, see the following oscilloscope plot:
http://www.autosportlabs.net/Ford_EDIS_ ... per_driver

When charging a single coil the EDIS module only dwells for about 5ms- this is at a nominal 12V on the bench, not 14.4v with a running engine. Therefore, even at your engine's max RPM you'll still get to enjoy most of that dwell time.

I hope this perspective helps! Please keep us posted on your findings- I'm sure we would all like to see what you find out.

Regards,

[Dean924s]

An update with some photos of the wheel mounted to the cam gear.

[Dean924s]

Some of the motor showing the cam mounted cap and rotor.

[Dean924s]

And a couple more of other things - Sensor mounting location and tdc on the cam. The nice thing about cam mounting is you can not be 180 out.

Also a mockup of the wheel in place. getting the alignment corect for mounting the wheel to a spair cam gear I have shown in the previous posts.

[alexander]

hi dean,
i completely understand that you may just wish to try out your ideas, and i know full well that sometimes that is half the pleasure, but on the practical side...

*i dont think it is true to say that a "true COP" setup requires your proposed EDIS8 on a 4 cylinder hybrid.

if you have the coil on the plug, that is true COP. the fact that two coils are being fired at the same time in no way detracts from whatever benefits accrue from having the coil on the plug. so while your idea may work, it is needlessly complicated to achieve no advantage.

*setting the trigger wheel is a problem. EDIS8, like 4 & 6, thinks that each tooth is 10deg of engine rotation. so, if you set the sensor 5 teeth BTDC, when the gap comes around, it thinks the motor is 50deg BTDC on no.1. if you set it that way, but on a 32-1 wheel on the cam, it will in fact be 100deg of engine rotation. EDIS8 will wait for four teeth to pass by, then fire output A, and that will be 20deg BTDC. in theory, you need to set the sensor 2 1/2 teeth BTDC, which is obviously meaningless.

*the tach output will be incorrect ie half real engine speed

*the miniumum engine speed MJ can register is 450rpm. that will be hit when your motor is turning at 900rpm, and you will not able to configure it for a lower speed. which may not be a problem; just making the point

*i think the load bin issue is a lot more complicated than you think. note that EDIS works like this: when the gap comes around, it 'knows' how many deg BTDC that is. it is 90 for EDIS4, 60 for EDIS 6 and 50 for EDI8. each tooth is 10deg. MJ looks at the number in the bin, converting it to teeth, then MJ takes the PIP, and delays it by the equivalent amount of time before sending the SAW. (brent, is that fair enough?) so it is not as simple as just halving the load bin values. if you set the vr sensor, say, 5 teeth BTDC, that will be 100deg in reality BTDC. well, you cant do that, because if you wanted to fire a plug at 5deg BTDC, you would be asking EDIS to wait 95deg before firing. i dont think it will do that at all. you cant set the VR sensor 2 1/2 teeth BTDC, so you wil have to choose another tooth. what about 3 teeth before TDC? that would be 60deg in reality. suppose you want to fire at 10deg BTDC. that means making EDIS find the gap, then wait 50deg. that will be 25deg on the map in your setup, so you will have to start with where EDIS thinks the gap is (50deg) and subtract the deg it needs to think it is waiting (25deg), and put 25deg into your bin. if you want to fire at 20deg BTDC, the same calculation sayd you must put 30deg in your bin. yes, there is a linear relationship, but there is also an offset determined by how many teeth before TDC, you set the VR. and you know you have to make some arbitrary decision, because 5 teeth is simply too far advance for EDIS to work.

does anyone think differently? i a feel a bit like nicola tesla conducting a thought experiment here, but i think his capacity for that was a bit greater than mine

*while it is possible that there could be a situation where you will limp home on 3 coils with one dead, rather than not limp home with 1 coil, and one dead, i think that this is so unlikely as to be unworthy of consideration. how often does anyone get stranded witha dead coil?? if it does happen, just call a towtruck.

*EDIS can multi spark, but MJ cannot. previous discussion on this subject indicates that its sole purpose was to slightly enhance emissions or running at very low rpm (cant recall which). there is apparently no benefit to multispark at the rpm you mention ie 3000+ rpm and if there is, i guarantee it will be beyond your abilitiy to measure it with the seat of your pants, or a stopwatch!


bottom line: i think your proposal is so complicated it will send you to early grave from thinking about it.

suggestion: make a 72-2 wheel for you cam pulley, and use an EDIS4, then use COP coils as others have successfully done. then everything will in reality work the way the software says it will work, and you will be MUCH happier.

regards
alexander
sydney, oz.

[Dean924s]

Ok This is what I was wanting to get.. . I have some serious thinking to do. Anyone have a 72-2 wheel around? I guess I could make up one by drilling holes in a spare cam wheel.

I think I had better get a 72-2 wheel mounted up just in case. The rest of the bits are easily obtainable from the local wrecking yards that I will just get the coils and what not to have around.

I see your point re the 100 deg. Hummmmmm I was hoping that this could be corrected with software. Hmmmm I have to think more

[Dean924s]

Ok I have found a source for 72 tooth gears unfortunately it is only 3" in diameter but I will have to go get a Mazda vr sensor (I know where one is in the junk yard.)

It was listed in stock and shipping tomorow. I hope I see it before the weekend.

"6325K67 Steel Plain Bore 14-1/2 Deg Spur Gear 24 Pitch, 72 Teeth, 3" Pitch Diameter, 1/2" Bore "

The source is McMaster-Carr

There are 36 tooth ones as well

Catalog page 1039

Ya I am hedging my bets lol. I have a feeling that you are correct. I just wish that mounting the thing to the crank was possible. I re looked at this and it is really not a good set up for this. The outermost pulley is for the PS and it is a separate unit that is just bolted to the crank. However there is no key to it note is it centered very accurately. The run out on my car is about 1/4 of an inch. I replaced it with another one and it was also out by almost the same amount.

I am not married to the COP. I think it is "cool" but completely not needed. It was just a case of if I could why not? My plan was solely based upon what I have on the shelf (I have the coils from my expedition on the shelf as they tend to die every now and again.) The edis I picked up the other day for a couple of dollars. I think I am going to try to re work the wheel I have to make

I am going to go to the junk yard tomorow and get an edis 4 set up complete with wires etc just to make sure

Has anyone ever re ground a 36-1 wheel to make a 72-2 wheel?

[brentp]

Dean,

I wasn't 100% sure what your motivations were behind the EDIS-8 and Coil on plugs. In principal, it would be a cool exercise to prove out your ideas- but, if you only wanted a system that simply worked really well, I think you will be impressed with the results from a standard EDIS-4 and coil pack.

As I was pointing out- at the relatively 'modest' upper RPM ranges your engine will be seeing you will be getting nearly full spark energy under all operating conditions.

An incremental improvement would be to emulate the Coil-on-plug configuration demonstrated by our user dr.occa- easy to implement using dual EDIS-4 modules.

Keep us posted on your project, and thanks for sharing!!

Regards,

[alexander]

hi dean.
i think a 72-2 wheel on your cam pulley will work well.
i personally put a 72-2 wheel in the distributor housing, and used a mazda CAS. the wheel is only a bit over 30mm across, so they can be made very small. have a look at these two:

http://www.autosportlabs.org/viewtopic. ... ight=#7241
top of 3rd page

http://www.autosportlabs.org/viewtopic.php?t=1648

my wheel was, just like the one you are looking at, a slice off a piece of gearstock i bought online from the USA (of course!).

note that there is a number of ways of constructing a wheel.

you can drill 72 holes in a metal disk then have it machined down to such that you are left with only semi circles at the periphery, which then look like teeth.

you could do that on the cheap by drilling the holes, mounting the disk on a bolt which is then put in the chuck of a drill press (or hand drill). start the drill, hold an angle grinder to the wheel till it is ground down enough to make the holes, into teeth. perfectly concentric to boot!

you could start with a round disk, and (carefully!!) cut slots in its periphery with a small angle grinder.

you can drill the holes in a round disk, then have the sensor read the holes from the side. there is a recent thread where someone has done that.

you can drill the holes into the periphery of a thicker disk, and have the sensor read them as if they were teeth. there is also a recent post where someone has done that too. in fact, he drilled the holes into the harmonic balancer. i have personally just done this for my other car (but still work in progress on the installation)

there are many ways to skin this cat, and with a bit of ingenuity, i am sure you will be able to fabricate a 72-2 wheel to fit on your cam pulley, and there is every liklihood that the teeth will still be big enough to be read by the standard ford VR sensor. the size of tooth which can be read by the ford VR sensor is the only question mark here. i know you have looked at patriq's thread:

http://www.autosportlabs.org/viewtopic. ... 5&start=15

his initial wheel/sensor combo did not work, and it seems that the small teeth could not be read by the ford VR sensor. his later adaptation worked:

http://www.autosportlabs.org/viewtopic. ... ight=#9080 (page 4 for the photos)
as you have already referred to the mazda item, that shouldnt be a problem.

regards
alexander.
sydney, australia.

[Dean924s]

alexander,

Thanks for the info. I had read most of that but not all. Really good info!!!. My biggest worry is getting the wheel mounted up and centered properly as it looks like the RX7 VR units need to be much closer to center than the Ford ones. I may have to mount it up and then true it with a rubber mallet and then finish it on the lath. The old spin it and hit it with a marker too see what hits first and tap it with the hammer to center it.

However I am confident that once I get the gear set up it will be a relatively simple installation.

I went to the Junk Yard today on my way to the office and scored two rx7 sensors, 3 edis4 units 2 coil packs a couple sets of wires, mounting brackets for both the coil packs and the edis units. Plugs and pigtails for all as well. The local yard charges for the coil packs ($20.00) each. I don't think they understood what all the other stuff I got was. It looked like a big mess of spaghetti when I picked it up and showed it to my friend at the counter.

I will sort it out and post up photos of my hall tonight. What this also means is I will have spare edis units and other bits when I am done.

[alexander]

hi dean
if it was too easy, it wouldnt be fun .... but if you have a lathe, then you definitely wont be having a problem. after all, you could weld it in place, then machine the trigger wheel to have 0 runout.

i assume you do have a Ford VR sensor as well? when you have the 72-2 wheel mounted, it would be good to know if the ford VR sensor will trigger from it, as that would give a useful benchmark.

anyhow, will be on the lookout for your updates.

regards
alexander

[Dean924s]

Arrrrrrrrrrrrrrgggggggggg.. .. Avast ye land lubbers we pillaged and had a good hall.. .. . . OK I am not a good Pirate but I feel like I was today. This is photos of my hall as I put it to the counter at the yard.