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

[Dean924s]

To add to dr.occa's post, yes, it counts teeth, but the spacing is critical as it uses that information to determine when to fire the plugs.

If you had an imprecise trigger wheel, yet close enough to prevent the EDIS module from malfunctioning, you would have a condition where the actual ignition advance would differ from the commanded ignition advance.

Hope this helps.

Yes makes sense. I also begs the question. Could the firm ware be modified to correct this?

I was assuming that the EDIS uses the tooth count versus an assumed 10 degrees of separation to initiate the timing of the spark. This would also allow for easy calculation of RPM as well. In other words it is only counting the pulse or the waves generated and it just assumes that there is 10 deg of separation thus it can calculate RPM and by the count time the spark. It then uses the -1 tooth as the reverence to then back calculate where it is and when it needs to initiate spark.

I assume this is how it works because it if actually measured the distance between the pulses and did not assume that it was 10 degrees separation then you would have no constant and could not calculate anything.

[Alexander_Monday]

According to Ford patent 4,922,874:

"Position sensor 48 produces position signals each 10.degree. of timing wheel rotation as teeth pass across the sensor, except during the passage of a missing tooth the occurrence of which is detected as an indication of absolute position. For purposes of SAW pulse conversion to a spark angle, teeth prior to each top dead center firing position are counted down as shown. Thus, the tooth corresponding to cylinder #1 top dead center is numbered zero with teeth which pass sensor 48 prior to tooth zero having corresponding numbers. Likewise, teeth counting down to top dead center for cylinder #2 are likewise numbered. The crank marker interval as shown in FIG. 5 is 10.degree.. A spark advance of 15.degree. BTDC, for example, would occur between crank marker 1 and crank marker 2.

..............

In operation, once a SAW pulse has been decoded, the ignition module monitors crankshaft position waiting for the occurrence of a position signal indicating the marker corresponding to the value of bits M1 to M3. Once this crank marker has passed, the fractional portion of an additional interval until the next expected crank marker equivalent to the value of byte 1 in FIG. 6b is counted out. Once the fractional portion has expired, the spark event is initiated."

So, according to this the 10 degree spacing and count of the pulses is critical to proper operation.

[Alexander_Monday]

Several posts after I started typing mine. To address them

According to http://www.dainst.com/info/edis/edis.html

"The first thing the EDIS module does is sync up to the missing tooth of the wheel. The EDIS module keeps an internal counter of teeth, if for some reason the counter loses sync with the wheel, the module will attempt to resync."

So it sounds like the EDIS has a built in tolerance level between what the timing should be and what the actual position is based on teeth count beyond which it decides it has an error. It would depend on what this error level is.

[dr.occa]

Several posts after I started typing mine. To address them

According to http://www.dainst.com/info/edis/edis.html

"The first thing the EDIS module does is sync up to the missing tooth of the wheel. The EDIS module keeps an internal counter of teeth, if for some reason the counter loses sync with the wheel, the module will attempt to resync."

So it sounds like the EDIS has a built in tolerance level between what the timing should be and what the actual position is based on teeth count beyond which it decides it has an error. It would depend on what this error level is.

then this makes sense and substantiates the need for consistent spacing between teeth. the EDIS bases its calculations on its initial sync up to the trigger wheel. if there is any imprecise "rhythm" prior to the -1 tooth it could theoretically put the EDIS unit in a re-sync loop thereby never generating a constant wave-form to base any timing off of. essentially leaving you in limp home mode despite the quality condition of your MJ.

my head's beginning to swell. i think i need to take something...

[Dean924s]

Figured I would post these up for comment.

I screwed up my first go at the 72-2 so I turned it in to a 36-1 (Don'T ask it was really dumb)

I think I may try to see if I can use it on the crank. I am still going with the cam mounted unit but eventually I will go to a crank unit if I can figure out a good/simple way to install the trigger wheel and the sensor.

Anyway here we go.

[Dean924s]

Some more

[Dean924s]

And more

[Dean924s]

First rule of this task is to count and then mark the teeth before you grind. Then re count again and even a third time!!

All three wheels The untouched 72 tooth spur gear at the top as I get them. The 72-2 on the bottom left and the 36-1 on the bottom right.

[Dean924s]

Something to give you an idea of the scale.

[brentp]

Looks good. I like the idea of using the spur gear!

[alexander]

First rule of this task is to count and then mark the teeth before you grind. Then re count again and even a third time!!

i think we know what happened to the first wheel!

[Dean924s]

First rule of this task is to count and then mark the teeth before you grind. Then re count again and even a third time!!

i think we know what happened to the first wheel!

To elaborate on my screw up so others will not make the same mistake I did.

1. Ware your glasses (if you need them) I need reading glasses and this was the real reason I screwed up. I had marked the gear, counted out etc. But I could not see. At the last moment I swear that the tooth next to the marked one looked like it was marked so away I went with the grinder. When I took it out of the vice I was.. . . well.. . . annoyed.

So how do you make one of these things. After screwing it up here is my method.

You will need a red permanent marker and either a black marker or a bit if "white out" (also called correcting fluid) or a paint pen works well. Phys get the fine point type as well.

The first step is to mount the wheel and see if it is out of round or of the mounting is off center. It is critical to do this using the sensor in its installed location. There are to ways to do this and it is important to follow the one that applies to you.

If your installation allows you to rotate/adjust the gear / trigger wheel 360 deg you can then install it and turn the motor and see if there is a tooth that is closer to the sensor than others. If there is adjust the gear so that this tooth would be the -1 position as installed in your motor and re check that it is still closes to the sensor when the motor is rotated. If it is then mark it and remove the gear. If it is not and another gear is now closer to the sensor then it is probably not the gear but something in the mounting of the gear that is not centered and you should try to correct this. From my experience with trigger wheels if you can See that it is off center you probably need to adjust it. Anyway once you get this straitened out you can move on to counting teeth.

If the gear is mounted so it is fixed relative to the crank position you have to install it and then set up the motor with the gear and sensor in place and count back to the tooth that needs to be -1 (the one that ends up getting removed in my process) and mark it

By checking the tooth alignment and making the tooth that is closest to the sensor the -1 you are removing the high point on the wheel and it will allow you to get the sensor that much close to the wheel. A Lot of nit picking but I think worth it in the end.

Now that you should have a gear with one marked tooth. Label it #1 and count around marking every 5th tooth (either direction it does not matter) until you get to 35. You then mark 35 and 36. Now go back and re count and make sure that you did it correctly.

Now repeat the process of counting and marking starting with the tooth that is after the tooth you just marked #36 (label it #1) and continue around to where you started marking and labeling every 5th tooth up to and including #35 and then mark and label #36 (it should be next to your original #1). If it is not re count review what you have done and find the error. (including verifying that you have a 72 tooth gear) When you are done you should have #35,36 & #1 marked 180 degrees out from one another. You should be able to put a strait edge across the teeth labeled #1 and it should bisect the gear in half exactly. If you are comfortable that everything is in line mark the teeth that you labeled #1 with while paint. Mark both the side and the top. I would also suggest that you put some tape over teeth on either side to make (I did not do this the last time but will do it if I have to make another one).

If you want to double check repeat the process on the other side of the gear starting with the tooth that you originally marked with the gear on the motor. You should end up with the sim two teeth marked. If you don't recount measure double check!!!! You have to get the correct!!!

Now that you are sure that everything is correct NOW RE COUNT AGAIN and check all your marks.

OK now grind away. I used a dremel to remove the two teeth on the 72-2. It is slower but allows more control and has a better finish. On the 36-1 wheel I used my 3" makita grinder with a very fine cutting blade, not a grinding blade. Making the 36-1 wheel is actually easier as you just start at one location and remove every other tooth then come back and remove the "1" in the location you want it. A Lot more work but much less technical and less chance of making an error as long as you pay attention. Even if you accidentally grind two adjacent teeth (you can only make this mistake once) grind a third and make that your -1 and then keep going around the gear. The mistake becomes your -1 location. Make a second mistake and you have a piece of scrap metal

I hope this helps. It really is not hard at all it is just a case that one mistake will usually result in you having to start over with a new part.

[Dean924s]

Brent,

You may want to move this thread some place else on your board as it seems to have morphed in to my build thread.

Anyway After trying several way of mounting the VR sensor I finally came up with what I think will be the final way to do it.

First off you should be aware that the sensors I am using are form a FC RX7 (86-91). The 3" gear / trigger wheel I am using is about as large a diameter that you can use with this sensor with out having to bend the sensor and or grind some of the mounting bracket away . I still may do this but for now it seems to be close enough to the gear and the magnet/mounting part seems to clear as well. We will see.

Anyway here is what the stock cap and rotar look line on my car.

[Dean924s]

More photos showing the cam gear cover removed.

Note the thing that the router is mounted to. It is in fact just a retainer that holds the timing belt gear in place and provides something to put a wrench on. Once the rotor is removed there is a cheese head bolt inside of it that you have to remove to get the thing off.

[Dean924s]

The bolt and gear retainer removed. Also note that this is on my spare head/cam assembly. I am going all my fitments here before I put it on the car.