Cam Sync Sensor Wiring

[Chris Duncan]

Accel Gen7 ECU
Fast Edist DIS

Using a cam sync sensor from a Ford Exploder, believe it's a 1996. The one that goes in the place of the distributor.

It's a hall effect (??) sensor with 3 terminals. The local dealer isn't too helpful and won't let me look at the manuals. Need to figure out what wire goes to which terminal.

from the ECU and DIS there's a cam signal wire and a signal return (ground). Although the ECU and DIS wire diagrams don't show a 3rd wire for the cam sensor I'm assuming it's a 5v reference which the tps and map sensor both have.

 

[Neal]

Not sure if the order is correct but connections should be...

+5v supply source
ground
Hall effect signal

 

Mitchell is sort of unclear but as best I can tell, the connector view is from the WIRE side of the conector. So looking at the wire side, with the locking clip up, GND/SIGNAL/12V (not 5v)

 

[Chris Duncan]

Mitchell is sort of unclear but as best I can tell, the connector view is from the WIRE side of the conector. So looking at the wire side, with the locking clip up, GND/SIGNAL/12V (not 5v)

So how would be the best way to test this before connecting it to the 3K worth of ECU's? Could you just run a 12v jumper, crank the engine and check the signal? I have a fluke 88 so should be able to test just don't know how?

 

[Ross Nicol]

Kalun an oscilloscope is what you need.Apply the DC voltage to the correct pins (sorry can't help there) and then hook the scope to the signal and ground.When you crank you should have a healthy series of square waves on the scope. Your Fluke 88 is probably like my old 77 and although it will read on AC it won't be accurate and you can't determine the waveform is a good shape. I've got plenty of oscilloscopes here in OZ but I don't suppose thats much use to you at the top end of the world.

Ross

 

[Paul Thompson 'Hooligan']

Hall Effect Sync Signal

Hi Kalun -

The signal for the hall effect output would, as Ross says, look square as the upper trace seen below. The lower trace is the mag output from an MSD Distributor.

 

So how would be the best way to test this before connecting it to the 3K worth of ECU's? Could you just run a 12v jumper, crank the engine and check the signal? I have a fluke 88 so should be able to test just don't know how?

Go here http://www.vems.hu/wiki/index.php?page=ElectronicDesign/SoundRecorder and download the one that says "soliton". Build the simple circuit shown towards the bottom of that page. Then you can hook it up to your sensor and read the waveform. TO use the program, you need to hit record, activate your sensor (crank engine), then hit stop and playback, it will then give you a waveform to view.
As for just adding 12v to the sensor and spinning it, try it, but that might not work. Most likely you will need pullup/pulldown resistors unless it is built into the sensor. Easiest way would be find one in a car and the black is gnd so the opposite one is 12v and the middle is signal. If that isnt possible, then measure the resistances between all the pins on the sensor and I will give you a best educated guess. In all honesty however, it is highly doubtfull that you could fry the ecu by hooking up the sensor wrong, it just wouldnt work.

 

[Neal]

Kalun,

I have access to a scope and would be happy to check the pinout.

Regards,
Neal

 

Did some checking in alldata. Pullup is internal to the sensor so all you have to do is hook the sensor up to the battery the way you think is right and put your voltmeter on dc volts between ground and the middle wire and slowly turn the sensor, you should alternate between less than 2 volts and above 8 volts.

 

[Chris Duncan]

Did some checking in alldata. Pullup is internal to the sensor so all you have to do is hook the sensor up to the battery the way you think is right and put your voltmeter on dc volts between ground and the middle wire and slowly turn the sensor, you should alternate between less than 2 volts and above 8 volts.

hooked it up with bat + on the right, ground on the left and the signal wire in the center. Put the volt meter on the ground and signal. Hand cranked the motor, all I get is battery voltage, no change.

So it's a bad sensor, or it's the wrong year, not a '96, for your data. I'm going to find another Ford store, one that has some more helpful techs. Or maybe Electromotive, they sell these things aftermarket.

 

[Neal]

You might also chack with MSD as I think they offer one also.

 

hooked it up with bat + on the right, ground on the left and the signal wire in the center.

May as well try it the other way around if you havent all ready.

 

[Chris Duncan]

May as well try it the other way around if you havent all ready.

THAT'S IT!

I was just afraid to try it. It goes back and forth from 12v to 0v

now some coolant and fuel and this thing may start

 

[Chris Duncan]

Running

It was good in a way that I found out how to check the cam sensor because it needed timing anyway.

It started and runs really nice after I got the Tbodies syncd

this thing is SICK

quick throttle response and nice little lump at idle

The Gen 7 is sweet, if I'd had the cam sensor timed from the start it would have started first try

Nothing like the sound of a crossover V8 to get you psyched to get this thing on the road

yeeeeeee haaaaaaaa

 

[Chris Duncan]

May as well try it the other way around if you havent all ready.

It's running but not very well and it's hard coding the cam sync in the Gen7.

I'm going round and round with Accel and Edist manufactures trying to figure this out.

(in review,
Accel Gen7 ECU,
FAST Edist distributorless with LS1 coils,
Ford Explorer cam sync sensor, (hall effect??)
Moroso magnetic crank sensor.)

The initial problems I see with this whole setup is the fact that the Edist and the ECU are sharing the cam sync signal.

Before I went this route I asked Accel, FAST Edist, and my local Dyno guy (that sells them both) if it was possible to use these 2 components together and to share the cam sync signal. They all said yes.

This cam sync signal sharing means 2 things though, first there's a voltage drop. While cranking I'm getting about 0 to 10 volts at the ECU with the Edist disconnected. With the Edist connected I'm getting 0 to 8 volts. No scope so I can't get a running measurement.

Secondly there's a timing issue. The optimum cam sync timing for the ECU is 45 deg BTDC, the optimum for the Edist is 60 deg. Both have a window of variance that is acceptable but these windows don't overlap very much. That and I don't have any way to time the cam sync dynamically, am now timing it statically with the Fluke. Is this done with a scope? I have been able to time the crank sensor with a timing light, so I know that's correct.

Now I'm getting conflicting information about what type of sensor the Ford unit really is. While first agreeing that it was a hall effect sensor, Accel is now saying it's a IPU (magnetic) sensor. Inspecting the sensor I do see that it has a magnetic pickup, but can a hall effect sensor be magnetic? I don't think it is an IPU sensor though, because when I changed it to this in the Edist configurations it won't fire at all.

The Gen7 has 3 choices in the software for cam sync sensor type, IPU (magnetic), hall effect, and 5V sq wave. And it runs on all 3 and codes the cam sync sensor on all 3.

the Edist has 2 choices, hall effect and magnetic, and it only runs on hall effect.

I'm about to construct a double cam sync sensor, with 2 cam sync sensors that are separately timeable. This would solve both the timing and voltage issues. I'm going to try one other thing though and that's set the cam sync to rising edge, although I seem to remember someone saying falling edge was better/more accurate. Don't know what else to do at this point other than try another cam sync sensor although Ford wants $200 and I suspect this type sensor either works or it doesn't.

On a side note the initial maps generated by the Gen 7 software were way rich for the 8 stack. Accel sent me another map they had for an 8 stack 402 and it's way leaner until full throttle.

 

The initial problems I see with this whole setup is the fact that the Edist and the ECU are sharing the cam sync signal.
While cranking I'm getting about 0 to 10 volts at the ECU with the Edist disconnected. With the Edist connected I'm getting 0 to 8 volts. No scope so I can't get a running measurement.
QUOTE]

I personally dont think that drop will be an issue. It is kind of hard because both FAST and Accel dont give much info on the web sites but all the ecu's are doing is looking for a change of voltage above a certain number. The fact that you actually get spark would tell me that the the sensor is fine, otherwise there would be nothing to sync #1 cylinder to. (I assume you are running a trigger wheel with a multiple of 4 teeth right, not a missing tooth?)
One thing you can do to raise the voltage is put a 1/4 watt 10k ohm resister between 12v and signal but there is a risk of more electrical noise. Which is another good point. It would be nice to see a scope reading but just make sure there is a good sheilding wire around the cam sensor wires and that it is only grounded on 1 side, this includes shielding your spliced wire.

Secondly there's a timing issue. The optimum cam sync timing for the ECU is 45 deg BTDC, the optimum for the Edist is 60 deg. Both have a window of variance that is acceptable but these windows don't overlap very much. That and I don't have any way to time the cam sync dynamically, am now timing it statically with the Fluke. Is this done with a scope? I have been able to time the crank sensor with a timing light, so I know that's correct.
QUOTE]

This would be the most probable cause in my opinion. I dont really know the system at all but with mine, to fine tune it, force ignition advance to 0 degrees, fire a timing light at the crank, turn the cam sensor like it is a distributer untill the timing mark is right on 0 degrees. It would be nice if the software let you set where you have your sensor positioned so you could fine tune it in the software if the sensor wasnt adjustable.

Now I'm getting conflicting information about what type of sensor the Ford unit really is. While first agreeing that it was a hall effect sensor, Accel is now saying it's a IPU (magnetic) sensor. Inspecting the sensor I do see that it has a magnetic pickup, but can a hall effect sensor be magnetic? I don't think it is an IPU sensor though, because when I changed it to this in the Edist configurations it won't fire at all.

The Gen7 has 3 choices in the software for cam sync sensor type, IPU (magnetic), hall effect, and 5V sq wave. And it runs on all 3 and codes the cam sync sensor on all 3.

the Edist has 2 choices, hall effect and magnetic, and it only runs on hall effect.
QUOTE]

I have never heard of the term IPU sensor. I assume that means inductive pickup which I assume means variable reluctance, which would explain why it wouldnt start with the edist system. Ford does use VR type cam sensors as well but the fact that you get a square wave signal and there is 3 wires at the sensor, there is no way it is a VR sensor.
Hall effect sensors vary, some have a little magnet that spins and the sensor reads that, others just need a normal hunk of steel as a tooth and pick that up, maybe that is the difference? Also, I think that shutter window type thing in the earlier 5.0l mustangs is a type of hall. Either way, they should give the same square wave signal so I dont think thats it but I would double check with the ecu guys.

I'm going to try one other thing though and that's set the cam sync to rising edge, although I seem to remember someone saying falling edge was better/more accurate. Don't know what else to do at this point other than try another cam sync sensor although Ford wants $200 and I suspect this type sensor either works or it doesn't.
QUOTE]

Generally digital signals trigger on the falling edge, vr trigger on rising. I dont think the accuracy would be anything significant enough to cause problems. I guess it depends on what hardware the ecu's use to interpret the sensor. Either way, I would give it a try, at worst, it wont start.
Double sensors wouldn't hurt and like you said, will help with the timing issue.

**The only other solution I can think of and this is what I would do, is to use the cam sensor for the edist only and dont use a cam sync for the accel. Get it all running nice in batch fire mode first. After that, if you think that the minute amount of gains and large amount of frustration you may get by running sequential are worth it, then go for it. Keep in mind, the amount of fuel you waste trying to make that engine that much more efficient will never be recovered in the improved fuel economy you will get.**

 

[Chris Duncan]

"" The fact that you actually get spark would tell me that the the sensor is fine, otherwise there would be nothing to sync #1 cylinder to.""

I agree it's always fired off fine. I think a lot of the popping is due to fouled plugs, the initial map was way too rich. The VE map estimator in the software is based on a 4 barrel carb. Accel emailed me a map for an 8 stack 402 and now the O2 readings are much closer to target. But the old map fouled the plugs so I need to clean them and restart.

But the cam sync keeps coding in the ECU even though it seems to be running ok.

"" (I assume you are running a trigger wheel with a multiple of 4 teeth right, not a missing tooth?)""

correct, 4 magnets and IPU sensor (inductive pick up)

""One thing you can do to raise the voltage is put a 1/4 watt 10k ohm resister between 12v and signal but there is a risk of more electrical noise. Which is another good point. It would be nice to see a scope reading but just make sure there is a good sheilding wire around the cam sensor wires and that it is only grounded on 1 side, this includes shielding your spliced wire.""

scope on order. all crank and cam signal wires shielded on one end only.

"""""Secondly there's a timing issue. """""

""This would be the most probable cause in my opinion.""

So now I'm trying a trick to put the DIS on the rising edge and the ECU on the falling edge of the cam sync. The falling and rising edges are 170/190 deg apart in the cam sync sensor so this gets them both closer to optimum. The Edist doesn't care which crank signal is after the cam signal as long as you tell it which number it is. It doesn't necessarily have to be #1. With the ECU it has to be #1.

""I dont really know the system at all but with mine, to fine tune it, force ignition advance to 0 degrees, fire a timing light at the crank, turn the cam sensor like it is a distributer untill the timing mark is right on 0 degrees.""

Yes I did this with the crank sensor, hooked up the timing light. But how do you do this with the cam sync sensor? My timing light needs a plug wire to trigger from and the cam sync on this system isn't used for timing of ignition/injection, just to tell the DIS/ECU which crank signal is for #1 cyl.compression stroke.

""It would be nice if the software let you set where you have your sensor positioned so you could fine tune it in the software if the sensor wasnt adjustable.""

agree

""I have never heard of the term IPU sensor. I assume that means inductive pickup which I assume means variable reluctance, ""

correct, Inductive Pick Up, magnetic or (flying magnet)

""which would explain why it wouldnt start with the edist system. Ford does use VR type cam sensors as well but the fact that you get a square wave signal and there is 3 wires at the sensor, there is no way it is a VR sensor.""

agree

""Hall effect sensors vary, some have a little magnet that spins and the sensor reads that, others just need a normal hunk of steel as a tooth and pick that up, maybe that is the difference?""

this cam sensor has a metal wheel with a 3 wire magnetic pickup. As opposed to the crank system which has a 4 magnet wheel and 2 wire IPU pickup.

""Generally digital signals trigger on the falling edge, vr trigger on rising. I dont think the accuracy would be anything significant enough to cause problems. I guess it depends on what hardware the ecu's use to interpret the sensor. Either way, I would give it a try, at worst, it wont start.""

both the DIS and ECU have the option of rising or falling. I need to get over the fear of damaging components with experimental settings or better yet learn just what you can get away with and what you can't. Had a scare today trying to splice off the 5v reference from the MAP to power the cam sync. ECU started hard coding the 5v ref signal. It cleared after a hard off and restart.... sigh of relief.

""Double sensors wouldn't hurt and like you said, will help with the timing issue.""

that's a last resort, much rather do it with software configuration than hardware fabrication.

""**The only other solution I can think of and this is what I would do, is to use the cam sensor for the edist only and dont use a cam sync for the accel. Get it all running nice in batch fire mode first.""

Tried that today, no longer codes the cam sensor but that's a bandaid, after all this money I'm taking the time to do it right. So with batch fire does it just fire each injector on every TDC of it's respective cylinder, instead of just the compression stroke? Or 2 injectors per every crank signal?

""After that, if you think that the minute amount of gains and large amount of frustration you may get by running sequential are worth it, then go for it. Keep in mind, the amount of fuel you waste trying to make that engine that much more efficient will never be recovered in the improved fuel economy you will get.**""

You may be right, I know it dumped a ton of fuel with the initial map, the whole exhaust is sooted up.

here's a copy of the email that's going out to Accel in the morning.


Ford 302 HO 1991 (approx 350-400HP)
AFR 185 aluminum heads
TFS cam, .500 lift, 280 duration
10.2:1 comp
Stainless 1-5/8" headers
TWM 8 stack induction, 48mm t-bodies
32 lbs lucas low res. injectors
Moroso 4 magnet crank trigger wheel, IPU (3deg index offset)
Ford Explorer 3 wire hall effect cam sync sensor (12v sq. wave)
FAST Edist distributorless ignition system
Vette LS1(8) coil on plug with resistor wires.
ACCEL Gen7 ECU with UEGO O2 add-on
Latest Gen7 CalMap software from the web.

Cam sync signal is shared between DIS and ECU, wires split about half way, all wires shielded and grounded on only one end.
some voltage drop with shared signal
Fluke 88 cam sync volt readings on cranking (back probed at ECU)
With DIS and ECU 8.2V to .3V
With ECU only 10.6V to .04V
(question accuracy, scope on order)

Cam sync signal static timed at 50 deg BTDC on falling edge in ECU (optimum 45deg)
Cam sync signal static timed at 60 deg BTDC on rising edge in DIS (optimum 65deg)
The rising and falling edges are 170/190 degrees apart in the cam sync sensor so having ECU on falling and DIS on rising means they are both closer to their respective optimum cam sync timing. Both have about a 20deg window so this should be ok.

Edist only has 2 options for cam sync signal, 12v hall effect or IPU
Edist requires a 12v signal if hall effect cam sync sensor is used. Attempted to set cam sync sensor as a IPU in the Edist DIS and it would not see signal or run so assuming this is a hall effect sensor. DIS module has LED's that show if crank and cam signal is present.

Tried cam sync sensor settings in CalMap on IPU, hall effect, and 5v Sq wave, hard codes 72(cam sync) on all 3 settings. But now realize ECU must be off for 20 seconds for changes to take effect, so may retry 5v Sq wave since I know it's not a IPU sensor.

Currently have ECU set on batch fire injection and code 72 is no longer present. Some popping in intake but this is probably sooty plugs because of rich CalMap VE estimator map. Emailed (8 stack 402) .ecm map is much better. O2 readings are much closer to target. Going to clean plugs and restart.

 

"But the cam sync keeps coding in the ECU even though it seems to be running ok. "

Is it possible to set the cam sensor trigger point to exactly what the ecu wants and then see if it still codes? How long does it take to code? I am thinking if it could code during 20 seconds of cranking then you wouldnt have to be too concerned about the spark and just disconnect the coils and crank, see if the sensor still codes. Then disconnect the spark module all together and crank, see if it codes then.

"So now I'm trying a trick to put the DIS on the rising edge and the ECU on the falling edge of the cam sync."

Any difference?

"Yes I did this with the crank sensor, hooked up the timing light. But how do you do this with the cam sync sensor? "

Sorry, I am an idiot, I wasnt paying attention when I read that. I have never acutally dynamically timed a cam sensor. I always just stick a degree wheel on the crank and my voltmeter on the sensor wire, turn engine over untill it is at the required degrees and adjust the cam sensor to turn on then, I have never had a problem doing it that way.


"correct, Inductive Pick Up, magnetic or (flying magnet)
this cam sensor has a metal wheel with a 3 wire magnetic pickup. As opposed to the crank system which has a 4 magnet wheel and 2 wire IPU pickup."

2 wires would tell me that it is NOT a flying magnet system but instead a stationary magnet (the sensor) and the teeth are just plain steel, this would be an analog signal (vr). Flying magnet is a hall type (digital)

"I need to get over the fear of damaging components with experimental settings or better yet learn just what you can get away with and what you can't. Had a scare today trying to splice off the 5v reference from the MAP to power the cam sync. ECU started hard coding the 5v ref signal. It cleared after a hard off and restart.... sigh of relief."

They arent as delicate as everyone thinks. They can take a fair bit of messing about. The only thing that is worth being concerned about is the motor, especially if it is not broken in. Best to break it in with a carb, then start messing about. As long as it is broken in, you can do basically anything under no load without worry of damage.


"Tried that today, no longer codes the cam sensor but that's a bandaid, after all this money I'm taking the time to do it right. So with batch fire does it just fire each injector on every TDC of it's respective cylinder, instead of just the compression stroke? Or 2 injectors per every crank signal?"

There are different types of batch fire, bank to bank, or in pairs (or other groups).
As much as it sounds "perfect" to shoot a stream of atomized fuel right into the cylinder when the intake valve is open, it only works up to about 1200rpm. I could do the math but I would require thinking. After around 1200rpm, the valve opening is so short that the injectors cant flow enough fuel to fill the cylinder, unless you have stupidly large injectors in which case the thing wont idle worth shit anyways no matter what.
So after 1200 rpm, the fuel will pool on the back of the valve and wait for it to open so it can dump in, which is what batchfire does right from the start.
The only benefits to sequential is better emissions and fuel economy at idle and less pulsation in the fuel rails. The benefits are worth it if you are a car manufacturer trying to squeeze every ounce you can out of a million cars, you tune 1 then program the rest of them. FOr a 1 time deal, dyno time (required, not optional) and fuel used, is never going to be recovered.
Definately get it running 100% on batchfire first, get experience and an understanding of it all, in the meantime, play with sequential if you are bored and need a new challenge.

 

[Chris Duncan]

""Is it possible to set the cam sensor trigger point to exactly what the ecu wants and then see if it still codes? How long does it take to code?""

set to ECU optimum timing, codes as soon as you go above 1500rpm (which is normal code threshold)


""I am thinking if it could code during 20 seconds of cranking then you wouldnt have to be too concerned about the spark and just disconnect the coils and crank, see if the sensor still codes. Then disconnect the spark module all together and crank, see if it codes then.""

wouldn't work because of rpm threshold, it has to be running

"So now I'm trying a trick to put the DIS on the rising edge and the ECU on the falling edge of the cam sync."

""Any difference?""

none, still codes

""Sorry, I am an idiot, I wasnt paying attention when I read that. I have never acutally dynamically timed a cam sensor. I always just stick a degree wheel on the crank and my voltmeter on the sensor wire, turn engine over untill it is at the required degrees and adjust the cam sensor to turn on then, I have never had a problem doing it that way.""

agree with the allowable variance it shouldn't be a problem

""The only thing that is worth being concerned about is the motor, especially if it is not broken in.""

it was just a re-ring bottom end but a new top end. With the 8-stack map it's running close to target AF ratio, it's incorrect mixture that's a problem right?

""There are different types of batch fire, bank to bank, or in pairs (or other groups). ""

according to "help" it's staggered batch, or "bank to bank"

""As much as it sounds "perfect" to shoot a stream of atomized fuel right into the cylinder when the intake valve is open, it only works up to about 1200rpm. I could do the math but I would require thinking. After around 1200rpm, the valve opening is so short that the injectors cant flow enough fuel to fill the cylinder, unless you have stupidly large injectors in which case the thing wont idle worth shit anyways no matter what.
So after 1200 rpm, the fuel will pool on the back of the valve and wait for it to open so it can dump in, which is what batchfire does right from the start.
The only benefits to sequential is better emissions and fuel economy at idle and less pulsation in the fuel rails. The benefits are worth it if you are a car manufacturer trying to squeeze every ounce you can out of a million cars, you tune 1 then program the rest of them. FOr a 1 time deal, dyno time (required, not optional) and fuel used, is never going to be recovered.
Definately get it running 100% on batchfire first, get experience and an understanding of it all, in the meantime, play with sequential if you are bored and need a new challenge.""

this makes sense to me. This is my first programmable ECU so might as well start with the basics and work up from there, no need to bite off more than necessary at first.

Going to post another thread with questions about PC oscilloscopes.