Cam sensor and ignition modifications

[gse_turbo]

I had a thought not too long ago about the cam sensor and reluctor. Moreso the reluctor.

So everyone is on the same page. This is about the exhaust cam gear on the 2.0 16v ECOTEC.

The gear has the mag pickup (pretty sure it's mag). I was thinking about using two intake gears and modifying the one to have an adjustable sensor pickup. The thought is to mechanically affect the ignition timing seperate from affecting the cam timing.

So the questions...
- does the cam sensor have that much control over ignition timing if at all?
- how accurate is the sensor and would in see a change with a huge movement?

...I'm sure other questions are there but I'm not thinking at the moment.

Sown at are your thoughts?

[benzino]

well the australian models don't even have a cam sensor so I don't think it can affect it too much, but then again they aren't OBDII, are the american models?

[gse_turbo]

Yeah. The US models are OBDII.

[PrecisionBoost]

All timing is off the 60-2 reluctor wheel.

The only purpose for the sensor on the cam is to get the sequential injection timing.

That is to say it's job is to tell you if cylinder #1 is in it's intake or power phase.

I thought of the same thing over 10 years ago, did some digging only to be disappointed that everything comes from the reluctor wheel.

I also thought about changing out the cam sprocket, but it would require that I figure out a way to go from sequential injection to batch injection.

The main issue was that the system would be injecting twice the fuel per revolution, which would be great for when the car is under boost but not so good at idle.

[PrecisionBoost]

In my case I was looking to retard the ignition, which is the easy part as you simply need a small programable delay ( any MSD )

However, if one wanted to advance timing, the only way to do it would be to reposition the variable reluctance sensor by several degrees ( to get more advance ) and then use an aftermarket ignition system to bring the timing forward to what ever level you wanted.

Alternatively I thought about installing a 60-2 timing wheel to the crank pulley, then I could make my own adjustable sensor bracket

There are quite a few guys making external 60-2 reluctor wheels, some for Megasquirt, some for use with MSD and Accel ignition systems.

[ls400x]

All timing is off the 60-2 reluctor wheel.

The only purpose for the sensor on the cam is to get the sequential injection timing.

That is to say it's job is to tell you if cylinder #1 is in it's intake or power phase.

I thought of the same thing over 10 years ago, did some digging only to be disappointed that everything comes from the reluctor wheel.

I was just reading this and thought "wow, concise correct technical information"

Then you dropped this bomb

The main issue was that the system would be injecting twice the fuel per revolution, which would be great for when the car is under boost but not so good at idle.

[PrecisionBoost]

I was just reading this and thought "wow, concise correct technical information"

Then you dropped this bomb

The main issue was that the system would be injecting twice the fuel per revolution, which would be great for when the car is under boost but not so good at idle.

Please enlighten me as to what your talking about.

Without sequential via the cam sensor the ECU would not be capable of differentiating between cylinders 1/4 or 2/3 and therefore it goes into a batch injection sequence where it fires the injectors 1&4 then 2&3 at the same time 180 degrees apart.

I have forced the ECU into this mode, the engine runs like shit at idle, but it still runs.

You would think that the ECU would cut the duty cycle by half since it's injecting into each cylinder twice instead of once, but I suppose GM only cared that the engine stay rich to protect itself, based on what I could see the engine went into open loop error mode and ignored the O2 sensor data.

So........ what I said is true, removing the cam sensor will make the engine run rich, which is bad for idle and ideal for someone who wants to run small amount of boost ( so long as they use their own wideband to monitor air/fuel )

[ls400x]

It still has a missing tooth once every 360 degrees of crank rotation, I'd imagine it would revert to batch 1+4 and 2+3 fuel and spark (2 cylinder ecm!) without a cam synch signal. Even if you were correct about the injector duty cycle doubling when there is no sync signal, you can still only reach a maximum of 100% injector duty cycle (on continiously). I would not call crudely doubling the injector duty cycle across the board as "ideal" in terms of a tuning solution for forced induction.

[PrecisionBoost]

Oh don't get me wrong, I wouldn't want to see anyone try using a cam sensor delete as a forced induction tuning method unless they have the ability to tune the air/fuel with a piggyback.

Sometimes it's nice to tune in open loop. ( where the factory ECU ignores the factory O2 sensor and automatically runs a preloaded ignition fuel map where the engine runs extremely rich and retards the ignition )

This way you don't get the ECU freaking out and doing stupid shit when you are purposely running a nice mean best torque air fuel around 12.3:1

The ECU is not stupid, it's running a map for this exact scenario where it looses the cam sync signal.

It will not run 100% duty cycle, when it switches to batch fire it pulls the duty cycle back to accommodate, but it's still stinking rich at idle because it can't decrease the duty cycle enough to give a stable idle.

That is to say if it's running 1.6% duty cycle at idle when firing in sync with the intake stroke, the ECU can not run 0.8% under batch fire, the dwell on most OEM injector is too large, you would have a real hard time dropping it below 1% duty cycle firing twice per revolution.

Thus, you get lumpy uneven idle with stinking rich air/fuel because the engineers at GM were smart enough to know that the engine will run ( poorly ) at 1.1% duty cycle in batch fire mode under open loop ( equivalent to 2.2% sequential or 40% more fuel than is required )

Do not quote me on these duty cycles, it's been 7 years since played with the ECU in this open loop configuration ( where I monitored the OBD2 port via scanner )

Needless to say, for me as an experienced performance guy, I have thought about moving the car to Open loop error mode just to make the ECU more stable, the factory Daewoo ( GM ) ECU's don't like tuning all that much.

Most people make the smart move to a standalone engine management system.

[PrecisionBoost]

(2 cylinder ecm!) without a cam synch signal.

Your not correct with this statement, running batch injection does not make it a 2 cylinder ECM, many engines, including new ones will work in multi-injection modes that make use of batch injection.

That is to say some engines will run sequential injection 90% of the time and during wide open throttle they will revert to batch injection.

By running batch injection you can decrease the duty cycle and you can pre-fill the intake runner with fuel on the compression/power stroke( which cools the intake valve )

That is to say if you have small injectors and you are running 80% duty cycle in sequential mode at 90% throttle opening, you can switch to batch injection and run a 40% duty cycle ( since your injecting twice )


Hopefully you can understand the logic, sequential with small injectors gives more precise fuel control and thus better mileage without sacrificing the over all power of the engine.

The multi-mode ( sequential and batch ) injection allows the best of both worlds, it helps with fuel economy while helping with top end power due to the cooling effect of evaporating fuel sitting at the intake valve for a full revolution.

I don't know what percentage use multi-mode these days considering the large shift to direct injection, but i'd be willing to bet that at least half of the non-direct injection cars use this multi-mode injection technique.

[ls400x]

What you have described indicates you have a fundamental misunderstanding of duty cycle. This is the proportion of on time to off time. I think you might be confusing it with pulse width.

As far as my GM Daewoo ecm is concerned it is operating 2 cylinders considering there are only 2 injector drivers and 2 ignition drivers.

[PrecisionBoost]

sorry man, you seem to be clueless


Pulse width is fixed between two identical engines running the same air fuel ratio with two different injection modes ( batch vs sequential )

Pulse width is the TOTAL time an injector is on during a cycle ( intake, compression, power, exhaust ) which is 2 revolutions of the crank.

No matter which type of injection mode, the total pulse width is the same.


When it comes to injection mode the difference is the duty cycle.

In a sequential it would have a very long "wait" time where the injector may be sitting completely off during the compression/power stage then it will use a pulse width modulation circuit to create the total required pulse width.

In batch it has a wait time, then pulse width modulation of the injector, then a wait time then pulse width modulation of the injector.

Lets take a value of 3000 RPM as an example.

At that engine speed it takes 40ms for the engine to complete one cycle ( intake, comp, power, exhaust )

Lets say your engine requires 10ms of pulse width to inject the total amount of fuel required for that cycle.

In sequential injection the injector may be off for 20ms, then it will run a 66% duty cycle during the next 15mS to give a total injector on time of 10ms ( 10ms pulse width )

In batch injection the injector may be off for 5ms, then run 33% duty cycle for the next 15ms, then shut off for 5ms, then run 25% duty cycle for the last 33ms of the cycle.

The fact is sequential injection needs a higher duty cycle because it has a smaller injector "on" window.


When it comes right down to it, the only reason for sequential injection is to get better fuel economy and reduce pollution.


When not using a cam sync signal you have to keep your batch injection duty cycles identical

When using a cam sync signal you can then use a non-linear batch injection duty cycle.

That is to say if you do not have a cam sync you must run two duty cycles of 33% ( one just before intake stroke, one just before power stroke )

If you have a cam sync they can change it, they can run 50% duty cycle just before the intake stroke and 15% duty cycle during a portion of the exhaust/power stroke.

Decreasing duty cycle on the back side during exhaust/power maximizes the cooling effect by spacing the injection events further apart allowing the fuel to evaporate and reduce the intake valve pressure.

[7-even]

ok so i dont know about the duty cycle and all that jaz. i understand it but just the very basics to get back to the original question GSE i do have a set of adjustable cam gears that are supposed to run just like stock. the exhaust gear does have the part for the sensor on it. also i think its not a magnet, i would have to try to find an email to be sure. i could send you a pic or the set if you wanted something for mock up. i would even be willing to sell them if your interested. that would give me some money towards getting work done i already have the parts for. the only thing is that there are no degree marks on the gears you need to use a degree wheel to adjust them.

[ls400x]

ok so i dont know about the duty cycle and all that jaz. i understand it but just the very basics to get back to the original question GSE i do have a set of adjustable cam gears that are supposed to run just like stock. the exhaust gear does have the part for the sensor on it. also i think its not a magnet, i would have to try to find an email to be sure. i could send you a pic or the set if you wanted something for mock up. i would even be willing to sell them if your interested. that would give me some money towards getting work done i already have the parts for. the only thing is that there are no degree marks on the gears you need to use a degree wheel to adjust them.

It sounds like the cam sensor is just a sync to allow the ECM to know which half of the 720 crank degree cycle it is on, the ECM derives timing from the tooth wheel on the crank. I doubt the range of cam wheel adjustment that will be useful will affect the cam angle sensors role in the system.

[ls400x]

sorry man, you seem to be clueless


Pulse width is fixed between two identical engines running the same air fuel ratio with two different injection modes ( batch vs sequential ).

Agreed, conditional on there being one injection event per cycle and all other variables the same. You contradict yourself further down though.

Pulse width is the TOTAL time an injector is on during a cycle ( intake, compression, power, exhaust ) which is 2 revolutions of the crank.

No, it's the time duration of the injector circuit being closed. It is independent of being in a cycle.

No matter which type of injection mode, the total pulse width is the same.

Agreed, give or take depending on the number of injection events per cycle, injector dead time will have an effect

When it comes to injection mode the difference is the duty cycle.

In a sequential it would have a very long "wait" time where the injector may be sitting completely off during the compression/power stage then it will use a pulse width modulation circuit to create the total required pulse width.

In batch it has a wait time, then pulse width modulation of the injector, then a wait time then pulse width modulation of the injector.

This is where I think you are off. The total "wait" time as you call it is the same for batch of sequential. The only difference is where the "wait" occurs in the cycle. The injector is on continiously for one event per cycle in both batch and sequential. The duty cycle = [on duration (refered to as pulse width) / (off duration + on duration)] *100

There is no sub PWM going on, not that it could reduce duty cycle even if there was.

Lets take a value of 3000 RPM as an example.

At that engine speed it takes 40ms for the engine to complete one cycle ( intake, comp, power, exhaust )

Lets say your engine requires 10ms of pulse width to inject the total amount of fuel required for that cycle.

In sequential injection the injector may be off for 20ms, then it will run a 66% duty cycle during the next 15mS to give a total injector on time of 10ms ( 10ms pulse width )

Are you saying that the injector is going to be driven open and shut a number of times within the allocated 15ms in your example? if so at what frequency? How do you propose the required mass of fuel be introduced into the combustion process with this reduced total on-time? Higher rail pressure? Larger flow rate injectors?

In batch injection the injector may be off for 5ms, then run 33% duty cycle for the next 15ms, then shut off for 5ms, then run 25% duty cycle for the last 33ms of the cycle.

The fact is sequential injection needs a higher duty cycle because it has a smaller injector "on" window.

None of the above is true. Same thing as the previous comment. There is one continious on period and one continious off period per cycle (720 crank degrees) batch or sequential.

When it comes right down to it, the only reason for sequential injection is to get better fuel economy and reduce pollution.

Agreed. At high duty cycles both batch and sequential are spraying at the back of a closed inlet valve for the majority of the cycle anyway!

When not using a cam sync signal you have to keep your batch injection duty cycles identical

When using a cam sync signal you can then use a non-linear batch injection duty cycle.

That is to say if you do not have a cam sync you must run two duty cycles of 33% ( one just before intake stroke, one just before power stroke )

If you have a cam sync they can change it, they can run 50% duty cycle just before the intake stroke and 15% duty cycle during a portion of the exhaust/power stroke.

Decreasing duty cycle on the back side during exhaust/power maximizes the cooling effect by spacing the injection events further apart allowing the fuel to evaporate and reduce the intake valve pressure.