cbr cam sensor overheating?

[fsae racer]

For teams with the f4i's, are you guys using the stock caxm sensor? If so, is it a falling edge mag sensor positioned about 45 deg from inline with the cylinders at the exhaust side of head near cylinder #1? If still yes, have you had any problems with it working fine for about 10-15 mins? It seems our sensor just stops sending a signal when the oil get good and toasty. Any ideas/soluntions would be greatly appreciated.

[fsae racer]

For teams with the f4i's, are you guys using the stock caxm sensor? If so, is it a falling edge mag sensor positioned about 45 deg from inline with the cylinders at the exhaust side of head near cylinder #1? If still yes, have you had any problems with it working fine for about 10-15 mins? It seems our sensor just stops sending a signal when the oil get good and toasty. Any ideas/soluntions would be greatly appreciated.

[Colin]

those sensors are usally very reliable i'd be checking your ECU setup, which ECU are you using? and how do you know it's stops sending a signal? have you hooked it up to a scope?

[Garlic]

How hot is your oil? Have you done away with the oil cooler?

We have also had no issues with the F4i sensor. Seems very reliable.

[fsae racer]

We're actually using a 04 600rr with an old m4 pro (still serial, not can bus/parrallel), but i am fairly certain the sensors are the same as the f4i. we not only still have the stock water cooled oil cooler, but we are running a dry sump with large al resivior in the air stream, which is def. cooling the oil a good deal. we have no idea what the oil temp truly is, but the water is running 115-120. It took us awhile to get all the parameters for the map setup correctly, because we were previously running f3s. However, after switching back the wasted spark ignition, finding the correct crip #, and determining the types of mag sensors, I am fairly certain all the ecu parameters are correct. As I mentioned before, it runs perfectly for 10-15 mins, and I think the heat just kills it.

[Garlic]

I guess I'd ask what Colin did- how do you know the sensor is failing? What diagnosis are you going through once it quits running.

Are you sure the coils are wasted spark?

[mysticv6]

After the car heats up, we start to see "sync error" and "no sync" in the motec logs. It is also painfully obvious something is wrong, because the motor that was running so well, starts to break up badly.

The coils on the factory RR are individual, sequntial coils. Due to the igniter being built into the stock RR ECU, we decided to return to the setup we used with the F3's, where we knew the coils' and igniter's impedences were properly matched. So yes, we have 2 coils for 4 plugs, hence wasted spark.

[B Lewis @ PE Engine Management]

It wasn't clear to me from the previous post, are you running the stock sensors? If not are you running 3-wire Hall effect units? I have seen several occasions where that kind of sensor has trouble in an oil bath and will act up like you described.

If you would like to discuss the problem, give me a call at 513-777-5233

[Dan Deussen @ Weber Motor]

Here is an idea: If you see sync errors, it might be due to the cam chain stretching just a little under heat and load, causing the cam sync pulse to move where the ECU does not expect it.

Best way to check it, is by using an O-scope and monitoring crank signal, cam signal and maybe one injector signal simultaneously.

I can't promise this is your problem, but it might be worth looking at.

[Matt Ahl]

After replacing every sensor possible, rewiring the car and finally just putting the car out to pasture I solved a similar problem. Freak issue, but I would never wish it upon anyone else to deal with.

We had the same heat related issues and it turned out to be a faulty magneto in the charging system. It made good voltage but must have put out some weird interference.

[mysticv6]

Yes, we are running the stock sensor. It is a 2-wire sensor, that we believe is a magnetic type sensor. This was confirmed by a builder who runs Motec ECU's on the 600RR. The only modification done was on the trigger wheel, it has three teeth stock. The motec needs there to only be one tooth, so we cut off 2 out of the three, and left the tooth that is 180* out of phase with TDC, cylinder #1 compression stroke.

[B Lewis @ PE Engine Management]

Is it possible that the sensor is wired backwards compared to how the Motec is configured. I don't know why it would be temp dependant, but having it wired in reverse will produce erratic zero crossings in the signal.

[mysticv6]

I had the same idea that it could be wired backwards, and throwing the timing off by more at higher RPMS, but I checked, and it is wired correctly. I tried every possible mode in the motec for rising/falling edge on the ref and sync, and the engine runs best in this mode. It is also the correct mode we would expect for both sensors being falling edge.

[Trans Am]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by mysticv6:
Yes, we are running the stock sensor. It is a 2-wire sensor, that we believe is a magnetic type sensor. <HR></BLOCKQUOTE>

The sensor you have is a VR (variable reluctance) sensor. They are the stock sensor for most engines because they are passive type sensors and extremely durable because of their construction; meaning high heat (300deg C),much higher than a hall effect sensor. I've only had one problem with this type of sensor and it was because one of the leads was torn out of the sensor. It is good to note that the output of this sensor is dependant on tooth speed to produce an output resulting in a weaker siginal at idle.

A question for you: Can your ECU running Batch mode fuelling? If you are already running wasted spark then you don't need a cam sync for spark, only fuelling. Then if your crank encoder has a good sync siginal and you can run batch fuelling meaning you don't need a cam sensor... But if your having crank sensor errors your engine won't even run, or run like crap if it does and then eventually die.

-Trans Am

[mysticv6]

I would guess the ECU can do batch mode fueling... But thats not a road we want to travel down. Its hard enough to finish the endurance without running out of gas, but to go from a sequential fuel injection to a batch fuel injection set up because of a sensor, that doenst seem to be the most efficient solution.

We don't really have a problem with it giving us errors at idle, so I think its a strong enough signal. There should be plenty of tooth speed at 9000 rpm for it to send a good signal.

Another question to anyone else running an F4i/RR with a motec: There is a long list of choices for sensor type. I know that it is not Hall effect, and that it is magnetic. Right now its on standard magnetic, but would possibly any of the modes like "low level magnetic" or "high level magnetic" work better with this particular application?

[Nick McNaughton]

Check your ref/sync voltage settings, and how they increase or drift with rpm. We had a case where the signal coming from an inductive cam sensor was getting high enough to make the Motec trigger the sync position early, effectively retarding everything by a ref tooth. While it started and free revved to 10k fine, it behaved like a soft RPM limiter beyond that and spat out all sorts of ref/sync errors. Fixed it by changing the voltage setting in the motec to something that wouldn't cause it problems at higher RPM/pulse voltages.

This happened a year or so ago, and its been a while since I've played with this stuff so the details might be a little sketchy... but check it anyway.

[mysticv6]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Nick McNaughton:
Check your ref/sync voltage settings, and how they increase or drift with rpm. We had a case where the signal coming from an inductive cam sensor was getting high enough to make the Motec trigger the sync position early, effectively retarding everything by a ref tooth. While it started and free revved to 10k fine, it behaved like a soft RPM limiter beyond that and spat out all sorts of ref/sync errors. Fixed it by changing the voltage setting in the motec to something that wouldn't cause it problems at higher RPM/pulse voltages.

This happened a year or so ago, and its been a while since I've played with this stuff so the details might be a little sketchy... but check it anyway. <HR></BLOCKQUOTE>

That sounds like sometihng that could be really helpful. I'll defintly start playing with that tommorrow some. The ref sensor seems to be fine, but the sync is defintly throwing some errors, and it behaves like a limiter would almost in the way the motor breaks up.

I hooked a scope up to it today, and when I'm getting the sync errors at the Motec, the sensor is defintly still outputting a signal.

[Colin]

all these suggestions are very valid, i had a huge amount of trouble setting up these sensors when first started trying to use them and the only way you can be sure about the problem is by hooking up a scope to the sensor output otherwise your just guessing, it will also alow you to get the correct trigger level for each RPM point
good luck

[BryanH]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>

Another question to anyone else running an F4i/RR with a motec: There is a long list of choices for sensor type. I know that it is not Hall effect, and that it is magnetic. Right now its on standard magnetic, but would possibly any of the modes like "low level magnetic" or "high level magnetic" work better with this particular application? <HR></BLOCKQUOTE>

Have you set ref and sync to type 2?
Set both to type -1, then set up threshold ve / rpm maps, I always run them this way, works very well. I don't usually bother to measure sensor output if it starts and runs. For a 13k engine set up rpm points at 250, 500, 1000, 3000, 7000. and start with v.threshold. 250mv,500mv 1.2v,3v,15v for both sensors. I increase voltage at each rpm point until the signal at that rpm is suppressed,then reduce by 20% thereby measuring exactly what level the motec is happy with. Do this for both sensors.
I don't remember if the dos software shows sync % on the sensor diag screen, If it does make sure that it is stable and doesn't drop below 10%. (50% means that sync event occours exactly in between 2 ref events) Sync should be set 40-60% to fully utilize error suppression. You may have cut off wrong teeth and are getting sync crossing ref at 9000rpm which effectively swaps coils around revlimiting your little baby.
Someone is trying to build a "Bird shifter"
maybe you could do "monkey with bucket of petrol and a sponge" you'd pass tilt table without a fuel cap.

[mysticv6]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Halfast:
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>

Another question to anyone else running an F4i/RR with a motec: There is a long list of choices for sensor type. I know that it is not Hall effect, and that it is magnetic. Right now its on standard magnetic, but would possibly any of the modes like "low level magnetic" or "high level magnetic" work better with this particular application? <HR></BLOCKQUOTE>

Have you set ref and sync to type 2?
Set both to type -1, then set up threshold ve / rpm maps, I always run them this way, works very well. I don't usually bother to measure sensor output if it starts and runs. For a 13k engine set up rpm points at 250, 500, 1000, 3000, 7000. and start with v.threshold. 250mv,500mv 1.2v,3v,15v for both sensors. I increase voltage at each rpm point until the signal at that rpm is suppressed,then reduce by 20% thereby measuring exactly what level the motec is happy with. Do this for both sensors.
I don't remember if the dos software shows sync % on the sensor diag screen, If it does make sure that it is stable and doesn't drop below 10%. (50% means that sync event occours exactly in between 2 ref events) Sync should be set 40-60% to fully utilize error suppression. You may have cut off wrong teeth and are getting sync crossing ref at 9000rpm which effectively swaps coils around revlimiting your little baby.
Someone is trying to build a "Bird shifter"
maybe you could do "monkey with bucket of petrol and a sponge" you'd pass tilt table without a fuel cap. <HR></BLOCKQUOTE>

Yup, we have them both set to "2." Thanks for the help, that will definlty give me a good place to start on this. I knew you could do it custom, but I was figuring that would be a pretty big chore with the O-scop and the motor running for a while...

I do remember seeing sync position at idle. I want to say it was 52 or something thereabouts, so its plenty far from the cam teeth. I didnt know what that number meant though, so I've never checked it at higher RPM's. Thats defintly something worth taking a look at.

We tried swapping sensors last night, and I re-wired the connection to the motec. The problem didnt go away. So hopefully one of these software solutions will do it.

Thanks for all the help so far.

[mysticv6]

So we rechecked the crip #. While previously using the lock timing at 0 window, the crip showed 90; however, it now looks much more like 102 is the correct crip #. Never the less, it did not correct our problem. We then moved on to adjusting the rpm site locations for the preset voltages of the sync level setup. when lowering the rpm sites, the sync error popped sooner and the sync positions %'s climbed sooner as well. However, we did not have the opportunity to test moving the rpm sites to higher values (lower voltages for the same rpm) as our electric water pump decided to kick the bucket. When we get the new pump, we would like to test your advice further, halfast. However, if we can see the sync level rising, is there something else wrong? any more advice would be greatly appreciated.

[BryanH]

It is very important to check crip at 6 to 8000 rpm. You need to accuratly mark the flywheel at around 25 degrees to do this without frying the exhaust. I use a optilux or Bosch timing light and never use "wind back" timing lights any more, they do not read correctly at high rpm.
You are not alone, incorrect base timing is one of the most common problems seen.

Because you managed to change the symptoms, I think you are on the right course now.
If you have no luck with the threshold/rpm map try reversing the sync trigger polarity (wiring not software) I don't recall seeing sync % change with rpm, The obvious answer is mag signal upside down but the final solution may be more subtle.
Bryan Hester

[mysticv6]

We tried checking the timing...again... And determined that the most appropriate CRIP is a 105. This makes sense as there are 12 teeth on the ref sensor, so thats 30 degrees a piece. When we first checked the crip #, it appeared the sync event was exactly between 2 ref events plus 3 ref events of 30 degrees each BTDC compression of cylinder 1 [15+(30*3)=105]. No idea where 90 came from or why it seemed to work, but at the same time we're still not ruling it out.

The RPM point values you gave Halfast, where did you get those from? We had the best luck when we used those values, plus 20%. Any more or any less made the engine throw sync errors at a lower RPM. With the 20% rpm values for the sensor, and the 105 degree CRIP, it seems to run the best we've seen yet, but it defintly still throws sync errors anywhere in the 10k rpm range. It still seems like it gets worse as the motor runs longer / gets hotter.

The other strange problem that I think is still contributing to this is that the Sync position is not stable whatsoever. At idle, it will bounce around, but be about 65. As we move higher in RPM's, it changes drastically, until it gets into the high 80% range, and even 90%. After 90% it starts throwing the sync errors, as would be expected. Any ideas why it would bounce around so much?

[BryanH]

I'm an idiot. 2nd look at m4 software and I realized that ve levels fixed. Not like M800
Did you change ref to -1? if not, do it.
Try changing last rpm site to 15000rpm, one sensor at a time.
Is the locked timing retarding from idle to 8000rpm? ( paint mark flywheel at 20-30 deg. & change locked value until mark is easy to see)
If yes, then ref sensor is wired back to front.
If timing is not changing the sync is wired back to front. OR.......
email photo of cam trigger wheel where teeth have been removed
and a photo of wiring from sensors to ecu,

[mysticv6]

We hooked a scope up tonight to the sync and ref sensors. They are both defintly falling edge.

One thing that didnt make sense to me: When the motor is not spinning, the sync was reading like 1.8 v. I would hve expected it to have been zero.

Also, at 5000 rpm, the sync is already putting out 8 volts. The user defined table in the M4 will only let us go up to 4.5 volts. So, I tried adding a resistor in series with the signl. wWe tried 1800 ohms first, and when we hooked it up we got sync errors immediatle, and the car would not start. Does the voltage from the sensor need to be within a certain ranged?

[BryanH]

The voltage in the table is the minium voltage required to trigger an event at or above the defined rpm. The purpose is to stop small voltage spikes (noise) causing false triggers between trigger events. Go back to original rpm sites.
Constant 1.8v? reset your scope. If the sync input pin is really reading 1.8v please confirm. Sine wave? maybe sync is not grounded correctly and
picking up noise from flouro lights? but I can't see it running at all with poor ground.
would love to see images of sync trigger wheel and wiring.

[Philip]

The sensor is definately putting out about 1.8 V when the car isnt running....i just put a multimeter on it and the multimeter was reading a DCV of 1.7. It is definately a constant 1.7 V until the trigger wheel starts spinning and causing the voltage to go up and down as it should, but the wave is centered around 1.7 V, not 0V.

Another thing we noticed when we put the scope on it is that relative to the ref signal, the sync signal seems to take longer to switch voltage as the RPMs increase. From what i understand about these falling edge sensors, the voltage goes up and then it quickly turns negative, and this rapid change is what triggers a ref/sync event in the motec. Could the sync sensor's slow voltage switching be the reason the sync pos % increases with RPM?

[Frank]

remember the ECU "0V" is above the chassis "ground"

sync position changing with rpm = scary bad

[mysticv6]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Frank:
remember the ECU "0V" is above the chassis "ground"

sync position changing with rpm = scary bad <HR></BLOCKQUOTE>

Scary bad is the way we feel about it too...

I wouldnt think the ECU's digital ground would be 1.8 V though. Maybe slightly higher than 0, but a voltage that high doesnt make much sense at all.

So is lowering the voltage via a resistor a bad idea altogehter?

[mysticv6]

Halfast: Here's a picture of the sensor wheel. The larger tooth is obviously the one we'reusing, and the other is the tooth we ground off.

http://plaza.ufl.edu/wheelerc/one.jpg
http://plaza.ufl.edu/wheelerc/two.jpg

[BryanH]

good morning Mystic, the 1.8v is a problem that must be sorted, start by disconnecting all sensors, injectors, ignitor while watching scope (which you have connected to sync input and sync gnd). Keep looking until you find out where the 1.8v is coming from. Unlikely to be ECU fault, more like feedback from somewhere else in the loom btw chassis gnd and ecu gnd must be &lt; 1.5ohm
Bryan H

[BryanH]

Mystic, if I were doing it I would have made up a new trigger disc, larger dia concentric with a single tooth projecting and s.s. bolts to eliminate possibility of oppisite bolt causing trigger event.
I would be surprised if the short tooth / bolt lump didnt cause a trigger at high rpm.
It's the change in terrain under the sensor which changes the flux which generates the voltage. Thats why the wheel surface must be very smooth/polished

[fsae racer]

Since Colin doesn't want to own up to mistake, Ill do it for him. JK, mr. treasurer. Last year we were using Hall effect sensors on the cam. However, the 600rr motors we're using this year obviously have mag sensors. Colin had done everything correctly except, remember to wire the correct pins of the motec to switch from Hall effect to mag. Bryan, thanx for all the help. There was definitely some time there where we had no idea where to go, but your ideas kept us searching in the correct direction.

[mysticv6]

"It was never wired wrong, it just wasn't wired right"

http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Ian M]

We are having major sync % problems, so I wanted to bring this thread back to life. Overheating isn't the problem, but I am at a loss as to what is.
We were running two different RR engines on an engine stand hooked to a Motec M4 running stock crank and cam sensor. We have cut off the two teeth on the stock wheel that are close to each other and left the one 180 degrees on the other side. (This is how we ran our F4i the past 2 years).
Our data logging before break shows our sync position stable at 40% and the engine ran fine. From Data logging over spring break 3/16/07 it looks like we lost a steady signal. We swapped cam sensors to no avail. The engine would run, but it would run like crap. We tried switching engines and it was the same issue after we had already ran this issue with a steady sync signal as well. I know you are going to say that our wires got switched, but that is not the case.
I have hooked it up to a scope and our cam and crank signals are wired right now to both be falling edge, so the ref/sync mode is set to 14. The signals look good and don't look like they are overlapping. Ever since we lost this signal our engine tuning has gone downhill to the point where it won't even start on the same maps and setups that it was running on before.
I am beginning to think that the Motec is not converting the signal properly. Before I had both sensors set as standard ref mag and standard sync mag. After looking at the scope it looks like the sync could be set to low level mag, which I have tried to no avail. (I realize this could cause noise interference at high rpms, but I am just trying to get the darn thing started)
We are getting fuel and we are getting spark, but it seems like the Motec sends a spark whenever it wants to and it will poof a little and then quit, and then sometimes it won't even poof, and then sometimes it will start and run like crap.
Is there any possibility that a Bosch Ignitor could go bad?
I am about to think the Motec is fried or something because about the time we lost sync signal our lambda sensor started reading 1.00 all the time and quit logging. We replaced it and the new one did the same thing. Last night our engine water temp. sensor was reading from 69C to 90C. I know this seems normal, BUT THE ENGINE NEVER STARTED! I am going to check the sensor voltage on the signal wire today to see if the sensor is all over the place or if the Motec is freaking out.
I am getting extremely frustrated, so any help or common experiences would be much appreciated.

[Supra]

guys i have a question, in our cbr we are using the motec m4 and the sync position is -1% do you have any idea what could be the probelm