Cam/Crank Postion sensor

[rjwoods77]

What are the advantages and disadvantages of the a proximity and hall effect cam and crank sensor? Is there any weird quirks or traits they have. I am looking at using a motec cam gear mounted magnet with hall effect sensor for the cam and a toothed wheel on the crank with a proximity sensor. I was thinking of running dual hall effects to have greater similarity of sensors. Opinions anyone?

[rjwoods77]

What are the advantages and disadvantages of the a proximity and hall effect cam and crank sensor? Is there any weird quirks or traits they have. I am looking at using a motec cam gear mounted magnet with hall effect sensor for the cam and a toothed wheel on the crank with a proximity sensor. I was thinking of running dual hall effects to have greater similarity of sensors. Opinions anyone?

[Chris Boyden]

Hall effect sensors put out a voltage that doesn't vary in amplitude vs. frequency (rpm).
Variable reluctance sensors put out a 20 mV signal at low rpm and up to 30 V signal at higher rpm. (Depends on air gap as well)
I think that one of the biggest issues you need to address is the temperature rating of the sensor. Make sure that it will stand up to hot oil immersion. VR sensors come stock on an F4i or similar bike engines, but you're running something different. VR sensors require a toothed wheel made of a magnetic material....i.e. steel, no stainless or aluminum. The air gap with VR sensors is fairly touchy, but reliable if you get it right.

[BeaverGuy]

This past year we used both a magnetic proximity sensor and a true hall effect sensor as cam position sensors. The only real issue we had was the air gap from the sensor to the sensed object. This was an initial problem with both sensors. However, it affected the hall effect sensor more. When we first installed it, we adjusted it to sense the cam lobe and it worked reasonably well. We could start the motor and it would idle, which was what we were after. When we started to work with it again a few days later the sensor was recieving an erratic signal and had difficulty starting, we had to adjust the sensor to be closer.

We switched to a magnetic sensor because we thought the hall effect sensor had deteriorated and we couldn't get any more. We had a magnetic proximity sensor because that is what we use as our crank sensor and we knew it could stand up to the environment. It turned out that it wasn't the hall effect sensor but a grounding problem. On the '06 motor we are planning on using magnetic sensors in both spots because it simplifies the parts we need on hand and our crank sensor design requires a magnetic sensor.

My opinion is that you try to use the same sensor in both places if you can. Interchangability and only needing half the spare parts makes things easier.

[rjwoods77]

How fuel injection computers(MoTec namely) recognize these voltage inputs? Do you set a range in the software were it reads? Do the injection computers reference a proximity sensor for rpm as well as position?

[Jarrod]

MoTec is user definable on almost everything, you just pick form a list what type of sensor it is, and you can either use standard calibrations and settings or set up your owm. We are running the standard variable reluctance sensor (new trigger wheel though) for the crank, with a honeywell gt101 hall effect picking up one of the bolts on the camshaft. (replaced the other with a stainless bolt). We just checked the signals with a CRO, set the sensor voltage levels for the crank sensor at various speeds and away we went. Never had a problem.

[rjwoods77]

I guess my real question was what is the advantage of a vr sensor variable voltage output(sine wave) versus a square wave output of the hall effect that is not variable. Is the crank sensor usually a vr sensor that is used for position and rpm pickup? What other options does motec have for reading rpm? I assume the vr sensor on the crank is the way to go because it kills 2 birds with one stone but i would like to know my options before i go that route.

[BeaverGuy]

I'm not sure about the Motec, but Haltech systems require the Sine wave of a VR sensor be converted to a square wave before the ECU can use it. This is built into the newer Haltech units and is probably similar on the Motec.

The crank sensor regardless of type is used to indicate crank position and RPM. With a magnetic sensor and magnets at the crank, if you are running some sort of non-sequential injection it is done with magnets of two different polarties, otherwise the required cam sensor provides the position information and only one magnet polarity is neccesary. With a hall efect or VR sensor it is done with a missing tooth.

[Colin]

Reluctance sensors' can be a pain to set up, like Jarrod said you have to set it up with a CRO both to check polarity (most don't produce a nice symmetrical sine wave) and to set the trigger levels, that said the reason they are used as a stock item on the engines is they are more reliable and have a longer life in harsh conditions, ie inside hot engine oil. So Rob the first thing you need to decide is if you want to run a fully sequential system ie not a wasted spark. I'd assume you would so you will need a cam sensor. The next thing I'd look at is what stock sensors you have if it's like an F4i that already has both sensor you'd be crazy to change them but if you have to add the sensor a good hall effect like the Honeywell gt101 will give you less headaches during setup than a reluctor will.

[Frank]

honeywell GT101, sooooo easy

[Ben Beacock]

Rob: I wrote a paper for an electrical devices course that compared hall-effect vs VR sensors for automotive applications. I'll see if I can dig it up.

edit: looks like i'll have to go back into my *ugh* paper archives.

Anyways.. there are a few things to consider: resolution, mechanical setup, temperature rating, and cost.

VR sensors have trouble reading at low speeds and the voltage output of the sine wave decrease the slower you go. You can add more teeth to your wheel the alleviate this, but you may then have problems timing signals at high RPM.
Hall effect sensors will read as slow as you want (up to the overflow limit of your timing code), so you can run fewer teeth and still have a stable signal at low rpm.

because of the sine wave signal, VR sensors need additional circuitry. I've used the National LM1905? Variable Reluctance Sense Amplifier with good results. It filters the signal well and has hysteresis built in.
Hall effect sensors have the circuitry on the sensor itself. This is where the reliability and temperature problems come in. Granted, the honeywell GT101 is robust and can take the temps, but I don't know the long term reliability.

Mechanical setup: typically the VR sensors run pointed straight at a toothed steel wheel. Hall effect sensors can be arranged such that there is a magnet on one side, and the sensor on the other. A steel 'gate' wheel would then pass between them and interrupt the magnetic field. The honeywell GT101 is different, and works more like a VR sensor. (advantage GT101)

cost.. VR sensors are generally cheaper. but whats a few bucks on a prototype anyways?

I beleive VR sensors can also introduce a fair amount of EMI. a 30V unfiltered sine wave running all the way to the ecu is noisy compared to a 5V digital signal.

My conclusions: If your engine is already set up with VR sensors then use them, since they should be more reliable and configured properly. But if you need to add sensors, then go with the honeywell hall-effect.

[rjwoods77]

Ben,

I am modding a normally carburated engine so there is no stock sensors on it that I can use. Everything I would be doing would be fromt he ground up. Thanks for the input though. When you find that report send it to bobbyfriction@hotmail.com.