[Codemonkey]

Does anyone know what the crank position sensor on the R125 actually senses ? I understand its a hall effect sensor and understand all that, what I mean is, is it detecting something on the end of the crankshaft itself rotating at (obviously) the same speed as the crankshaft, or is it something on a geared down shaft that rotates once every 2 crankshaft rotations ?

What I'm trying to understand is how the ECU works out the position of the crankshaft in the combustion cycle. If its sensing from the crankshaft itself (2 rotations per combustion cycle), the ECU must have some way to figure out if its on for example a compression stroke or an exhaust stroke so it can decide when to open the injector or create the spark unless it just creates a spark every rotation (wasted spark) and opens the injector regardless of the inlet valve being closed. If its just sensing from some other half crank speed shaft then its a bit more obvious and no further explanation required!

Anyone know for sure ?

 

[Killachipmunk]

I imagine they can have a very simple SR flip-flop or such to skip 1 rotation of the crank, so it's entirely possible that it's on the crank itself.
I seem to remember seeing a picture of the sensor, but can't for the life of me remember where it sat, it was next to the cam chain somewhere.
I could be wrong, but I'm pretty sure it was on the end of the camshaft.

 

[Codemonkey]

I imagine they can have a very simple SR flip-flop or such to skip 1 rotation of the crank, so it's entirely possible that it's on the crank itself.
I seem to remember seeing a picture of the sensor, but can't for the life of me remember where it sat, it was next to the cam chain somewhere.

That would be pointless. The moment you disconnect the battery or if you swapped ECU or engine, they could end up out of sync. On a bike with a "missing tooth" type sense wheel, it would maybe be possible to sense the slowing of the rotation on the compression cycle while starting the engine allowing it to work things out, but despite looking through the service manual for the bike, I can't see one in the R125.

The sensor location is shown in the service manual but its not clear what its actually sensing.

 

[Benji]

The sensor is located on the left hand side of the engine (looking from when you are sat on it) and senses direct from the crankshaft no gearing down. Workshop manual the image on 5-29 shows it. It is a wasted spark, and because it is such a small amount of fuel, it's not a problem to fire the injector, as it has another useful side effect in that it helps cool the valves from the fuel being injected against the back of them, and also brings down the temperature of the inlet tract.

Yes if we had a camshaft sensor then it could be done so it only fires the once, but it's more money on sensors and requires more complicated processing as well. The other thing of note, is that the crankshaft sensor missing tooth is not at top dead centre (tdc), as the ecu needs time to be able to make its fuel/ignition calculations and still be able to get back from processing this in time to be able to fire the injector for the amount of time required. It's an interesting piece of interrupt processing.

You are right as well Code about the issue with swapping ecu's it would just be a nightmare, this is actually an issue with Piezo injectors as they take a pulse to open them, and a pulse to close them, so you want to make sure you know what state they are before you start swapping injectors about. Interesting to watch on an oscilloscope. The other interesting thing you can see even with a normal injector as on the R125, as the revs rise you can see the timing changing. On Diesels you can see a pre injection, and on some of the direct injected you can see multiple pulses to create the swirl.

 

[Codemonkey]

The sensor is located on the left hand side of the engine (looking from when you are sat on it) and senses direct from the crankshaft no gearing down. Workshop manual the image on 5-29 shows it. It is a wasted spark, and because it is such a small amount of fuel, it's not a problem to fire the injector, as it has another useful side effect in that it helps cool the valves from the fuel being injected against the back of them, and also brings down the temperature of the inlet tract.

Yes if we had a camshaft sensor then it could be done so it only fires the once, but it's more money on sensors and requires more complicated processing as well. The other thing of note, is that the crankshaft sensor missing tooth is not at top dead centre (tdc), as the ecu needs time to be able to make its fuel/ignition calculations and still be able to get back from processing this in time to be able to fire the injector for the amount of time required. It's an interesting piece of interrupt processing.

You are right as well Code about the issue with swapping ecu's it would just be a nightmare, this is actually an issue with Piezo injectors as they take a pulse to open them, and a pulse to close them, so you want to make sure you know what state they are before you start swapping injectors about. Interesting to watch on an oscilloscope. The other interesting thing you can see even with a normal injector as on the R125, as the revs rise you can see the timing changing. On Diesels you can see a pre injection, and on some of the direct injected you can see multiple pulses to create the swirl.

Cheers Benji. I knew some systems cooled the valves this way and know about the sensor being before TDC. I just wondered if there was something I was missing. I don't suppose you know if there is more than 1 pulse from the sensor on the R125 (ie like a missing tooth wheel), or is it just the one event per rotation ?

 

[Benji]

I haven't scoped that i'm afraid so don't know the exact answer, nor have i had that part apart. It was on my list of things to do. It would be unusual to do a one event pulse on a crankshaft, but it is common on a camshaft.

 

[Codemonkey]

I haven't scoped that i'm afraid so don't know the exact answer, nor have i had that part apart. It was on my list of things to do. It would be unusual to do a one event pulse on a crankshaft, but it is common on a camshaft.

Guess I'll just have to get the scope out when the weather's a bit warmer! Cheers

 

[Killachipmunk]

What you wondering for anyway Code?
I don't suppose you're still trying to remove the timing retard?

 

[Benji]

Guess I'll just have to get the scope out when the weather's a bit warmer! Cheers

If you do, can you capture the trace and put it up here, I assume you have digital capture scope.

 

[Codemonkey]

Nope, not trying to remove the timing retard as such, actually thinking of building my own ECU from scratch. Always fancied designing one and a single cylinder fuel injected engine is a good place to start. Still need to look into the ignition part since I don't know much about capacitor discharge ignition (CDI). The rest of the sensors (manifold intake pressure, air temperature, coolant temperature, throttle position) are all pretty straight forward as is the idle air valve actuator.

Yes Benji, I have a digital storage scope, so if/when I probe it, I'll capture the plots and post them up. Don't hold your breath though, could be a while until I get around to any of this and any of the following could happen:

1) I simply can't be bothered
2) I sell the bike for something bigger therefore don't have a test mule to develop it on
3) I just get bored
4) We end up with a remappable ECU with programmable rev limiter!

I've already found a source for the same connector used in the Moric ECU so it should be possible to build a drop in replacement.

 

[Benji]

Have you looked at the microsquirt ecu, that is open source, and available as a kit to build.

 

[Codemonkey]

Have you looked at the microsquirt ecu, that is open source, and available as a kit to build.

Yes, seen that and read up on it. I don't think we'd be able to use it as a drop in replacement though and besides, where's the fun in taking one someone's already done ? Building one from a kit is just plain boring. I work in R&D. I'm doing the research bit, it just remains to be seen if I can be bothered or get the time to do the development bit ;-)

 

[Codemonkey]

Still need to look into the ignition part since I don't know much about capacitor discharge ignition (CDI).

Well, it seems I shouldn't have bothered reading up on CDI cos from looking at photo's of a disassembled Moric ECU, it doesn't use it anyway, its just a standard ignition system. Not sure why the ECU gets referred to as a CDI so frequently cos it definitely isn't!

Anyone wondering what i'm ranting about:

Standard ignition uses a high inductance coil whereby a magnetic field is built up by feeding it with a 12V supply. When the supply is removed, the field is allowed to collapse and this causes a current to flow in the windings of the coil which due to the number of turns, creates a very high voltage (>30KV) for the spark. The downside to this system is that it takes a while to create the magnetic field (this is the dwell time) due to the high inductance of the coil. This long dwell time means you have to start charging the field in the coil before you want to create the spark, and if the revs are very high it can become a limiting factor.

CDI uses a small inverter circuit to charge a capacitor to about 400V (a bit like the flash circuit in a camera) which is then in turn discharged through a small low inductance pulse transformer which steps the voltage up to >40KV. The benefit of this system is that the capacitor can be charged very quickly, thereby not becoming a limiting factor in RPM. Its so fast in fact that in some systems it can generate multiple sparks.

 

[Codemonkey]

For anyone that was interested in this, it looks like it uses a variable reluctance sensor and senses from the magneto rotor. If anyone's got some decent detailed photo's of the rotor itself, especially showing the side on view I'd be interested in seeing them!