Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jimbob]

My cranksensor was replaced around 3 or 4 yrs ago, after infamously failing.

Been good as gold since - I originally ran it same way as original, but MDTM reran it the easy way just over a year ago whilst he was spitting in my pistons


Jimbob - just a thought, but your cable isn't flapping around at speed, causing the cable to prematurely go intermittent?

iirc it is loosely cabletied, so has limited movement


my one is prob a similar age to yours, this was bought a year ago last may

I'd like to think, as you run multiple V6s, you'd have a spare anyway, if the codes return

I keep one at the dealers down the road 

as I can get them next day, dont see a lot of point in carrying a £60 spare part

Whatever you're dealers EPC tells him, the 3.0 all take the cheaper one, so around £37

Your right, I remember now, they supplied the wrong one (invoice I found)  and swapped it and refunded 

I thought it was dear when I read the invoice! 

[Jimbob]

Plenty of short - medium runs today and no 19 so far 

[deviator]

Slightly off the wall suggestion here.

I am not sure but I haven't heard of anywhere as many CS problems with the vectra. Now this might be down to the in-line versus the inverse engine mounting. Or is it a different sensor?

Well I have it on reasonably good authority that the plugs are different (round and square). I don't know about the sensor. If the only difference is the plug, why not try a vectra sensor with an Omega plug soldered on.

Not sure if it would work or even if it would be any better, but someone may want to try it.

[Abiton]

I spent a bit of time today dissecting the crank sensor that I replaced the other week.  It was still working, as far as the ECU could tell (no fault codes, no EML), but the outer insulation of the cable was all but gone.

Having carried around a genuine spare for a year or so, I thought I'd fit that, and maybe mend the cable on the old one.  After looking a bit closer, this seemed impossible, so I thought I'd sacrifice it for 'learning'.  Towards the sensor end of the cable, not only had the outer insulation cracked and disintegrated, but the inner insulation on the 3 wires inside the screening was also in a disastrous state.  I'm amazed that it was still functioning TBH.

They're tough little beggars to hack to bits, I'll say that much.  Eventually uncovered the fact that it appears to be a completely passive sensor, no electronic components in there at all.  There seems to be a pair of coaxial coils right at the tip, wound on the inner and outer surfaces of what looks like a 'pot core'-type ferrite, with the open end facing into the crankcase.

I guess this must work as a sort of transformer, where the inner and outer coils are coupled more or less depending on the presence/absence of metal close to the tip of the sensor. 

I'm not sure that all this waffle really helps anyone, but maybe someone knows about the circuitry inside the ECU that energises and reads back these coils, to make a bit more sense of the thing???  If anyone has a duff ECU they want to sacrifice to the cause, I'll happily attempt to find out this stuff if no-one has prior knowledge. 

To sum up, the cable is the most likely thing to fail on this sensor.  Oh, and make damn sure it's pushed fully home when you fit a new one, 'cos the distance of the tip from whatever it 'sees' in the crankcase is surely critical.

[TheBoy]

it will be a hall type sensor.  It appears to be the cable, not the sensor itself, that fails.

[Abiton]

I certainly agree with your second sentence, but I'm inclined to disagree with the first.

Don't Hall Sensors tend to have slots through which metal bits pass?

This site seems to be quite useful:
http://www.aa1car.com/library/crank_sensors.htm
From which:
"The third type of crankshaft position sensor is a magnetic pickup that reads slots machined in a "reluctor" ring in the center of the crankshaft, on the harmonic balancer or flywheel. This setup is used on GM engines with the Direct Ignition Systems (DIS) on the 2.0L, 2.5L and 2.8L engines, and the Integrated Distributorless Ignition (IDI) on the 2.3L Quad 4, and also many Ford, Chrysler and import engines.

On the GM applications, the crank reluctor ring has six equally spaced slots 60 degrees apart. A seventh slot is spaced 10 degrees from one of the others so the crank sensor will generate an extra "sync-pulse" every revolution. The PCM then uses the information to calculate proper ignition and injector timing. This type of sensor must be carefully positioned so the air gap is within .050 in. of the crankshaft reluctor ring."

 :-/

[Kevin Wood]

Sounds like a variable reluctance sensor to me.

Typically, it'll either be a coil wound on a magnet or a pair of coils wound on a non-magnetic core.

In the former case, as teeth milled into the sensor disk pass the end of the magnetic core, some of the magnetic flux is dissipated by the teeth causing a collapse of the flux around the magnet and a corresponding voltage to be induced in the coil.

The latter type of sensor energise the core with an AC signal to one coil, and watch for fluctuations of the returning signal from the other coil. These have the advantage that they work right down to "DC" whereas the magnet type require a minimum cranking speed before you get a useable signal level from them.

Looking at the wiring diagram I'd say the crank sensor is the former. It only has a single coil (electrically, at least - maybe split into two halves physically).

The AC sensors are often used for cam sensors where you only have one tooth and it's moving at half engine speed, so low speed performance is more important.

Kevin

[Abiton]

Great stuff Kevin.

I honestly couldn't say for sure what the 'pot core' that has the two coils on is made from; I said ferrite 'cos that's what I expect coils to be wound on, and because it seemed harder and 'crunchier' than the other (plastic) materials used to pot the whole caboodle.  All three wires did seem to go down as far as the coil(s) though.

Do you (or anyone else) think it's possible that the wiring between ECU and connector may deteriorate in the same way as the sensor cable itself (haven't even looked on mine to see whether it's screened cable for that run or not)?

Alternatively, can you see anything in the circuitry that may be adversely affected by shorted wiring, and thus make a replacement sensor potentially not 'fix it'? (Would have thought it would be fairly nailed down, protection-wise  :-/)
Or generally any capacitors etc. nearby whose natural ageing may be a problem?

Just to mention; systems may not be identical between my ageing 2.0 and newer cars/V6 engines.  All I can remember about my crank sensor/replacement is that it was the more expensive of the two possibilities for the 2.0. 

[Kevin Wood]

Ahh, of course, it's a 2.0. I was looking at the V6 diagrams so it's possible that it's a different sensor type. Is the sensor core magnetic?

Indeed, if, as you say, you've got 3 wires inside the screening it looks more like AC type. I didn't pick that up the first time round.

I can't recall a failure of the loom wiring between ECU and connector, although anything is possible. However, the wiring in the loom is better protected, and not subjected to anything like as much heat and vibration.

I guess it's also possible for a short to damage the driver in the ECU but I would expect it to either work or not in this case, and I would expect the ECU to be adequately protected internally. :-/

Kevin

[Abiton]

...Is the sensor core magnetic?...

Dunno. I'll stick a magnet in the pile of bits/crumbs later and see whether anything sticks. 

[Kevin Wood]

I meant is the core itself a magnet? See if bits of the core stick to a pin, for example.

Kevin

[Abiton]

Nothing that will stick to a pin/needle.

The coil-former bits stick to a magnet though.

Unfortunately it seems that the sensor I have dismantled isn't the type that most of you have. 
Haynes (wiring diags) seems to show at least three different types, some with only two wires connected to the sensor's 'guts', and one that looks like it does have active electronic bits in.

[deviator]

Here is my (simplistic & probably incorrect) view....

Think of it as the oppersite of a speaker, where as the speaker takes volts to charge the coil and move the diaphram, the sensor is the other way around. The magnet on the crank spins and causes the ferrite core to move thus generating a pulse of a low volts/millvolts.

^I may be wrong but I think the basics are there.

As for taking the sensor apart, if it's normally a wire failure, can you get the solder tags inside without destroying it too much? IE could you chop the back off or drill holes? I think you can see where this is going.

[Jimbob]

Still no 19 reappeared.

Wonder if it was a truely phantom code  :-/

[dippydave]

Thought i'd add here that despite the Vauxhall info  that sensor part #90458251 is for 2.0 16V cars up to chassis 14194716, i've discovered mine (with number 128xxxxx) needed the later (and cheaper  ) part #9174621. I hope it lasts for yonks now!

now just to get me money back on the old one....