x20xev crankshaft sensor problems

[vexorg]

You can do a comparison, not a proper test, but if you had a good loom/ecu then you can measure each pin to ground, and to each other.

But without a good loom/ecu then you'll not know what the reading should be.

[ilovedmymantas]

My apologies guys.
Taking a step back from being overloaded, re-reading this post helped me understand the method of your teachings a little more, I think

Finally got a dry day to test the sensor. There was a bit of a hiccup with the new test leads, i thought the red one was faulty but it turned out to be the plastic sheath going to the meter was a bit too long, cut both of them back by 1/4" and all was good.

The plug is parallel with the fuel rail and is a pig to get out. When I finally managed to get a finger on the right hand side metal clip it came off when i removed the plug, don't know how i'm getting that back on . It's situated directly under the highlighted area.



Only the sensor has been unplugged and tested so far, the (good) battery's inside on trickle charge.

Pin 3 + 2, ground/power as expected 0 ohms


Pin 3 +1, ground/signal


have no idea what the ground/ signal reading should be for this sensor but am I correct in thinking it's a little light?

From the link I posted this apparently should be between 700 and 3000 ohms

For the sake of it I tested pin 1 + 2, signal/power

I changed the meter setting (not sure why) Is that 10 ohms?

Do you think the readings show I have a faulty sensor?

[vexorg]

It wont be 10 ohm, just below what the meter can meater accurately. Anywhere between 20 and 0 ohns would probably give similar.

[Robsey]

As it is showing 0.01, I would try turning the dial to 2k and see if that makes a more accurate reading.

If it still shows 0.01, then try 200 ohms.

I would expect the instrument itself to register a couple of ohms across the probes... a tad more if your batteries are tired.

[ilovedmymantas]

Weather and painkillers permitting I'll re test that tomorrow at 200ohms and stick the battery in to do a couple of DC checks.

The four 'how to test...' sites i've focussed on say the sensor resistance reading should be anywhere between 200-3000 ohms. Since mine is showing 11.3 can i assume the new sensor is faulty?

-Just found the cambiare troubleshooting page. It says the resistance should be between 200 and 900ohms, dependent on rubbing a metal object over the top of the sensor. I can't do that because it's on the car but would I be right in thinking with no mechanical input it should still be a minimum of 200ohms?

Should I buy a different brand one online ( though I suspect they all come from the same overseas manufacturer ) to test and compare? I don't want to disturb old wiring unless I have to.

[vexorg]

Is it the right one, because there were some changes made to them that needed connectors cut off.

Then for the v6, I know the omega crank sensor looks identical, a few digits off in the part number and it gave an ecu error (code 19), but cav/calibra v6 sensor gave no error.

[ilovedmymantas]

Hi David,
I think this engine only needed the wires cut for a camshaft sensor modification?

i'm 99% sure it's the right one. There are two for the xev in the cambiare catalogue https://cambiare-ve.com/catalogue/?sear ... &Year=1995
The supplied sensor is the bottom one, it has a 'spanner' back like the original siemens unit, the others are flat.
The look and numbers also tie in with no. 10 in the now restricted 7zap https://opel.7zap.com/en/car/j89/p/1/82-3/#10

I've ordered a 'kerr nelson' brand ckp that's due to arrive tomorrow, i hope testing that for comparison will confirm it's the cambiare unit at fault and i don't have to mess with the wiring.

* Does anyone know where the reluctor ring is on this engine? It seems to be the only sensor without an 'O' ring.

[vexorg]

The ring is on the crank, will be in line with where the sensor is.

[ilovedmymantas]

My inability to understand this stuff is frustrating. The new sensor arrived today and I still can't get anywhere near a 200ohm lower resistance. I started on the 2k range expecting more than 200 when rubbing metal across it but nothing changed

As it is showing 0.01, I would try turning the dial to 2k and see if that makes a more accurate reading.

If it still shows 0.01, then try 200 ohms.

I would expect the instrument itself to register a couple of ohms across the probes... a tad more if your batteries are tired.

Today's readings were done on the 200 ohm setting and aren't much different from the cambiare one, surely i should have one reading in the hundreds?

Probe to probe - 00.3
Ground + power - 00.7
Ground + signal - 8.00
Signal + power - 7.7

[thomas]

Status report needed. Is it running? If so, what are the symptoms, any error codes?

Thought the thinking was the 92 was routine due to failing to start, and the 93 the same?
The only way to get a clean 56.1 wiring-diagram and key it seems is to haunt 2nd-hand book-stalls and jumble-sales in Australia?
Are the diagrams correct? Are we barking up the wrong tree?

The diagram (56.1) does not seem to use the same symbol/connectors for the crank-sensor here as for the c18nz/c20ne etc.
The cam-sensor is the same symbol, both still though presumed simple inductors, not sure about 56.5.
Excited by the reluctor, a screw nail anything iron passing it, it'll generate a tiny little ac signal, a dirty spiky, sinusoidal wave.
All by itself it can do this, it needs no power applied.
Don't rub or strike it, keep magnets away. The little coil inside, its wires are hair thin, their attachment fragile.

The power applied is usually a DC offset (sometimes AC, the combined signal separated later) that the fluctuating ac wave generated in the sensor winding rides on, induced both on the signal out wire and on the supply.
The DC part is discarded and you have the little zero-crossing ac wave again. One half of it, the negative usually is kept, the other discarded.
What's left is further conditioned till it resembles a square-wave the computer can count.
The faster the thing moving goes, the reluctor, sprocket teeth, move towards and past it, the spikes get closer together and their voltage peaks and troughs steeper.

The inductor coil's DC resistance doesn't change, other than through normal wire resistance increase due to temperature rise and vice-versa.
The resistance will appear to fluctuate a few 1/100ths or 10ths of an ohm as it is excited, but it is just the induced voltage fooling your meter.
This side-effect is useful, but the induced voltage at cranking speed (faster means more frequent 36 minus 1 for the missing tooth, per revolution, and bigger higher spikes) might be just 0.2-0.3 V AC, hardly measurable, with vibration, the variability of temporary connections, electrical noise, but after its (the sensor's) coil-resistance is a useful if crude and fickle test.
More certain done with sensor on the bench than in place.

A scope needed to visualise tiny and short-lived signal forms.
A logic tester that as well as indicating + or - polarity, also detects ac-like signal forms/pulses might be sensitive enough.
A scope because watching it dance, soar and fall as you wave things past an inductive sensor, is unambiguous, unmistakable, unforgettable.

Don't test wire resistance/continuity with the ecu connected.
Your meter could apply up to 1V, about double the forward voltage of most semi-conductor devices.
Measuring voltages, your meter's very high input impedance is of the order of 10megohms and is fine, safe.

It is a computer, there are few things I would use a multi-meter with on a pc or laptop.
I would test what you can, as you have with battery removed -a wandering earth-lead has a springy attraction for the battery post.
Unplug the ecu plug once familiar with the pin-numbering, while testing the loom.

Can see too that the ecu or plug, hasn't got wet, corroded.
It carries some risks, of actually breaking a wire or a connector pin.
Just observe as you replug it that plug pins or wires are fully home, push each one in with something blunt, e.g. wooden cocktail-stick, every one, see that none are simply being pushed out/back.
The important things here are that pin x on sensor gets (only) to pin y on ecu.

I don't have the foggiest what you might expect to find -signal input lines, voltage + or - at the end of any of the leads leading from ecu/loom to/from sensors. Or about the official modification to the cam-sensor plug. Any printed matter?
What was that about, to use a different sensor than originally used?

Too many unknowns with this to advise any power on, ignition on, or engine running testing, it just has so much potential for crossed over probes or clips, of connecting supply, or an earth where it can do harm etc., for things to go from bad to worse.

[vexorg]

The resistance doesn't change when metal moves in front of it.
It will generate a voltage, a pulsed voltage, so that will potentially mess up the ohms readings.

The best option is to fit the sensor, connect an oscilloscope type device, and then measure while cranking.

I think it's kind of gone beyond what yo can do with a meter now, assuming you have buzzed out all the wires back to the ecu.

[thomas]

How are you getting on with this Matt?

Diagrams from the previous page are gone with that hosting.
I was too busy with the subframe for quixotic battle with 'award-winning' (means offered the most freebies/bungs) Minnow-tech.
Hope they were without error, thought correct, derived from the original scrappy postimage diagram for 56.1.
Nothing new. Can make any available, at other url but will have to whitelist your ip address in advance.

If you have a new sensor that you think is the correct part and good.
With your meter on 2K resistance scale measure, note:

between 1 and 2
between 3 and 2
between 1 and 3

Join 1 and 3 together, measure between those joined two and pin 2

Join 1 and 2 together, measure between those joined two and pin 3

lastly

join 3 and 2 together measure between those joined two, and pin 1

Just curiosity more than anything.

[ilovedmymantas]

Still frustrated with this.

Had a good weather day Saturday before last so put the fully charged battery in and did some voltage testing, with the sensor plugged in and unplugged (also with the kerr nelson replacement plugged in but not bolted to the engine). I've since realised I should have taken far more exact notes, most of it's a jumble.

I was worried about unplugging and testing the ecu wires in the footwell due to a left shoulder problem, haven't done that yet. That's been overshadowed by my lower back being tender on that Sat after testing and escalating on sunday to a trapped nerve, haven't been out until today, yesterday was the first day I could put a pair of socks on. Yay!
End of moan.
-----------------------------
Testing with battery in and fuel pump fuse pulled for cranking:

power feed to sensor wire, ign off - 0.00v (20v setting)

power feed to sensor wire, ign on

This drops to 2.19v when cranking.

Earth and signal wire ign off - .003 (2v setting)

Earth and signal wire, cranking
https://youtube.com/shorts/pWyfdPqvBZQ

Sites i've looked at seem to suggest the power wire should be either 5v, 7v or 12v. Mine seems to read half that, could this be the problem?

Out of interest, does tech 1 show any of the wiring faults or is it as basic as the paperclip test?

[vexorg]

tech1 can show live data too, but not much help if it's not running. At most you'll get code 31 or 19 (or 0340 or 0335 ... whatever the 4 digit code is).

[ilovedmymantas]

tech1 can show live data too, but not much help if it's not running. At most you'll get code 31 or 19 (or 0340 or 0335 ... whatever the 4 digit code is).

The reason I asked is i can't find any data for what these readings should be. If the engine was running would tech1 identify which wire was at fault e.g. feed voltage too low causing incorrect signal? If i can't solve this my last resort is the garage asking for a deposit/blank cheque.

If my back eases up my plan this week is to reconnect everything and retest feed & signal wires (also alternator output) with the engine running.
- It will run on limp mode but i don't really want to, I just don't like the hard starting on cam sensor only or the battery draining 15s starting time, apart from that mot's due January

[vexorg]

No, the tech1 couldn't identify the wire, just tell you the fault and possible the live engine speed might show something off, but not as detailed as identifying wires.

I'm not sure many garages would want to look at it, potentially charging as much as the car means they'll know you might just say no.
Where are you based? maybe someone close can help.

[ilovedmymantas]

I thought I might be grasping at straws re tech1 and wiring. I'm in West Lothian and have tried all the Lothian's auto electrician, no one is interested.

The sites i've looked at say the data is available in the workshop manual. I assume they mean vauxhall rather than Haynes but I can't find this data anywhere.

[vexorg]

You're not that far from me, I'm in ayrshire, just south of glasgow. The problem is finding time right now, between work and getting stuff done in the house, and for kids and family, doesn't leave a lot of free time.

[ilovedmymantas]

I can appreciate that, it's hectic at this time of year.

[ilovedmymantas]

I got out to the car today to put the stuff i'd undone back together - my back finally felt mobile enough and it was a mild day.

The battery had been off trickle charge for three days when fitted, to rule out the alternator first reading was with ignition off.


After cranking with more time and noise than i'm comfortable with (it was probably 7-8 seconds but seemed like an eternity) the engine started, albeit in limp mode. Alternator seems to be ok.


The last thing I checked was the voltage feed to the crank sensor. This dropped to 1.1v randomly three or four times to start with then settled


There is only 0.04v difference between ign on and engine running. From my (very) limited knowledge i don't think this is a problem but i still don't know what the voltage should be. How do i find this information? I'm a decent searcher but this has me stumped

After warming up the engine for around ten minutes i stopped it, left for a minute, then restarted.
This time it was totally different, It took slightly longer and there was a lot of farting. I barely knew it had started because it seemed to be running smoothly and quietly at 200revs
When it built up to normal revs there was a rough sounding noise i couldn't pinpoint, sounded like it came from somewhere below the alternator but who knows? It'll probably have disappeared next start-up

It's got to the stage that friends and relatives are telling me about cars for sale, arrgh!
I'm not good with change, i'd rather put that money into fixing my cav, i just can't find anyone willing to do it

[ilovedmymantas]

Good news! Got the car back just before christmas.
as previously mentioned a hefty deposit (£500) was required to investigate the problem. I'm still no wiser as to the cause but it's running ok now, fitted with the kerr-nelson crankshaft sensor I supplied instead of the cambiare. There was also a bad fuel leak, this cost £300 in total to resolve. No matter, I'm back on the road now
I've left the surplus £200 with the garage, the mot's beginning of Feb and i'm getting my gates timing belt kit and gm water pump fitted (£150). I've ordered a gm oil filter, just trying to decide on an oil, fully synthetic oil or ester.

[Robsey]

I spoke to my colleague who used to be a Vauxhall mechanic in the nineties.

He says that the design of the XEV with three identical connectors was very often a cause for running issues.
Being plugged in to the wrong sockets.

It used to drive the mechanics nuts.
However they always had one mechanic who knew how to sort the running issues.

It is good to know that your garage were able to get to the bottom of your issue.

[vexorg]

And why the xev was held in such low regard after the c20xe

[ilovedmymantas]

And why the xev was held in such low regard after the c20xe

I thought it was because of more electronics, little ability to increase power and worst of all being an interference engine - i'd be happier without that feature! Apart from that I've been happy enough with the XEV for over twenty years regarding emissions.

[Robsey]

There are definitely more sensors on the XEV to tighten up emissions, hence the X prefix engine code.

There is just one sensor (crank angle sensor) on the SPi NZ engines and MPi 8 valve NE and SEH engines.

The XE kept things quite simple, but has the 16 valve interference engine block.
Liked for it's moderately simple but raw 150 bhp engine.

Sadly the XEV lost about 14 ponies compared to the XE, but if well maintained and not messed with, it is a very capable and reliable unit.
Unfortunately more sensors = more to go wrong.

As with any Vauxhall, where possible original Bosch or Pierburg parts are essential where still available.