Okay, so I recently had a thread (Linky), where I had a no-crank condition with no interior / trunk / headlights / puddle lights / no wipers.

Central locking system is still inoperable (door locks / rear trunk lid), but I'd rather not focus on that now.

Starter / engine now cranks, but engine doesn't fire up. I notice the tach shows zero RPM during a crank, which the typical answer is, "bad CKPS". I replaced it, I think, two years ago (maybe 2-3k miles) - do they go out that often? Do I have to spend $110 USD every couple of years for this part?

Anyway, before I go blaming the CKPS, I took the horn relay (under-hood, left fuse panel) and swapped it with the ignition relay (under-hood, right fuse panel) - still zero RPM on crank. So, one at a time, swapped out the two ignition relays behind the kick-panels under the rear-seat. Still zero RPM on tach.

I found an article at jimbutterworth.co.uk where he bench-tests the CKPS - has anyone confirmed these two tests? He states 1,300 OHMs resistance and inductance of 1 Henry. The other test is to wave a bladed screwdriver in front of the CKPS's 'pip' - he shows a reading of 250mV to 300mV.

My other understanding is there should be 12V to the CKPS when the key is on, but I have to ask, is that 12V with the key ON, or 12V when the key is at cranking position?

 

There is NO voltage TO the crank "sensor". It is a magnetic electrical pulse generator. It doesn't really "sense" anything other than the metal passing close to it, which makes it pulse. The faster the metal passes by it, the stronger the pulse generated. It does generate a AC signal.
I don't know about bench testing. I do know that my old one that was bad had the same ohm resistance as the new one. This tells me that measuring resistance isn't a sure thing. Not sure about the mV either. Not even sure what 1 Henry is.
I will say that there is a great possibility that new ones are maybe not of the same grade quality as the originals. I too, hope I don't have to replace mine every couple of years as well....

 

Good to know how this sensor actually works - thanks!

I thought it odd that 12V would go to the sensor and it exhibit OHM output, seeing as there are only two wires (on mine anyway, some have three blades). I mean, the main CPU must be reading "bad values" from this sensor to portray the zero-RPM condition - there must be some variance of values arriving at the CPU.

Sad we can't bench test these - you'd think you could. I guess the only bench test is the "no RPM reading on tach on engine crank". I wonder if wrapping the CKPS in some sort of heat shield material would extend the life of this thing. I'm sure the engine heat places a toll on the sensor. But I'm a bit confused because, although it's been about two years, I've only added about 3,000 miles since I last replaced the CKPS.

I may still have my old CKPS laying around - I should look for it and see if I can bench test the two of them, plus a new one. Maybe make a slotted wheel (like the ring on the crank) and attach it to my Dremel and run it at the slow speed and see if I get any reading variances.

 

What the CKPS is is a variable reluctance sensor. That means it is a magnet with an electrical coil wrapped around it. As a piece of magnetic material cuts the magnetic flux from the magnet, the field is disturbed and an electrical emf is generated in the coil. A Henry is a unit of inductance, which is a characteristic of the coil. Resistance represents the DC impedance and Inductance is the AC impedance.

So, if the electrical characteristics are confirmed correct by a bench measurement, that would imply that the magnet or the magnetic circuit (the surrounding steel) has deteriorated. To actually test the output, I would say you would need a setup with a steel toothed gear passing teeth at a calibrated distance from the sensor, at a calibrated speed. Then, you would only get meaningful data by testing one sensor vs another. For the test to be meaningful, you would load the sensor with the same input impedance as the ECU, which is available in the schematics, then monitor both the peak voltage AND the waveform (using an oscilloscope).

Even with all the test equipment available, I would say it is far simpler to keep a known good sensor and compare it against the suspect running the engine.

 

You can test the Crank Sensor in situ but you will need a scope to do it. You can get a USB scope to attach to a laptop quite cheaply these days and it's a very useful diagnostic tool for all sorts of things.

I have one of these and it works well - not in the same league as a Picoscope but for a tenth of the price it does the job I want.

Hantek 1008C PC USB 8CH Automotive Diagnostic Oscilloscope DAQ Program Generator | eBay

 

Thanks Brendan and Ross, for the follow-up explanations.

I've always wanted to add a scope to my toolbox - there's been a time or two, in days gone by, that I've wanted to pick up this uScope (Linky) . . . the price of the Hantek is certainly tempting!

I just ordered an official Jag CKPS (LHE1640AA) from Jagbits, so when it arrives, I'll bench test new and old and report back.

 

Did anything ever come of this?