camshaft position sensor 850/1994

Post by **stone36** » 17 Oct 2007, 11:31

Also check, no, just replace the fuel pump relay, it is under the main fuse box section under the hood. Had the same problems with mine, replaced both fuel pump relay and camshaft position sensor, both very easy. To replace camshaft sensro make sure you have a torx 25 I think.

stevegs · Post by **stevegs** » 19 Oct 2007, 06:49

Hi y'all,

Many thanks for your replies, Lee and Stone. Now I've finished getting my motorbike through its annual test ('63 Royal Enfield Continental, no engine management system

, and it passed OK) I can concentrate on the car.

Today, I took off the camshaft position sensor. I made sure no fault codes had been logged, spun the starter, and it wouldn't fire (this was with a cold engine). Only fault 3-1-4 (CPS signal absent) showed up, so all was as expected.

I then checked what voltages appeared on the wiring harness (female) connector. Looking at it with the two recesses and the retaining spring at the top, these were: 0V on the left pin, 5V on the middle pin and 11V in the right pin, all with respect to the battery negative terminal.

I am an electronics engineer, so I have the kit for further testing, which I did as follows:

With ignition OFF (important), I measured the resistance from the left pin to battery negative. It was less than 10 ohms, which was as I expected - it confirms this is the 'ground' pin.

With ignition ON, I measured the current drawn to battery negative from the other pins, with a 1kohm resistor in series (important, since without it you could damage something). These were: 1.43mA from the middle pin and 11mA from the right pin. Again, all as expected - the right pin is a 'cleaned up' battery supply, making it about a volt lower than battery positive. The middle pin is the signal to the ECU - the CPS output is a transistor switch to ground, with a resistor pull-up in the ECU to its internal 5V supply. From the figures recorded (5V and 1.43mA), this resistor works out at 2.5 kohms - I expected anything between 470 ohms and 47 kohms, so again all seemed well.

Since I haven't got a breakout lead, I bench tested the CPS using 12V and 5V supplies, with a 2.2 kohm pullup resistor to 5V (this is the next lower 'preferred' value in the industry standard E12 resistor value series) to simulate the ECU. As has been noted by someone else earlier in this thread, the CPS is a Hall effect sensor: this has an internal amplifier and switch. The output is therefore digital (ie. a square wave when the car is running, not a sine wave). The CPS includes a magnet, which seemed to be quite strong - any weakness here could give problems at high temperatures owing to the Curie effect, whereby all magnets lose their strength progressively as they get heated up.

Since it was now out of the car, I had to simulate the interrupter (the non-magnetic steel half-cylinder bolted onto the end of the camshaft) with a 1/2" wide piece of thin steel. With no steel present, the output was 0.2V - anything less than 0.4V should be OK. With the steel present, the output was 5V - anything less than the actual 5V power level indicates leakage in the output transistor switch. I monitored the output with the oscilloscope to check for glitches - didn't see any. So far, so good....

I heated up the CPS with a hair drier while passing the steel through the gap - all worked as advertised, even when it was quite hot. I then used a freezer spray on the sensor itself, which gave it a temperature change shock - it still functioned fine, so I reckon there's nothing wrong with it.

I refitted it to the car: to secure the connector pair properly, I had to remove the air filter top for access, which involved temporarily disconnecting the mass airflow sensor. Lee, I entirely agree about cable routing, and paid attention to this when I first had the problem. During reassembly, I clipped the HT lead into only the two outboard recesses in the air filter housing, leaving it out of the inboard one since otherwise it would have been touching the airflow sensor cable. There is a further recess for the airflow cable on the main body of the air cleaner box - the two cables are at least 1/2" apart and cross at right angles. The CPS cable is a good 3" away.

Before starting the car, I rechecked the codes - this time I got 1-3-1 (no RPM sensor

). Note that I had not cranked the engine since the last reset; I'd only switched the ignition on and off a few times. Wondering if I had got a confused ECU, I reset the codes again and cranked the engine - it started fine.

Stopped it, checked codes - none logged. I locked the house and went for a 3 mile drive. Got back, checked codes - none logged. I waited for 30 minutes - just the time it should take it to cool to luke-warm after a short journey. It started fine. I waggled the leads to both the CPS and RPM sensor with it running, stopped it and checked codes - none logged.

I've yet to check the fuel pump relay, but I think that must be OK since I can hear the pump start every time I switch on the ignition, and I usually let it reach pressure (when I hear it stop) before starting the engine as a matter of course.

I'll be doing about 120 miles this weekend in a number of journeys. At the end of each, I'll check for any logged fault codes and let you know.

To quote Sir Arthur Conan Doyle "When you have eliminated the impossible, whatever remains, however improbable, must be the truth." The cynic in me thinks Volvo might have put in a deliberate bug (no - feature!) whereby after you have done something totally unrelated (like engage reverse gear) a certain number of times, it will flag a CPS fault and then refuse to start when at a particular temperature.

BTW - the CPS fixing bolts are Torx T40.

Post by **stone36** » 19 Oct 2007, 09:54

WOW!

If I can plug in an extension cord I feel like I have a victory... With your background you may disagree with my next statement but it has been the case as I have observed it. Electronic components on my volvo have worked until they haven't. The reason I say that is if you do find it to be the CPS, when mine went out, I went to the parts yard and pulled 2 off a couple of dead volvo's figuring, it probably wasn't the CPS that caused the wreck and they should still be good, I was charged $10 for both, relaced one and it fired right up, two years later (with a scrap FP relay) and it hasn't missed yet. I even sold the second one untested on ebay for $40!! I think even FCP groton asks $160 or something for this part. If you have the 15 minutes and a close enough parts yard, get a couple from them.

As for the fuel pump relay, mine was very intermittent until I just replaced it, even new from volvo they are maybe $20. I keep three spares in my car.

When my CPS went out for good I had three codes, the two you had and then one other, that wasn't too worriesome, after talking to volvo, they figured that the CPS going out caused the next step, RPM senson to fail, when I replaced the CPS it cleared all three. Don't know if that helps or not, but information ususally doesn't hurt. Good luck, stone36

Post by **Ozark Lee** » 19 Oct 2007, 18:57

The cynic in me thinks Volvo might have put in a deliberate bug (no - feature!) whereby after you have done something totally unrelated (like engage reverse gear) a certain number of times, it will flag a CPS fault and then refuse to start when at a particular temperature.

I bought my wife an Audi 5000 years ago and I was convinced that they needed to add a credit card reader somewhere on the dash as standard equipment.

Every 60 days I had to take the damn thing to "Hans & Franz" (the indy Porsche - Audi guys) and the bill was always $650.00. I posited that it would be much more convenient if I could just swipe my card, get debited $650.00, and have the car run again.

Us EE's think alike!

...Lee

stevegs · Post by **stevegs** » 29 Oct 2007, 16:57

Hi again,

120 miles later and still no fault codes logged. One trip was another short one - 1/2 mile to take an old bed to the tip, 3/4 mile into town to get the replacement and 1/2 mile back home. It started fine every time.

So the fault might have been in the wiring or a connector, but if it was, I would have thought I'd fixed it first time round.

There aren't any car breakers near here, and I've found the nearest (15+ miles away) seldom have 850 parts. There aren't that many 850s on British roads - far more 740/940s; hence there are many more of these in breakers' yards. I did see somewhere that the 740/940 and V70 use the same CPS - it would make sense if they did.

BTW - the road wheels off a 740 do not fit an 850, as I found some time ago. The stud spacing is the same but the offset from the wheel centre-line (dishing) is different.

stevegs · Post by **stevegs** » 08 Jan 2015, 13:16

Much later now (I've been away for a long time) - and no further problems having replaced the silly thing with one from eBay in 2008 after a further non-starting bout around Christmas 2007. The only difference I could find between the new CPS and the old (when repeating the bench test with the CPS out of the car) is the 'low' voltage (ie. with no steel present) is 20mV on the new one and 200mV on the old. As I said earlier, up to 400mV is OK as a logic 'low' output (and up to 800mV should be valid 'low' to a logic input circuit). The output from the CPS goes to the main ECU - so I can only conclude I've got a fussy ECU.

The fact that the new one pulls down to 20mV indicates Volvo must have become aware of the problem and done something about it. I'm sure there is nothing inherently wrong with my old sensor.

gwb · Post by **gwb** » 18 Jan 2015, 13:01

The thorough nature of these posts is most helpful. I've been having similar issues with my '94 850 and will see if the CPS can resolve the issue. Many thanks!

Post by **precopster** » 18 Jan 2015, 14:16

Has anyone here mentioned to the OP that the high tension lead from the coil going to the distributor should be routed far enough away from both the Cam PS and Crank PS so to avoid interference?

There is a plastic clip on the engine bracket for this purpose. They often break. Or perhaps the insulation on the high tension lead is beginning to break down and is able to cause radio interference to both the crank and cam sensor wiring. Check the dates on the high tension leads. Bougicord place a build date on them.