I've had my S2 for around 9 months now and I have always thought, surely 211hp should feel quicker than this. So the first thing I did when I got an iphone is download Dynolicious. Ran umpteen runs and the result was always around 195 engine hp (using 17% losses) give or take a few. I also did some 1/4 mile runs and 0-62, and the results were around 15.3s and 7.5s respectively. Published data seems to be around 14.7 and 6.7s. And the car does smell like it's running a bit rich. By the way I have a Promax chip installed. I'm fairly sure that this chip would still rely on the hall and knock sensors (and still retard ignition if there is a fault)

Hmmm. A read of the workshop manual provides some possible clues - words to the effect of "Timing is retarded 6 degrees if a fault in the Hall sensor or knock detectors is detected by the ECU". Right, so I disconnect the hall sensor and make another couple of Dynolicious runs - and the result is the same. Then I reconnect to see if there was a dry connection - same result.

So the possible faults are hall or knock. I wouldn't think the knock sensor would let go - but someone out there may suggest otherwise.

So how do you test a hall sensor without an oscilloscope? The hall sensors are very expensive so I don't want to replace it "just in case". Would it be possible to test with a DMM for a signal when idling, or perhaps by turning the crankshaft by hand?

Anyone out there ever diagnosed a faulty sensor?

 

Anyone had a hall sensor fail?

 

My hall sensor completely crumble apart in my hands when I was doing a belts/fronts seals/valve cover seals job.

I never ran it without the sensor. I just ordered a new one right away.

So, it's possible that your sensor has failed, but there are more common and less expensive problems you may want to look at first.

- vacuum leak
- O2 sensor
- cap/rotor
- plugs / plug wires

Maybe try swapping the stock chip back in, and see how it runs.

 

O2 sensor is checked and OK.
Cap/rotor is OK - no misfires - but I am replacing these just because they are getting old (original rotor and cap)
Plugs / plug wires seem to be OK - the car only has 80,000 miles.
How do you get the hall sensor out? Access to the bolts is fouled by the cylinder head. Maybe a specially bent/cut allen key?

 

Did you take into account that it's 20 years old and probably has 100k miles?

Internal engine wear increases tolerances and will cause HP to drop over time. In fact, I'd say that your car sounds quite healthy, if it's original.

 

Clark has instructions for sensor testing

http://clarks-garage.com/shop-manual....htm#speed-ref

And you don't have to get new OEM part which are expensive, some BMW's have that same sensors, Bosch also which is much cheaper. Paragon sells them.

http://www.paragon-products.com/Refe...61.210.002.htm

 

Mine has 80,000 miles, which isn't a lot for these engines as I understand it. There's a UK site that has a list of around half dozen S2 dyno test on these 20 year old cars. The results indicate that almost all of them were making at least 200 bhp and they all had more mileage.

Secondly, there is no difference to performance when the hall sensor is disconnected, yet the ECU retards timing by 6 degrees if it doesn't see the sensor. So, I am quite sure that something is retarding the timing. It is either the hall sensor or knock sensor.

 

Clark only has procedure for testing the reference sensor. 16v cars also have a hall sensor behind the camshaft drive gear.

Thanks - Paragon does sell the hall sensor - quite cheap too.

 

Well I found where my horses went. The rear knock sensor wiring in the S2 has a near 180 degree kink built into the design where the boot comes of the plug to the sensor due to the intake manifold above and the route of the wiring back down. So I pulled off the fuel rail (are these always that difficult to remove - I had to lever it up carefully with a large screwdriver. In the end I had to remove the clips off the top of the injector to get the rail off and then leave the injectors in the manifold. You wouldn't think that rail held that much fuel - what a mess). I have a new sensor to go in - but I could not extract the sensor from the plug. Upon inspection the insulation around the boot was split and the wires inside frayed and "shorting" to each other and also to the shielded part of the cable. Before I disassembled the wiring looked perfect from above - I was surprised at the damage. I decided to reinstall the knock sensor and plug but cut the wire at the end of the plug and wire a (temporary) new connection to the wiring harness - very difficult with manifold still in place.

I took it for a drive tonight and what a difference - the bumometer reckons an increase of perhaps 10-15 hp, as the ignition is presumably no longer retarded. So this is what an S2 should feel like - no wonder I was only a second or two faster on a 1:10 lap than regular na's on the track.

So now I need to buy new connectors and wire to splice to the harness. I spoke to an auto electrician and you can buy the plug (with boot) for $12. They don't have shielded wire, but the electronics store sells shielded microphone twin core round cable for $2.50/m. He also recommended splicing the new plug and wire to the harness using screw terminal connectors wrapped in heat shrink. You also need to connect the new shielding to the old shielding. But to do this properly I have to remove the intake manifold - how hard is this on the S2? I know you have to take the fuel rail off and ISV to get access to the pan head bolts. Looks like a tight fit to the brake master cylinder/reservoir.

So - if there are any S2's feeling a bit slow - try this. The car actually seemed to run perfectly with the faulty sensor wiring. I didn't know how it should feel as it had this issue since purchase.

I'll need to re dynolicious and compare results.

 

Good detective work! There are many applications that require shielded wire, check with an avionics repair shop. AFAIK the shielding doesn't need to connect continuously along a circuit, they are usually just grounded. That would save pulling the manifold from what you describe?