Hi Folks,

I've had my 993 since December now (a 1995 Carrera in Midnight Blue, will post avatar pic shortly ;-)), and have been lurking here ever since.

Recently I had the chance to attach a friends iScan to my car, trying to hunt down an occasional misfire under load. The machine came back with the message "Knock sensor 2". I cleared the fault and drove the car again for another week. This time the diag was clear. Another check a week or so on and again I get the knock sensor message.

I've done some searches here and can see some talk of knock sensor "counting", though nobody in those threads seems to be using the iScan. Does anyone know if I should be worried? To be honest I've not even done the basics yet in tracking down the misfire (I'll do caps and rotors end of the month, leads after that if I must...)

Any thoughts greatly appreciated!

Thanks

-- Chris

 

very important. if you were doing knock test where you drive the car and it counts the knocks, this is nothing, it tells you when the knock sensor activates, and this is due to fuel quality. you fill up from one gas staton and you have several knocks, from another gas station and you have zero.

if you have a fault error code knock sensor 2, then
check this out

 

Thanks, good info. I'm a bit scared now though

Maybe this explains the misfire:

"When a fault is stored, the ignition angle is retarded for all cylin-
ders in the range in which knock control is active."

..though I had this knock sensor message for months before I detected this misfire - maybe I was still being too nice to the car :-/

I'll get on to checking the stuff off in that troubleshooting table (I like that much better than opening up the motor right now) but assuming worst case scenario (I am a bit glass-half-full sometimes) does anyone have any idea what kind of engine damage triggers these sensors while the engine actually feels really nice and smooth?

Thanks!

-- Chris

 

hoy Chris,

You're strolling down worry lane, IT IS NOTHING. When the engine knocks because of a broken valve or pierced piston, I bet you will notice it before plugging a scan tool.
even the proceedure in this porsche sheet tells you to check the knock sensor for bad connection or rust, thats it. my own opinion, I bet its the fuel quality.
Have fun mate, enjoy your car.

 

The ECU can pulse 5v on the diagnostic connector for every knock registration but only the PST2, PiWiS and Bosch Hammer will count them. The iScan is a Bosch hammer clone so it might but a regular OBD2 scanner will not.

The MY95 has a different ECU harness so if you follow these directions for diagnosing a MY02+ 996 (M7.8) OBD2 faultcode (MY95 993 is M2.1) use pins #29 for knock sensor #2 and pin #30 for knock sensor ground.

 

"When a fault is stored, the ignition angle is retarded for all cylin-
ders in the range in which knock control is active."

Actually, only the pinging/knocking cylinder gets retarded & not all of them.
As in the case of the 928, if the Hall sensor fails, the timing is retarded for
all the cylinders. Even with a failure of the Hall sensor, the DME could still
determine the knocking cylinder, but Porsche/Bosch decided to use the Hall
sensor for cylinder I.D. which is really only needed for sequential injection
without a DIS ignition (individual cylinder coils). And even a Hall isn't needed
for DIS.

 

Thanks all for the good info. From this I'm beginning to suspect that the misfire and the knock fault are two different issues.

I've been running the car on 98RON, so normally I wouldn't suspect fuel problems, except the car has had a remap gaining almost 35bhp - I wonder if this map is to aggressive and is causing the knocking?

The misfire is very likely just a dizzy cap / rotor / leads issue. I plan on replacing these in the next couple of weeks to find out.

Thanks again for all the advice.

-- Chris

 

You should have started there. The answer is YES.

Which chip did you get?

 

It's a custom map made on rollers by a company callled AMD in the UK - a couple of years ago and before I bought the car. I just have the graphs and stuff in the history file. The power went from 240 at the wheels up to 272 - that seems like quite a lot for a simple remap.

I think I can get access to chip blowing tools, so I'll try to find an original chip and run it for a while like that and see if the fault codes re-occur. If that fixes it I guess I'll have to do some work to get the horsepower back :-)

Thanks!

-- Chris

 

Wow, that's a lot of gains for a NA 993 with just a chip. Getting so much power from just a ECU remap means very aggressive timing advances. You are basically running on the verge of detonation and possible engine doom.

That means that, as long as the fuel you are using is perfect,that there are no carbon deposits, having a perfect ambient air temperature, running at sea level, etc, you may be ok, But, it just takes for one of these variables to deviate a little for your engine to start detonating and your knock sensors to retard timing to prevent engine failure.

See if you can reload or swap back the stock chip, then get a reputable less agressive chip that balances performance and reliability in there if you wish.

 

Yeah, it seemed like a lot to me once I thought about it a bit. I actually have a flyer from them in my history folder too, estimating about 10-15hp gains on a 993 job, so I'm not sure what happened on the day they did my car...

As you say, I'll go find an original map / chip and run that. Hopefully there'll be no more knock messages after that - I'll keep scanning :-)

Thanks again!

-- Chris

 

"Wow, that's a lot of gains for a NA 993 with just a chip."

Right! To the point of being unrealistic!

 

I wonder if the car was running badly on the before run? Is 240 at the rollers unusually low for this kind of car do you think?

I've also heard rumours that playing with the brakes etc can give impressions of false gains on these type of plots, but I don't really know anything about this area. All I have is a couple of printed out graphs and a sheet of numbers.

-- Chris

 

Looking again at the graphs I think I can see why the hp difference is so much: the first run hit peak at 240hp at 5500rpm and the after mapping run at 6250rpm (275hp).

The before graph drops real sharp after the 5500 revs - not sure why but I guess there was either a problem or they got out of the throttle for some reason.

I'm going to go back to a standard map for a while anyway, but a question I have now is whether a claimed ~290bhp is unreasonable for a NA pre-vario 993 with a remap?

Also, is it legal / acceptable for me to ask if anyone has an EPROM image of the standard chip? My car never came to me with the original but I do have access to blank proms and a programmer...

Cheers

-- Chris

 

Hi

Mine has just been measured at 287.5 by Weltmeister in Silverstone, allegedly an accurate machine

Pete