Hello to all,i will make new engine loom for my 2wd sierra cosworth,i have bought raychem 44 cable,so what is the correct type of cable for map and knock sensor lines,and what thickness? thank you very much

Thanasis

 

IIRC, the MAP is shielded back to the ECU to prevent EMI.

Not sure on the knock.

Martin

 

Shielded or twisted pair for both to stop interference. 24awg I think? Probably need to get that confirmed though

 

Thank you very much

 

Knock sensor would need to be shielded.

Map sensor does not.

 

Thank you

 

I disagree, otherwise why else would it have been done?

Martin

 

I shield Crank, phase, knock and Map sensor, grounding the shielding at the ECU end.
Its also vital to ground the ECU pins 1-19 at the correct point to avoid introducing noise into the ground path.

I see loads of looms unshielded and most do have issues, had one in particular that wouldn't even start on crank RF from the starter picking up on the map sensor, was interesting watching it emulating

more info here.

http://shop.motorsport-developments....loom-947-p.asp

 

It would be very rare to see an OEM loom have a MAP sensor shielded. If Cossie's have them, they would be a rarity. Must admit I've never even looked at a Cossie, it's mostly jap cars I'd see

Maybe Cossies are strange that way, but in 15 years or so mucking about with looms, ecu's on Jap cars etc, Ive never needed to shield map sensor wiring, nor seen a car with it.
Any custom wiring looms I've bought or patch harness' from ecu manufacturers, again have never shielded the map sensor wiring

And have never had an issue where even thought to shield it.

 

Yes the OE YB loom has shielded Map sensor wire

 

It must be one of the only cars with such wiring ?

Maybe just due to it's age ?

Typically on a modern car, the only shielded wires you'll find are VR sensors, NB Lambda and MAF signal.
All of which are essentially high frequency/oscillating signals

Bog standard 0-5v voltage that are fairly linear in operation are never shielded. Even rotational hall signals like crank, cam, wheel speeds etc arent shielded in most cases

 

Modern ECUs are far better at dealing with noise and interference than the old webers, The P8 is far better at dealing with it then any of the L1-L6-L8s,

 

Don't know an after market ECU that would require shielded map sensor wires. Cam, Crank & Knock only require to be shielded. If you are having trouble with an unshielded map sensor may be best to throw the ECU in the bin its either very very old or shockingly designed.

 

I can only go from what i have seen and tested here while developing my Looms, And the fact that Ford themselves shielded the Map sensor 25 years ago.

 

The topic starter is talking about a Weber ECU Rod, not an aftermarket one.
And yes, his Weber ECU may well be 30 years old... as was the one in your car.
They suffer immensely from RFi and have virtually non existant internal noise filtering.

Put an unshielded loom in a Cossie engine bay and pop a decent scope on it and watch what happens to the voltage output when you move the king lead near the sensor in the engine bay.
Better still, emulate it and you can watch it change load sites yourself. The loom needs quality shielding and needs the grounds installing in the correct place.

Even better - Try the same test with some of the old Magnecor leads on and you wont even have to move the HT leads as the blody engine will go so rich it missfires sometimes. LOL

Also, while the P8 has the best filtering of all the Weber ECU's of that ERA, Ford still produced a sub loom for the map sensor back in 1993 with even better shielding as they had problems of the car not starting under cranking due to the interference from the starter motor.

We are talking about cars from the 80s here that still had interference suppressors fitted to coils and alternators so that you could listen to a radio or watch a TV in a house on the same street as the car was driving.

 

I presume the shielding is probably due to the nastier environment of the Cossie engine bay.

ie long HT leads, Dizzy, old conventional single coil etc. It was probably a very noisy environment compared to some more modern stuff.

TBH I've never even seen anything on any logs on any ecu's or engines Ive ever worked at where there has been noise on the map signal.
And that's even making a plug in patch harness to run an aftermarket ecu directly in a Cossie using all OEM sensors etc

 

Yeah i agree im sure it is to do with the noise in the YB bay, but that's why my YB looms are shielded in the way they are, because they are still in the same bay they were 25 years ago.
They do seem to become even more picky about noise when converted to wasted spark too, i think many of the issues people see with the WS kit are down to poor Looms, and unsheilded Looms, it may even be down to the extra RF from the HT leads etc.

For what it costs extra to shield the Map i wouldn't even consider not doing it on them TBH,

 

The problem a YB on Weber management has is that the ecu is very sensitive to changes of a few millivolts which on other sensors are not really critical as they are all filtered electrically within the ecu and in the software to slow them down and just look at averages.

However, the map sensors signal is sampled every 90 degrees and recorded by the ecu so that it can build a model for each cylinder.

If the pressure suddenly rises then it assumes that the throttle has been opened and a transient fuel correction factor is tagged onto the injector pulsewidth to ensure no lean out occurs.

The whole system is speed density and that means its entire fuel calibration and delivery system is based almost entirely on map sensor voltage, so any variation in voltage output results in a fuelling change. This is why its essential that the signal output is a genuine reflection of manifold depression/pressure and not just a niose spike. Remember, we are talking about millivolts here, not volts.

 

So the old Weber system doesnt use TPS for transients ?

 

The original map sensor screened cable the outside core is grounded with pin 19 in the ecu plug,but i don't know in the other side in the plug where is going the outside core...

 

Screens must only be grounded at one end, usually the ecu.

Never connect at two ends otherwise the screen is lost

 

Sound advice, connecting the screen at both ends creates an earth loop that can cause more noise problems.

 

Thanks a lot