Forum » Practical Harness Construction - Club Level » Couple of Questions on Shielding.

Couple of Questions on Shielding.

Practical Harness Construction - Club Level

Discussion and questions related to the course Practical Harness Construction - Club Level


Page 1
Author
74 Views

Planning out the last few feet of my loom here. Shielded wire is the next thing to order but I have a couple of ways I can go.

Intending to use 20AWG M27500 cable for the few shielded wires. On this Honda B series harness they are for distributor signals, CKF Sensor (2-wire, so VR), and knock. I am omitting the shields on the HO2S signal wires.

My questions:

The service manual shows that the original distributor sensors are shielded only on the power side (CKPP, TDCP, CYPP) of the VR signal wires. I'm wondering if I shouldn't do a fully shielded 6-wire cable since I am integrating coil trigger wires into the loom for a COP conversion and they will be running alongside the signal wires all the way back to the ecu. Same goes for the CKF sensor across the back of the engine to the oil pump. Is it overkill to shield all 6 wires? Is there any possible benefit? More importantly, is there any reason I might NOT want the ground side of my signal wires contained within the shielding?

Secondly...

I originally designed to run the wire for the Bosch knock sensor through the loom parallel to the stock knock sensor. I'm running a stock ecu/Hondata S300 setup with no knock control and use a Knocksense MS to a spare analog input on the ecu to mark knock. I have to revert to the stock system for emissions testing, and was planning to use a DTM connector at the sensor end to allow me to swap between two sensors for emissions testing with the stock ecu. Would it be possible to to tee the knock signal wire off to the ecu and the Knocksense on the cabin side of the loom? Would there be any interference caused by having one leg of that fork not connected to either the ecu or the Knocksense at any given time? The other concern is that the Bosch sensor has the 2nd pin for the shield whereas the stock knock sensor has only one pin and runs through a single conductor shielded wire, with the shield draining back to the thermostat housing ground while the Knocksense drains the shield on its negative terminal at the box. If I were to set this up as stock, with the shield draining to the engine side, would this cause an issue while running the Bosch sensor? I can't think of any, as it doesn't produce a ground loop, but trying to make sure I don't inadvertently build any headaches into this loom. Thanks!

It would be more common in the aftermarket world to run a separate 2 core screened cable to each VR sensor. So in your case that would mean three 2 core cables. I have never seen it done with 2 or 3 sensors all going through the same cable so although it may work, for the sake of say $30 extra worth of cable I wouldnt risk being the guinea pig.

You cant share a knock sensor signal between two devices, the impedance change will mess with the filtering. You will need to run two separate knock sensors.

It seemed strange to me too, but nonetheless there it is in the attached pic. I can run the four separate pairs to the sensors, not a big deal aside from adding bulk to the loom.

As far as the knock sensor, it wouldn't be connected to two devices, it would be either to the knock box or the ecu, depending on which one is plugged in. I t would leave a loose lead to whichever arrangement is not currently plugged in though. Just wondering if that extra lead would act as a sort of antenna for noise while not being terminated to an ecu or knock box.

Attached Files