Electrical noise and suppression.

[2fast4u2c]

Just a quick question for you electrical guru's. 

My bike is wired like a lear jet with more sensors and engine mgmt. gear than I care to think of.  With everything going on and wires running next to each other, does anyone have a suggestion for the kind of wire I should use to help alleviate any noise generation or emf problems?  I thought about using RG-58 or 59 cable with the shield (CB cable) but I'm wondering about the impedance issue?

any helpful knowledgable thoughts welcome.

Guy

[1212FBGS]

A lot of those sensors don’t generate or run on very much voltage...usually 5v...if ya want to be safe, isolate the ecu and ign wires by wrapping them separate and avoid any wires that may cross or run close to high tension wires... but your probably using stick coils anyhow...
Kent

[Dean Los Angeles]

Electrical noise comes from two main sources. One is radiated by high frequency signals that can be detected at some distance, and the plug wire is a huge source. Don't run anything close to the plug wires. Even if you are running coil-on-plug the trigger is still carrying a high frequency signal. Anything that is high frequency could be a problem.

The other source is induced by the transformer effect. If you put two wires next to each other and one has a changing voltage it will create a magnetic field that will carry over to the wire next to it. DC won't do it, but any kind of sensor that is changing state, like the tach or mph pickup, certainly will. The longer the distance that the wires are in contact the bigger the effect. Two wires that both have a changing signal can cross over to each other.

#1. If it ain't broke, don't fix it. No sense tearing the bike apart if it's already running well. If you are running new wires or putting the vehicle back together, then try and determine what that wire is doing and separate it or re-route it if possible.

RG-58 or RG-59 would work for sensor wires. I have a couple of inputs that might be coax.

[Seldom Seen Slim]

I agree that RG-xx cables would be okay for sensors -- that carry a relatively stable signal.  But for something that pulses I'd stay away from them.  Am I right, Dean, ikn thinking that the capacitive effect of the coax might start doing funny things to those pulses?  How about using plain wires, carefully dressed to avoid parallel runs if possible (where necessary) and then have a ferrite bead on the ends of the wire, to restrict RF that might be on the wires?

[2fast4u2c]

Great info.  I do have the trigger wires from the ECU to the stick coils wrapped and isolated.  Your right too, if it ain't broke.  I guess I'm just so skittish about doing well this year that I'm trying to over engineer things again.  I gotta stop doing that, you'd think I was a Honda engineer.

Thanks,
Guy

[smcleod007]

My memory of wiring techniques is pretty rusty but mainly try not to run any low voltage control wiring along side something that produces an alternating field. I try to make sure that the bikes sensors that I tap into for data logging ( TPS, MAP, TACH, CRANK, ect...) don't get routed along side wiring from the fuel pump, 12 volt solenoids that get pulsed off and on, fuel injectors and the items that Dean mentioned. On my bike I have twisted the power leads to the fuel pump, fuel injectors, stick coils, air shifter and boost solenoids, water injection pump and air compressor to name a few. This is to keep them from inducing an AC signal in to their adjacent wiring. I would not use coax on a sender unless it was factory wired that way or I had determined it was definitely giving me an error. I've seen a lot of noise get amplified in a wiring system when a coax cable gets installed and the shielding is not properly grounded to a good grounding bus.