Sale ends todayGet 30% off any course (excluding packages)
Ends in --- --- ---
Discussion and questions related to the course Practical Motorsport Wiring - Professional Level
I've built a couple street/club level harnesses with less components but this will be my first time using mil spec wiring and concentrically twisting the whole thing and not just smaller sub harnesses. My plan is to have a two firewall bulkheads. One for injector/coils and the other for sensors and solenoids. The solenoids are with the sensor harness due to location. My 2 areas of concern are the turbo branch and main trunk
-A solenoid that is powered by an injector driver on the ECU (Holley Dominator)
-2 boost solenoids controlled by pwm
-Pre turbo iat
-5 pressure transducers including Wastegate Dome Pressure
-Turbo Speed sensor
- runs parallel to passenger injector wires. Engine is a gm ls v8
My concern is getting clean data from the dome sensor for tuning and from the pressure transducers for safety's. they will be run right along a speed signal and 3 solenoids.
- My concern with the main trunk is having injector, coil, and sensor wiring and not knowing what I should shield. I plan to run Low impedance injectors in the future but not sure if I have to oversize the twisted pairs If I run them twisted per cylinder from the ECU. I don't know which injectors I will run but I've seen they can be 2-4 ohms and at 12v that could be 6 amps per cylinder. Obviously they don't all fire at once. Previously I have run a 10 gauge wire split right before the injector sub harness to 4 pins on a dtm 12 pin connector. On the sub harness side two pins went to each 4 cylinder bank. I have also split 5v and sensor ground at ecu and run a shielded cable to each sensor which made installation easy but that would add bulk and probably make it more difficult concentrically twist. I'm also running IGN1a coils which can pull a few amps at 4.5-5 ms of dwell time as well as a DBW throttle body.
I looked and found a short article about shielding so feel free to link me to stuff that I missed. I also find more information on the engine bay harness side but less about before the bulkhead and integrating the wiring from the pdm. Not knowing if I'm running 10 gauge or 8 18 gauge twisted pairs is making it hard for me to plan my concentric twist. It seems like this is a common thing people ask or maybe stress about early on so feel free to include theory or articles or reasoning with why something should or shouldn't be shielded. I've also previously gotten information previously from areas where stuff is oversized and higher quality mil spec wiring wasn't used. So for instance i've seen 18 gauge per injector for power, 16 gauge per coil power with each bank going to a 10 or 12 gauge wire, and sensors using 20 gauge wire and Wideband sensors and DBW throttle body using 18 gauge. If someone here with more knowledge has the time to drop some theory here and give me some advice on how you would integrate power into the harness (ie split at ecu or split before bulkhead) I would greatly appreciate it. I've made excel sheets with all the wires and where they go but can't plan a concentric twist until I figure out wire size, what's shielded and where i'm putting the splices to accommodate shielded pairs. I can't emphasize how much time is wasted stressing about this... Thank you for any help or direction.
I seldom see shielding used on anything but low-level signals (knock sensors), or high-speed digital signals where the timing edge is critical (cam/crank sensors, perhaps wheelspeed sensors). An active pressure sensor should be able to easily create a solid output signal without any special precautions.
As for wiring gauge, I do nearly everything in 22 awg, only using larger (18 awg) for coil power supplies, and DBW motors. Even low-impedance injectors are normally run in a peak and hold fashion, so they only draw 4A for about 1ms, then the current reduces to 1A typically, so 22 awg will be fine.
Remember most wiring chart listing maximum current are referring to a maximum constant DC current. Most power usage in a running engine is for a much reduced duty cycle, which allows much smaller wire to work without issue.
Now - if you experience an issue, and the solution is to shield a signal, then that experience is incorporated into the next harness. I think it's overkill to shield every sensor run unless you have a known problem.
I've got a truth table I follow to determine which wires need shielding and which don't. It basically looks like this:
VR sensors and Knock sensors are high frequency analog signals, so they need shielding. CanBus and Hall sensors are high frequency, but they're digital, so no shield needed. Temp sensors and most other sensors are analog, but they are low frequency, so no shield for them either.
There are exceptions to this rule of thumb, but it's a good start point if you aren't sure where to begin.
I generally use 20 awg spec 55 for injectors and branched to each ignition coil, though IGN-1A's in your application could be an exception as they can draw a fair amount.
Pulling from log data I have of cars I've done PDM's on, I think I've seen around 3.5 amps continuous (25 Hz log rate) for 8 injectors N/A.
Ignition coils from the same car were around 10 amps.
Up-rating amps of a wire gauge for lower duty cycle is a fairly common practice, as well as de-rating for temperature, wire bundle, length and insulation temp rating. I'd advise you to look into the Military standards for the wire you're going to use. There is a lot of debate on amp ratings regarding wire size because of the amount of variables involved. My background is as an electrician in commercial buildings and the calculation tables in that industry change about every 4 years (with a new code book or supplemental bulletin), so even among professional engineers there is debate.
But hey when in doubt the larger size is usually safer.
As for where to split power all depends on how you choose to run it and the amp ratings of conductors and terminals. If the pins are only rated for 7.5 amps your going to have to account for that if the load continuous load is higher. Lots of reading one could do there as well; I believe Deutsch does their ratings in continuous amps at a pretty high ambient temperature. If you have bigger pins you can use 1 or if smaller split it into 2 or even 3. Sometimes concentric layering drives my decisions to use multiple wires vs one.
In my experience there is often a wrong way to do something, but multiple correct ways.
Shielding wire, see above lol. Solid answers there