Help with planning wire loom (Details inside)

I'm building a MilSpec harness for my heavily modified 2003 Subaru WRX. It's the first motorsport-related harness I've built. (I built several harnesses when I was in the US Navy - but they were all point-to-point: no branch circuits).

I've spent nearly a year researching, documenting, etc. I'm getting very close to beginning construction.

My engine harness is coming from the firewall via 61-pin MilSpec connector. I'm concentric twisting 5 layers including a single core wire. I'm trying to plan which wires will be in which layer so that I don't have wires from layer 2 pulling out through the other layers to connect to a sensor, etc.

So I started making a diagram showing which wires are included in each section of the loom... but that's where I also realized that not every wire home-runs back to the 61-pin connector. There are wires which are common which must break from the core loom onto a branch, but also continue through the core loom to another branch on the other side of the motor. For example the injector and coil common wires, and some grounds.

Also, when showing a branch circuit, at the main loom where the circuit branches off, some of the wires go in one direction in the main harness and some go the other way. What is the best way to show this in a diagram?

Does anyone have experience creating a diagrams like this? Even better, does anyone have a template?

I'll attach some screen shots of what I have currently. Thanks in advance for your assistance.

Attachments:

Branch Diagram 1: The diagram I've started creating to show what wires are in what section of loom. The heavy black line represents the main loom.

My Engine 2: Photo of the top of my engine with lines added to show sensor/connector rough locations

G'day Joe.

You're 100% on the right track with your documentation. I haven't found a cheaply achievable way of digitizing loom designs easily. Microsoft Visio, with a lot of time put into connector diagrams and the like seems to be the industry standard. Although that RapidHarness program looks very interesting, it's licensing scheme looks pretty damn expensive :-(. To be honest? I usually draw my branch diagrams by hand still, I've experimented with a couple of ways of doing it in software, but its just much quicker to do it by hand. I use google sheets for the bulk of the actual harness design, specifying what connects to where and so on, this coupled with a tidy hand-drawn branch diagram with lengths is enough to get the construction spot on I find.

For your wire layup schedule, I'd focus more on trying to group the physical type of wires in layers first (shielded, twisted, size, etc), and then making the branch points tidy as a secondary goal. For example, your Crank and Cam position sensor wiring is most likely going to be shielded, so will be pretty firm and bulky. This might make a really good core layer. A few months ago I made a post on the Guild of EFI tuners facebook page that might help out a bit:

https://www.facebook.com/groups/737420992943719/permalink/1766767276675747/

It's inevitable that you're going to end up with wires buried under layers that need to break out at points, hell, sometimes you lose a complete layer at a single branch, and the direction of twist on other layers needs to change! As branch points end up very rigid and well supported, its okay for them to have a certain amount of 'messiness' to them, within reason. They end up wrapped in kapton tape, and then booted so they can't move, meaning the wires underneath wont see any excess strain. I've attached a couple of pics shamelessly stolen from RaceSpec on instagram, as they show what I mean pretty well. The guy's a true master, his work is really great to learn from.

Thank you Zac. I read your linked post on Facebook. Good info there. I'm glad to hear that it's not imperative to have wires in the exact layer corresponding to where they leave or enter the loom. That was a "must" at a previous employer where we built industrial control panels.

I think I had a mental block based on the assumption that if I start a concentric-twisted loom, that the wires would just "peel away" from that group at the branches (leaving voids in the original twist) rather than just retwisting the loom after a branch based on the number of conductors are remaining. The "Sample2" photo attached above cleared the mental block.

I'm not sure if any of this makes sense. I'm suffering from caffeine withdrawal at the moment. ;)

It all makes sense Joe, except for the Caffeine withdrawal, which is completely unacceptable and I think it's imperative you do something about ASAP! ;-).

Because we end up using so many different types of wiring in a motorsport loom I find I need to end up varying from any specifications that actually exist, they just don't really seem to apply to the motorsport world. Even the rules and techniques I make for myself end up getting modified on a case by case basis when required.

Keep us updated on what you;re up to with it though, really keen to see how it comes along :-).