fpscolin

Hey all, I just picked up my Innovate MTX-L wideband and I'm going to be installing this weekend. I couldn't really find it online, but what is the recommended gauge of wire to run from the O2 gauge to the engine bay? I decided to guess at bought some '18 AWG' at the local Canadian tire. No idea how wire nomenclature works, just wondering if I can use the wire I picked up and if not, what I should be using.

Very excited about this install, since it's the first 100% turbo-oriented car work I've ever done! Megasquirt2 PNP is on its way as well...

cyotani

iTrader: (8)

22 awg is pretty typical for analog signals like this.
18 awg is more used for higher amp power wire and is overkill but it would work just fine.

acedeuce802

iTrader: (1)

Googling will find that an MTX-L draws around 3 amps. You can look up charts that have a safe wire gauge for each amperage at the length of wire. 18 gauge wire is safe for 5 amps at 25', so you're definitely safe to use that.

Edit: These 3 amps would be for the power and ground wires. If you're talking about the signal wire to Megasquirt, then signal wire is often 22 or even higher, up to 26 or even 30 gauge.

fpscolin

Wow, neat chart. I know next to nothing about electrical, so this was super helpful. Thanks a lot

mopnbucket

Re-opening this thread as I have a follow-on concern:

The MTX-L is set up with a wiring harness of what appears to be all AWG-22 (-ish) wires, including for power and ground.

As pointed out earlier in the thread, during sensor heating, the unit draws up to 3 amps which is well over the recommended current for 22-gauge wire. Clearly, given that this gauge/controller are widely used and it has been this way for years, this is not a problem in practice.

Can somebody explain to me why? Is it just because the heater circuit is theoretically only going to run for ~30-60 seconds at a time so it won't build up enough heat to melt? What's the reasoning here? (Looking to learn.)

Thanks!

Roland T.

There are several factors in correctly selecting a wire size such as insulation temperature rating, load and load duration, acceptable voltage drop, wire bundle size, if it's located in a conduit or not, location of wire bundle, if the end of the wire is terminated on a fuse block or located in the engine room, conductor material (**** aluminum wiring, that is all).

I usually rely on this calculator: Circuit Wizard - Blue Sea Systems but it will never recommend anything below 18AWG (ABYC standards, I'm a marine electrician). After you press calculate you can press 'explain results' to get an idea of the derating factors applied and why, which is nice if you're curious.

In my job we aim for 3% voltage drop for panel feeds, navigation and electronics and 10% for other loads such as pumps. It kinda sucks having to run 1/0 to a pump with a 14AWG pigtail just because the run is so long and the entire system is 12v only.

curly

iTrader: (25)

I was going to say, the chart doesn’t even go under 18, probably because of a standard. I will say, a lot of **** works well for 18 or 16 gauge, so I’d suggest staying there. Crimpers, crimps, connectors, etc, all auto store **** is usually 16-18 min. Yes the 0-5v signal only needs a ~26 gauge wire, but I can’t imagine what the average Joe/Josephine’s crimp on that wire would look like.

mopnbucket

Right, so that's the thing, that super tiny gauge makes sense for 0-5v signal wire, but that same wiring is also carrying the 12v/3amp O2 sensor heating power....and that should need a much beefier wire, no?

EDIT: ok I think I understand now. Please confirm if the following reasoning makes sense

1. From a heat perspective, only 3amps running for 30-60 seconds is likely not going to cause a problem even in a 22-gauge wire.
2. From a voltage drop perspective, the run of 22-gauge wire is so short (12" or less) that there will be no noticeable voltage drop.

Therefore, 22-gauge is actually fine. Is that right?

I was confused about (1). I thought that with that much amperage running through a tiny wire it might melt like a fuse, but sounds like I mixed up voltage drop rating vs. heat.