So ideally, the charge voltage at the battery will be ~14VDC. As the battery voltage approaches this value, the charge current drops. The resistance of the wire from the alternator reference at the Main Relay box actually may be beneficial to some degree, it will serve as a current limiter that the battery reacts against. The instantaneous voltage at the battery terminals is static (ignoring the battery internal resistance) regardless of load. The battery stabilizes the system voltage in this way. The direction of the current at the battery terminal is dynamic with load in this scenario. If the alternator voltage minus the losses (both resistive and otherwise) across the source network to the battery is LOWER than the battery voltage, the current will flow out of the battery. If it is HIGHER than the battery voltage the current will flow into the battery. The fuel pump described in this thread may approach a 30A load, which is significant. This will theoretically result in the said voltage being LOWER more frequently, which will result in a discharge cycle. Couple this with the reverse temperature coefficient of the reference in my alternator (at least), and you get alternator voltages of ~13.5 volts at temperature, and battery voltages in the high to mid 12's. While I haven't measured the current into the battery under these conditions, I would venture to guess that they are nearly zero, so not really charging, not really discharging.

 

I think this is a very real possibility (1V+ drop)

 

If I were to go as far as run 10awg wire back to the fusebox from the fuel pump I would probably just run a pair of like 4awg welding cables or whatever from battery to engine bay. all of the voltages erywhur

 

also a valid solution

 

If it poses a problem it's easy enough to route the fuel pump lead to the alternator or fusebox at a later date. I assume the logic here is that the charge/discharge cycles will decrease battery life?

The DW300 at ~85psi draws ~13.5A. The Walbro 450 at 85psi draws ~18.9A.

 

Yes

 

As well as decreasing the OC battery voltage.

 

Relevant section of the wiring diagram for 01-05 Miatas.

 

And thats why I rewired mine. I wasnt concerned about voltage drop and getting every bit of flow, I was concerned about melting the stock wiring and dieing in a fireball.

 

Any plans of a kit with all this stuff?

 

Ships tomorrow

 

Yes, I'll do a parts kit. It won't be cheap.

 

I'd be interested in some wiring/relay kit too maybe depending on what all it is. Already have the DW300 and Fuelab FPR. Need to read through this and a few other threads again a little more carefully but need to start planning out my fuel system for my TSE EFR install. It's great to read up on what you guys are doing.

 

Is fudging with the bulkhead required? I've heard the short run of wiring from the bulkhead connector to the pump itself wasn't long enough to require the increase in gauge.

 

As stated earlier. Its not enough to cause a significant drop. But some of use are worry warts and I don't like the idea of 20Amps going through the stock 26 year old connector on my pump.

 

Just run a 2nd wire in parallel with the red/blue wire from the relay to the pump to augment it. AWG 12 or 14 is good enough for 20A.

 

Not if you want to do PWM. Or have less voltage drop. Or not have a toasty relay under the dash running lots of current through it. Also that is a lot further to run a wire than to the battery.

 

Did you build the 12V 1.0uF capacitors or the 0.001uF Signal capacitors into the harness?
Based on the slim harness, it doesn't look like it.

I'm planning to re-make mine to make the lead to the OEM harness connector longer so that it's not all stuffed between the motor and firewall.
Originally I used a single 0.47uF capacitor on each of the 2 harnesses (coils 1&4 and 2&3) since it was the largest I could find on the day I made them but didn't include the 0.001uF capacitor since I'd read that few people included it in their harness:

- the capacitor is clad in clear heatshrink, so that I can see if/when it leaks, and zip-tied to the harness

As you said, making harnesses is time consuming and if you haven't used these capacitors that's good enough for me not to use them either since it makes the harness easier and tidier to make.

 

Hang on that's not right.
10 ft of 8 gauge is 6 milliohms.
The 20A fuel pump will cause 0.12 V of drop. Not worth worrying about.

Wire it to the battery using 14 gauge.
Be sure to fuse it with a 30A fuse, sitting near the battery.

 

The run from the battery to the pump following reasonable paths (along existing loom paths and through OEM grommet) ended up being nearly 8ft of wire, which will be 0.21v of drop with 10awg. 14awg would be 0.53v at 13A (DW300) and 0.81v at 20A (Walbro450).