Discussion and questions related to the course EFI Wiring Fundamentals
I wanted to get folks opinions on a high level power distribution design of mine for a 1974 Mk1 Escort (in the UK) running a Megasquirt MS3X. My main issue is the battery has been relocated to the boot (to the rear) and I'm in a position where there are two power feeds going to the front; one 10mm² inside the cabin to power the ECU and most of the original Ford wiring and a 35mm² underneath to the starter, alternator and engine bay fuse box (injectors, coil pack, elec rad fan and water pump).
Am I going to run into an issue having the injectors and coil pack effectively fed from a different source to the ECU. I'm thinking the alternator output not being smoothed by the battery is going to cause issues especially with the injectors?
Please find attached high level design.
While not completely ideal, that setup shouldn't give you any major problems. The alternator output will find the path to the battery a lower impedance than to any other part of the system, so the battery should still smooth any spikes out. As long as the current draw along the 10mm^2 wires doesnt get too large, the voltage seen by the ECU should be very close to that seen by the injectors and coils.
If the draw along your 10mm^2 cable does start to bring the voltage down, the ECU will adjust things like the injector dead time, as it will assume the injectors are seeing the same supply voltage as the ECU, which could lead you to have issues.
Could you move the supply to the ECU and other electronics to the 35mm^2 cable? Leaving the original escort electrics powered by the existing 10mm^2 cable shouldn't be a problem.
Thanks for your reply. The problem you mentioned in the second paragraph is the sort of thing I was concerned about, I've nearly finished building an injector testing rig to measure dead time and that got me thinking about the wiring.. Better to know potential problems, than not know about them.
I can move the ECU and wideband across to the 35mm2 without much issue. Do you think this would be a better solution even though they will be right next to the noisy alternator?
Going back to your first sentence, what does an ideal setup look like when the battery is in the rear?
In addition, I recently rewatched the entire wiring fundamentals course and throughly enjoyed it, keep up the good work! I'd be eager to know when the practical wiring course is coming out.
G'day Dan. thank you for the kind words, glad the kiwi drawl didn't get on the nerves too much ;-).
Would be interested to hear more about your injector test rig, its something I've had marinating in the back of the brain for a while. Apparently the biggest issues to overcome are the pressure spikes as injectors open and close. You can get into harmonic regions where the pressure does really funky things, and the same injector pulse width at different engine speeds will deliver a drastically different amount of fuel per injection event. Needs some fast logging (>1000Hz) to see it. Would love to log it happening install and aftermarket pulsation dampner and log the results too... One day... One day... :-).
Your current setup shouldn't give you any problems as long as the current draw along to 10mm^2 wire doesn't get so high that there is a noticeable voltage drop across it. If this same feed is powering the original electrics, the original lighting might draw quite a bit of current if everything is on, so its something to watch out for.
That being said, yes, I'd switch the feed for the ecu to the 35mm^2 cable. What sort of alternator is fitted to the engine? If its relatively modern (internally regulated) its output should be fairly clean. The 35mm^2 cable back to the battery will be a very low impedance path to any ripples or spikes, so I wouldn't expect there to be an issue.
I'm never willing to say something is the ideal setup, as there is always someone out there with a better idea I'd love to learn from, but the way I've done vehicles like this (classic vehicle conversions to EFI for racing purposes) is to mount a kill-switch within reach of the driver, which doubles as a handy power distribution point. The 35mm^2 cable heads to the kill switch, and the other side goes to your ECU and power supply relays. The alternator output also connects in here, but must be on the battery side of the switch, so its output it isolated from the electrical system if the switch is used while the engine is running.
My dead time rig is based on ones that others have built who are also running Megasquirt, I have my doubts how effective the setup will actually be but I won't know until I try. I got a wiring harness to finish building first though before I end up in the trenches chasing dead times haha.
It's mostly the original wiring in the cabin, however I reverse engineered the original engine harness so I could create my own version (without cutting up the original). The head lights, main beams, horn now have relays in the engine bay powered off the 35mm2. To give you a better idea of my design, I've attached two diagrams describing the front and cabin fuseboxes.
It's a fairly modern 70A alternator from a Ford KA, its a Bosch compact style with internal fan that uses ring terminals. I can move the ECU, Wideband and MAF over to the 35mm2 without much issue, just need to reconfigure my firewall multi-pin connector.
In your ideal setup you didn't mention the starter, which I'm guessing mounts to the same point as the alternator. Any reason why you wouldn't attach the alternator to the starter?