To Summarize the project:
After reading all the posts, here on

, on just how lousy the headlight voltage really is, I decided to take measurements with my Fluke Model 87 III digital multimeter and those values recorded were
verified with my Fluke Model 8021 B digital multimeter. Both meters have been through their calibration cycles within the last year so, I have faith in the accuracy of the values reported.
All of the voltage readings were taken as follows:
Both bulbs were removed from their respective buckets and left dangling on the factory harness. The meters were then connected across the
terminals of the bulbs and the voltages recorded. The engine was off, the key was off. I regarded this as the "worst case scenario" for the voltage.
After verifying the length of the wire run, I made the decision to use 16 gauge wire as the current draw is less than 10 amps and the run length is less than 10 feet. Direct Current wiring is susceptible to line loss to a much greater degree than is Alternating Current so wire size is important. On the other hand, utilizing wire in a gauge larger than is
required is just a waste of monies. It does not buy anything is the way of performance.
The feed line (42" plus the 6" jumper between the blocks) from the battery to the fuse blocks (There are now two of them on the firewall...) is 10 gauge with each headlight relay being fed by a dedicated (one 6", one 7") 14 gauge wire.
There are
two relays, one for each "beam". These relays sit in a box on the firewall. For those that remember what my firewall looks like, the single relay that was there has been relocated into the relay box along with the two new relays for the headlights. This was done to tidy the firewall....
The 16 gauge wires that run to the headlight connectors are not solderless. In fact, every connection made was soldered as I have a "thing" about solderless connections.... Personal thing, have seen too many of them fail in the field.
To quantify the ground situation, I disconnected the headlights from the harness and Ohmed the
connector pin to the
battery ground post. I found 0.04 Ohms resistance. My leads have 0.02 Ohms resistance in and of themselves so the actual displayed value was 0.06 Ohms.
Side note: To eliminate a variable, the same leads were used with both meters. A small point to be sure but..... I do tend to strain at gnats...
Beginning Voltage:
Battery = 12.16Vdc
Headlight = 8.5Vdc.
This is an initial voltage loss of 30.09% an improvement should be rather easy.
Ending Voltage:
Battery = 12.16Vdc
Headlight = 12.09Vdc.
This is a final voltage loss of 0.57% an improvement to be sure.
Wire gauge selection was assisted by this calculator:
http://www.bulkwire.com/wireresistance.asp
To be sure, I had already selected the wire size based upon experience but, I never trust my experience, I always double check. Very Old Habit from when I was a working Engineer....
BTW, me being me, I polished the pins at the headlights to improve connectivity. As it turned out, it was a somewhat wasted effort as the performance did not improve. But then again, these bulbs are only 5 years old...
Once the LED headlights go in the beam dedicated fuses will be replaced. The factory headlight fuse (10 Amp) has been replaced with a 5 amp fuse as all it is now protecting is the relay coil wiring and the relay coils themselves.
As for attempting to improve this result, I fail to see how it can be done. The total system loss at the headlights is now a whopping 0.07 Vdc... For all intents and purposes, there is no loss.
I did not purchase a harness, I built mine. Again, a personal thing as I just hate to buy what I can build myself. Was it less expensive than the commercial harness? No, it was not. It cost way more but I feel the investment is justified fro not other reason than it forced me to think it out and build it. At my age/health condition any sort of build event that I can pull off has value...............................................