You are getting carried away in wanting absolute definitions with absolute rules of installation and then mixing and matching situations.

-If radials are not in contact with the ground, they are elevated

-If the antenna is not in close proximity to the ground it is elevated

*If an antenna and/or the radials are raised 1 inch above ground does the performance of the antenna dramatically change?   No.
 
*Do rules about the number of radials required dramatically change?  No.  It's not cut and dried that this '1 inch elevated antenna' now only requires two resonant radials per band, whereas 1 inch lower and it need 36 radials for optimum performance. 
 
It's relative.  The lower the antenna system, the more it is affected by earth. The higher it is the more it performs like an antenna system in free space.

Why would you ground a Moxon antenna?

There is a lot of weight on that antenna.  Either hang it higher from the ends, or get a center support.

I think a center support is more practical and will cause less stress on the the hardware.

b.

Ideally each mast should have it's own ground system.  However, with the mast 2-3ft apart I think it is acceptable to tie them together.

-The ground wire should run directly by the shortest path to ground.  There should be a ground rod at that point.

-The ground wire should not be connected to any other metal parts such as the rain gutters.  -

-Cables going over a roof edge should be supported at a point immediately above and below the edge, with a sufficient loop to prevent contact or rubbing.  You may use commercial TV twinlead supports, or bracket or fabricate them from aluminum or wood with appropriate cable clamps.

-One major purpose of grounding wires is to keep lightning outside the house.  Do not run ground wires into the house.   The electricians and engineers have indicated in discussions that it is permissible to run 'Bonding wires' between rods through the structure if it significantly shortens the run.

-Your plywood wire support at the roof edge sounds fine. No need to make extra work replacing it with plexiglass.  After all there is not need to insulate it.  The gentle curve sounds good, as does your service/shack ground!

Good questions, you are on the right path!  bill

http://vk1od.net/calc/tl/tllc.php

Owen has a great line loss calculator.

Forget the antenna literature.  Since it didn't work for you location and situation let's get real.

Measuring antenna element angles down to five or ten degrees is not a common problem.  You are chasing minutia.

The basic problem is the antenna is too low.   Not just the center. The antenna as a whole.  Get the thing up 25 or 30 feet in the air as a minimum.   Get those drooping ends near level.  Letting them droop causes them to couple with earth and increase losses.

Basics review:  For 40m, 30ft elevation would be only 1/4wl.  The antenna will be essentially omnidirectional.   You won't see much directionality typical of a dipole on that band until you approach 1/2wl, or 20meters/60ft.

Stand up and take a walk in the sun, Bunky!

Sorry, but you are essentially re-inventing the wheel.

Hams have used multiconductor cable, like flat rotor control cable with the wires cut to different lengths as radials for years.  It works!

Radials is radials; let's leave the word 'fan' out the fight! 

I read the literature.  I just modeled it in EZNEC.   20m 1/4wl, 1 radial, elevated 30ft over Real Ground, it shows a slight cardioid pattern with about 5dB of F/B difference.  It's a lumpy cardioid pattern.   Bill

There are only a few choices, which should be obvious and they are the related harmonics lengths.   1/8, 1/2, 5/8, 3/4.  of 40m.

You don't say why you won't just replace the fallen antenna. but from the above choices 1/8wl would be the easiest to install.  

Congratulation on your antenna farm it's definitely one to be proud of!

p.s.  You have have made a very simple question confusing by including so much extraneous information.

Nope. 

I note the text says the Serial is 00050387, and the photo shows 0050287.  *shrug*

The photo shows "Made in Japan"  vice the claimed Made in USA/California

I wonder if the silver label wasn't taken from some other Kenwood product...

The ARRL Handbook, and Antenna Manual cover this topic and many!

I strongly encourage every beginning ham to buy a copy.  You will use them as references for years to come.

Welcome aboard.  Bill

Joining the chorus, but not saying you can't do, be aware that overhauling  an amplifier is jumping in to the deep end of the pool, if you have no prior repair experience.

-Get the manual
-Read a lot about repairing amps
-Talk to people.
-Find somebody local who is experienced with amps to help you!!!!!
-Get a thicker hide...people WILL disagree with you on boards.
-DMM with diode function
-capacitor checker (optional)
-Dummy load, appropriately rated (You need it anyway for testing and tuning up)

-Tubes can't be checked on tube testers, they draw too much power.  All you can do is a basic short check with DMM, then cross your fingers when turning on the amp.

-Caps.  One school of thought is run caps till they blow; that they can be reformed; that slow voltage bring up is best.  Other school is replace them if they appear old; turn the power on see if they hold.

-Power supply diodes and resistors should be checked.  Replacing with modern diodes is not  a bad idea.

-Resistors in the meter circuits should be checked, be precisely accurate or better replaced. 

-Check tube bases for overheating, cracked pins and springs.

-Check fuse for correct amperage, clean socket.

-Absolutely confirm that transformer is properly strapped for your voltage!!

-Check wafer/rotary switches for arcing, carbon tracks, corrosion,  clean with an alcohol swab.  DO NOT use any conductive sprays like DeOxit.

-Remove and measure parasitic suppressor resistor.  If drifted more than a few percent, replace with proper value 'non-inductive' resistor, e.g. carbon composite, etc.

-Check overall security, burn or scortched parts, bad solder joints, general cleanliness  (use a damp rag).

This is a good start and hits many of the key points.

Jeeeez, guys.  They aren't using your grandpa's Kraft gold box!

$40 Million drones on missions don't get controlled with model RC protocols over unencrypted frequencies, and no they don't use HF.  It's done from satellites with SHF.

2.  I'm not convinced the Iranians "captured" our UAV.  It think it crashed after losing the command link.  Their claim to manipulate the GPS and 'spoof' the craft to land in the wrong place, while theoretically possible is a far stretch technically. Also unlikely because the military uses an encrypted GPS mode, which is what GPS was designed for.