Continuous copper wire to earth ground

Why waste time and money on cadweld? When you can use a barrel splice and acrimp tool.

btw cadweld contains cadmium hense the name. very toxic fumes

you create future fail spots, the bend stresses the metal and has a good chance of it cracking or similiar in the future. cracked wires overheat and cause fires

neatness only counts if it doesnt cause troubles.

as for the OP I would run a new ground line thru the clamp at the copper line, jump out the meter even though it doesnt matter, ideally the OLD water line could of been left in the ground disconnected at both ends from water but still used for grounding.......

one thing the OP may find it very hard to drive the ground rods, standing on ladder with 8 foot rod waving in breeze.

if the ground is hard it might be easier to have a electrician do it one with proper driver for ground rods

Bends in wires that have low frequency or DC will not have a noticable effect in most cases. When you get to higher frequencies it will. A lightning strike is a very steep pulse. It will not follow the bend in the wire. It may jump off at a bend and go to something else. Also it will act like an inductor in the wire and in effect you are disconnecting the wire at the bend.

Sharp bends in wire may or may not be bad as far as the physical routing goes.

You obviously haven't been advised properly. LIGHTNING does not turn corners well. It will NOT follow the copper if it has sharp bends. The best way to protect a piece of equipment from lightning is to tie a knot in the cord. You'll also notice on houses with lightning rods, the cable coming from them will have nice big easy bends where it comes around the gutters and down the side. Lightning will blow right out the side of a conductor that has too tight a bend in it. NOW, back on the topic of the ground cable in an electrical panel, the tight bends don't matter, because that ground is not for lighting protection anyway.

A steel post driver works great down to the last 2 feet.

Fixing the ground will not fix your erratic voltages at the receptacles. That is a function of the neutral wire going back to the utility

I don't think that would satisfy the NEC.

I've used a torch on the shrink tubing that is used to cover UF splicers all the time. It seems to work fine and I haven't had any call backs because of it.

Would you care to say what section of the NEC it would violate.

... and, at some level, lightning can exhibit RF properties where bends CAN make a difference.

Why tempt the fates?

I agree you shouldn't bend it to tight like a near perfect 90 degrees but you don't need to bend it that much to get good looks. Bending copper work harden it and further bending operations can make it break, but exactly how many bends before it breaks. When you wrap a 12 gauge wire around a post, tighten down the screw, and bend the tail to make it break off, how many bends (back and forth motions) do you make? For me it is at least four each way at a very sharp angle. But when you install romex, you bend it and then how much additional bending do you do, and how sharp are the bends compared to a single stripped wire? My guess is that you would have to do a lot of very sharp bending to get the wires work hardened enough to crack.

Perhaps the worry is grounding for a lightning strike. At that kind of power level, the current can do all sorts of strange things. It might decide to spark across the gap as well as flowing around the corner.

OT: Bending radius is a significant concern for coaxial cable. Bending the wire deforms the outer shielding layer away from the desired cylinder. A sharp bend deforms the cylinder significantly, making one side closer than the other to the signal-carrying center conductor. This does "funky things" to the signal. Don't ask me to define that technical term. I'm doing well to remember my installation training at all. I remember the trainer putting a meter of some kind on the line and showing how the needle moved when the cable was bent severely.

All wire (and fiber optic cable) has a maximum bending radius. For romex, wire bend should never exceed what would go around a large fist. Sharper bends eventually cause insulation breakdown. Install wire to last 100 years. No sharp bends.

Reasons for no sharp bends for lightning conductors are different. Others have described symptoms. But the reason involves something electrical: higher wire impedance. Lightning protection requires a conductor to be lowest impedance. That means wire length must be shorter. Sharp bends (like inductors) increase wire impedance. High current through higher impedance will find other conductive paths - such as arcing into an adjacent wall.

Problems created by sharp bends and splices are why AC wall receptacle safety (equipment) grounds do not properly earth lightning; are not earth grounds.

Two reas> Why would sharp bends do anything? As long as the wire is continuous

Quote from (2000) NEC, Section 334.24:

"Bends in Types NM, NMC, and NMS cable (Romex) shall be made so that the cable will not be damaged. The radius of the curve of the inner edge of any bend during or after installation shall not be less than five times the diameter of the cable."

When the acceptance testing was done on type NM cable by Underwriters Laboratories; low these many years ago; the incidents of conductor insulation damage went up sharply at bend radii of 3 cable diameters. The US NEC requires bends of five cable diameter in radius as a safety margin. The problem isn't conductor fracture but rather that the insulation will fail over time if it is bent too sharply. Since there is a bare EGC run inside the jacket of type NM cable the possibility of arcing which can cause a fire is real.