On 12/28/2015 4:04 PM, John G wrote:
Don Y has brought this to us :

Do the *easy* things, first.
- measure at the *main* panel
- measure at the subpanel (already done)
- open the disconnect at the subpanel and remeasure at the main panel
Assuming this isolates the problem to the subpanel:
- remove individual loads from the subpanel (open breakers) and measure

Take good notes.

You should be able to diagnose the problem just from these observations.

It has taken an unbelievable amount of drivel and misinformation to get down to
this simple list of things to do to isolate the problem. :-?

I am basically lazy. : As I think most folks are. I would rather do
the easy things -- take data and THINK about the problem -- than run
around "try this", etc.

Occam's Razor: chances are, it's something in the subpanel/workshop
and NOT something with the utility -- *or* the main panel (not
counting the subpanel tie-in).

Eliminate the workshop and see if the problem persists. If not, tells
you the problem is *related* to the workshop (though may still manifest
in something else).

Far better to be able to tell an electrician or the utility: "I did
this and this is what I saw" than to just throw a bunch of unrelated
observations at them ("Well, did you try *this*?" "No, but I tried
something (totally unrelated)!")

We had a neighbor around the corner have his metercenter catch fire!
Faulty connection on one of the mains and it just arced it's way into
flames. Likewise, had a vault "explode" (coincidentally, near that
same home!).

I.e., when things start deviating from normal by too much, Bad Things
happen. "You have been warned" ;-)