> Edward W. Sleight wrote:
> > Yes, the meter is a Fluke model 7 and it's ok. As I said Brian, it
> > seemed like a unique opportunity to watch "something develop" as the
> > process goes on.
>
Peter VE6YC wrote:
> Sorry but a Fluke model 7 won't cut it for measuring ground
> resistances. You need a "Megger" Although they don't put out
> 50,000 volts, they do run 500 to 1K volts.
>
[K0iL] I have to agree with Peter on this. Measuring the "DC
resistance" of the tower's concrete to ground "system" tells you very
little about how it will react to a lightning strike which has both AC &
DC components or even to RF which as AC. Therefore the system will have
an impedance to lightning as well as RF that is different from the
DC-only resistance. Compare the DC resistance of coaxial cable to its
50 ohm impedance.
A megger is the best device to use to test a ground system
because it supplies a high di/dt pulse. PolyPhaser's book on grounding
(The GLEP book) refers to the megger as a glorified ohmmeter (w/voltage
source), but I believe a specially made megger (EHV) is what they use to
test all of their equipment. (There used to be a guy on this list who
worked for them? You out there Bob?)
Also, the "Interference Handbook" by Bill Nelson(WA6FQG) & Bill
Orr(W6SAI) on pgs 199-201 gives two methods for measuring ground
systems. The first method is meggering of course! But most hams don't
have access to a megger so they suggest another method using three test
points (main point A, B, & C) each using "rods" the same depth and 20ft
apart from each other to avoid interaction between test points. In this
case one "rod" could be the concrete base.
I'm not sure how valid this method is since I feel it introduces
some error into the total measurement compared to meggering and only
uses DC resistance readings, but I'm presenting it here to give another
option that could be used w/o a megger.
Using a high quality VOM, six readings are taken between the
"rods" or test points: A-B(93ohms), B-A(67), B-C(83), C-B(113),
A-C(103), and C-A(71). The averages between each set of readings are
computed, i.e. A-B and B-A are averaged. So the result is three
averages written as so: A+B=80, A+C=87, & B+C=98. Here's where it gets
kinda dubious because he treats the averages as a summations. He solves
the averages results as equations: (A+B)+(A+C)=167 or, 2A+B+C=167. Then
subtracting the third equation:
2A+(B+C)=167
- (B+C)= 98
-------------
2A = 69
And therefore, A = 34.5 ohms (this seems high, but is an
example only!)
You could substitute A into the above equations and solve for B.
Then solve for C using the A or B values in one of the original average
"equations".
One thing the second method shows is that taking a single DC
resistance reading between only two points in only one direction is
probably not going give the desired result. I'd really like to see a
comparison done between these two test methods sometime, but I don't
have the time or opportunity right now. Maybe the ARRL Lab guys could
take a crack at it, or maybe someone on this list with some time on
their hands?
73,
de ed -K0iL
k0il@qsl.net
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|