[TowerTalk] Ground Measurements

ROBERT A. WANDERER bobwanderer@polyphaser.com
Thu, 17 Apr 97 10:47:22


     I would like to add a few comments (and acorrection) to the 
     posting by Steve Z KJ7CH. Pretty good job going from memory!
     Any responses should come directly to me as I'm not a subscriber 
     to this reflector, but it looks like I ought to be!
     
     What Steve is describing is called the 3-stake ground resistance 
     measurement. There is also the 4-stake measurement, which is used 
     to ascertain the depth at which the first conductive soil layer 
     occurs and its value. This obviously is done prior to installing 
     the ground system!
     
     Please refer to the drawing in the original posting. E1 is the 
     electrode under test and it should be the last (furthest out) 
     ground rod in the system, otherwise the sphere of influence and 
     presence of the ground system will give you erroneous readings 
     which will usually be on the low side. Hopefully you are not 
     dealing with buried utility lines in the vicinity, as they will 
     affect the veracity of your readings too. Without going into it, 
     the ground testers I'm familiar with have methodologies to 
     minimize the impact of this situation.
     
     E2 and E3 are NOT part of the ground system under test. They are 
     the probes provided by the earth tester manufacturer. One 
     manufacturer is J W Biddle of Blue Bell, Pa and their product is 
     trademarked as the Megger. E3 is placed 100' past E1 and E2 is 
     placed about 62' past E1. Again, both of these probes (and their 
     connecting cable) should go out past the ground system so the 
     readings aren't compromised. Make a reading. Move E2 a few feet 
     either side of the initial point and make a reading. Go both a 
     few feet closer and further from E1. Make several readings. When 
     the readings start significantly departing (both higher [as E2 
     approaches E3] and lower [as E2 approaches E1]) from the initial 
     readings, you are finished. The readings (you can average them) 
     obtained prior to the marked increase/decrease is the value of 
     your ground system.
     
     The Megger uses 100 Hz. As previously mentioned, there is NO 
     correlation between this reading and the impedance of the ground 
     system at the "important" lightning frequencies (10-100 kHz and 
     up to about 1 MHz; energy components exist into the GHz range, 
     although once past the 50 MHz point they are pretty low in 
     level--but the idea is to provide a low impedance path to ground 
     for ALL the energy). As a general rule, a ground system measuring 
     5-Ohms or less should do a good job if a low-inductance ground 
     was installed. As a real-life example (and without revealing the 
     customer), I did a site survey where the ground measured less 
     than 5-Ohms, but equipment damage was still occurring. The ground 
     grid was composed of #6AWG wire. In a nutshell, this wire was 
     just too inductive at lightning frequencies to be a good ground 
     despite the low measurement. If wire must be used, #00AWG 
     (commonly written as 2/0) is the minimum although you can "get 
     away" with #2AWG. However, 1.5"-wide copper strap of 26-ga 
     thickness (0.0159") has even less inductance than 2/0. In 
     addition, it has much more surface area. 
     
     If I can leave you with two buzzwords (concepts) that need to be 
     taken into account when dealing with grounding, they would be 
     "inductance" (impedance) and "surface area." The ground system 
     must have low inductance (impedance) to work properly and its 
     surface area must equal or (preferably) exceed the circumferences 
     of ALL the cables, waveguides, etc. going to it. It is not 
     uncommon (and I saw it at that aforementioned site) to see 
     several 1-7/8" Heliax(TM) cables going to a ground plate and 
     having a #6 (and if we're "lucky", a #2) wire going to the ground 
     system. Once again, the beauty of copper strap is (1) low 
     inductance and (2) surface area, as both sides count! So a 
     1.5"-wide strap  has 3" of surface area.
     
     The voltage drop formula to keep in mind is Vd=L (di/dt). For 
     those who have forgotten, di/dt is the change in current over the 
     corresponding change in time. Using average values, di would be 
     18,000 Amperes and dt would be 2 microseconds. A typical 150' 
     tower would have a L of 40 microHenries. Doing the math, there 
     would be 360 kV between the top of the tower and the bottom! 
     
     I hope I haven't made this more complex than necessary. In 
     actuality, describing the process takes longer than actually 
     doing it.
     
     Of course, when all is said and done, take the most important 
     step of all and install a #1 Iron golf club at the top of your 
     tower. As Lee Trevino once said, "Not even God can hit a #1 
     Iron!" If your neighbors are unhappy with your tower(s), perhaps 
     you can point out that you are performing a public service since 
     the lightning strikes will be more likely to hit your tower than 
     their home, and they should be grateful rather than resentful of 
     their presence. :-)
     
     Hope this helps!
     
     Sincerely and 73,
     Bob Wanderer AA0CY
     Senior Applications Engineer
     PolyPhaser Corporation

--
FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com