Lots of hams are going to install radials in a grass yard area at their home, so this is how I "do grass". (Pun intended!)
You will need a plot plan for your yard in order to make a sound plan.
I started with a 15 radials plan, went to 21 and was able to fit in 13 more for a total of 34 radials, each time adding the new radials to the plot.
In the interim years, the radial wires have "magically" drifted down to about 3 inches.
Guessing ~900' of #16 THHN wire, the shortest `9', the longest `36'. Most common length was `24'.
Radial Plate from DXE. Plate end of wires have lugs soldered to them, far ends of wire dabbed with Liquid Electrical Tape.
The Red lines point to the fiducial point so I can realign the cardboard template to the property marker (chip).
Here are 3 photos; Initial 15 Radial Plan, Radial Tool Set and Radial Birdseye View
 |
|
Initial 15 Radial Plan ~2007 |
 |
|
Electric String Trimmer and Template. Note there are 20 Radials on the template ~2010 Note Arrow to Concrete "Chip", Circle to insert Re-Bar to hold alignment. |
 |
|
Over the course of time I boosted to 34 radials, those grass slits disappear in ~1 week. Year ~2013 |
Note:
1- Some time ago I read an article by L.B. Cebik W4RNL (SK), a very highly regarded antenna analyst, where he commented that he had done modeling using Nec-4 to evaluate below ground and also for radials just slightly above the ground surface. He concluded that there was not a significant performance shortcoming for surface radials as long as there was a fair number of them. I liked that answer, so I merely scratch the surface of the earth where the grass root meets the earth. What's good for Cebik is good for me!
2- I use the installed radial lengths in my NEC-2d modeling.
3- NCJ had an article in the 2004 March April issue dealing with radials and it's worth reviewing.