I assembled a bird feeder from Wild Birds Unlimited for an elderly friend yesterday, It was an iron stand that went in the ground with arched arms from which a couple of birdfeeders hang. There was one 4 foot section of pole, which served as the support of a longer pole.. the 4 foot section goes into the ground via a coiled steel wrapped on one end, and then insert a screwdriver through a hole in the pole to give you a grip for spinning it into the ground.. It was quick and easy to drive the pole.
It instantly occurred to me the same proccess would be perfect to install a ground rod.. expecially if you just wanted a temporary one, because it come back out as easily as it goes in. It seems surprising gound rods arn't sold with this coil at the end - at least I cant find any
It is a good idea, Rich P., and I hope you or another reader of this thread consider suggesting it to the ground rod factory.
When I sunk a ground rod for carrier current broadcasting I hired my brick and mortar guy to pound it into the dirt with a fat hammer. It didn't look easy.
Oh, and, thanks on behalf of the bird community.
Well.. a google for "screw in ground rod" provides a 6 foot rod with a T handle on top, but the prices range from $139-$179 each - same rod, just different companies. Sounds a bit overpriced to me
Here is a link to a screw in anchor for a 4X4 post at only $31. Maybe it could be adapted for use as a ground.
Click on the "Installation" picture for a shot of the screw. Can't really see how long it is.
Here is a 4 foot screw anchor for only $9.
Looking at these various "screw-in" and "anchor" products, it seems perhaps that some of them might also serve in designing:
- 3-Meter Tower Structures;
- Guy-wire anchors for holding up towers.
Keep the ideas flowing!
l saw a few of the 4x4 post mounts but didn't mention them because the overall legth is only 24", with the post holder portion of it to remain above ground, so it looked to be only about 18" of grounding.
The 4 foot anchor might be a little more promsing though it's "Ideal for use in Softer Soils". I wonder if maybe if such a contraption as shown on the end could be attached to a heavier, longer rod by drilling a hole througn the rod to secure it??,, Just pondering.
Ground rods are good for lightning protection, not for enhancing a transmitter's signal output. Electrical ground rods are typically 8 ft. long. A screw-in thing is likely to be much shorter than 8 ft. and will probably not be galvanized or copper clad like the electrical ground rods. The earth in my area has a large percentage of clay and shale rock. When I dig down, there are numerous small rocks in the 2" to 4" and above size. Even one of those would stop a screw in rod dead in its tracks.
For even slightly rocky soil it's best to rent a medium size hand held electric "jackhammer" also known a rotary hammers, impact hammers, and demolition hammers. They have a chuck attachment for driving ground rods. Rental price should be around $50 to $75 a day, less for a partial day. Check equipment rental places. Home Depot stores have a rental department. Do a YouTube search on how to drive ground rods to see these things in action. One ground rod is sufficient for lightning protection.
After you drive the ground rod, BE SURE to install radial wires. They are absolutely necessary because a ground rod performs very poorly as an RF ground. I cringe every time I read about an installation with the transmitter connected to just a ground rod. It's not that hard to lay out a minimal radial system with 4 10ft wires. 16 20ft or 30ft wires is still not that hard. Use insulated wire, 18 to 22 gauge, laid on top of the earth, and fasten them down with plastic landscape staples. Over a growing season, the grass roots will bury them and they will disappear.
Ground rods are good for lightning protection, not for enhancing a transmitter's signal output
I thought it was also good for enhancing the signals range.. better ground propagation, better connection to moisture in the earth... I recall diagrams suggesting the more ground rods at the end of each ground radial the better..
?
Back in April, member Rich posted some graphs comparing a single ground rod to a system of radial wires: http://www.part15.us/forum/part15-forums/rf-ground-loss-antenna-system-efficiency. The graph for a single ground rod is in post #1 and the graph for a radial system is shown in post #4.
There is a dramatic decrease in ground resistance when radials are used compared to just a ground rod. In a region with poor conductivity of 2 mW/m, the single-ground-rod ground resistance is 150 ohms compared to just 5.4 ohms for the radial system, and in a region with good conductivity of 15 mS/m, the ground rod yields 25 ohms and the radial system yields 1 ohm. In both cases the reduction of ground resistance with a radial system is about 25 times.
To put these differences in perspective, consider the fact that the radiation resistance of a 3 meter antenna at the high end of the band is about 0.1 ohms (the power dissipated in the radiation resistance is the actual radiated power), and the radiation resistance is in series with the much higher ground resistance. So, much more of your precious 100 mW is being wasted in the ground resistance than the power actually radiated. Reducing the ground resistance by 25 times with a radial system will increase the radiated power by 25 times.
There is a lot of information on the web about radial systems for vertical antennas. Here is one that isn't too technical, but still provides some good information to help select an appropriate number of radials and lengths. http://www.steppir.com/wp-content/uploads/2013/07/Radial-Systems-for-Verticals-Rev2.1.pdf
Sorry RichPowers, I didn't comment about putting ground rods at the end of radials. I can't really comment on the effectiveness of doing that. My gut feeling is that it won't help. The RF current in each radial starts at a maximum at the transmitter end and tapers to zero at the end, so a ground rod at the end would not have any current to conduct to ground. However, the end is a maximum RF voltage point and I have read that for very high power transmitters, that can cause a safety hazard. So possibly a ground rod at the end would reduce the hazard. Rich Fry is the only one here that has the expensive version on NEC software that can accurately model buried wires and ground rods. Perhaps he can shed some light on this subject.
Below is a NEC4.2 study of this, for the conditions shown.
The r-f currents flowing in the earth near a vertical monopole as a result of its radiation penetrate the earth to a "skin depth" that depends on operating frequency and earth conductivity at/near the antenna site. Skin depth is defined as that distance where the amplitude of the wave is reduced to 37% of its value at the surface of a conductor.
For the conditions in this NEC study, one skin depth is about 6 meters.
So there is a slight benefit in adding ground rods to this system in that they are able to collect a small amount of additional r-f current to add to what is collected by the buried radials.
Adding these 8 ground rods reduced the real term of the load resistance of the antenna system from 22.7 ohms to 21.8 ohms.
Other things equal, the field improvement at 1 km when adding the ground rods is about 2.4% -- which doesn't have much practical value to a radio receiver (YMMV).
In my mind the slight benefit of adding ground rods to the far end of radial wires as Rich's analysis shows, doesn't warrant the physical work involved with driving the 8 additional ground rods. To get the same result, you can add a few more radial wires. That's a lot easier.
On further study, I found that the results for the three studies posted by Rich may be confusing or misleading to some when we talk about just the Ground Resistance (Rg) of vertical antenna systems. Rg is a function of the antenna/ground configuration and is totally independent of the transmitter and external base loading coil (if used).
Rich's first graph for a single-ground rod at http://www.part15.us/forum/part15-forums/rf-ground-loss-antenna-system-efficiency shows Rg + Rr. That's OK because Rr=.1 is negligible. His second graph at http://www.part15.us/comment/31793#comment-31793 shows just Rg. In Rich's latest post in this thread concerning adding ground rods to the far ends of radial wire, he shows the "load resistance of the antenna system" as being 21.8 ohms and 22.7 ohms with and without the added ground rods, respectively. I believe he is including the coil resistance in this one. So the 20 ohms coil resistance should be subtracted from these values to get Rg + Rr. That would result in 1.8 ohms and 2.7 ohms, respectively.
Comment Rich?
Rg is a function of the antenna/ground configuration and is totally independent of the transmitter and external base loading coil (if used).
Rg is independent of the R of a loading coil no matter where that coil is physically located.
... That would result in 1.8 ohms and 2.7 ohms...
Correct. I used the notation and values of "load R" to highlight the reason why the field at 1 km changed by only 2.4% when Rg changed by 67%.
The NEC4 results posted earlier in this thread assumed the resistance of the loading coil to be 20 ohms. This value is an estimate of the a-c resistance toward the upper end of the AM broadcast band of the internal, ferrite-core loading coils included in some Part 15 AM transmitters, based on private communications I have had with those who measured it.
If loading coil loss could be reduced, then radiated power would increase (other things equal).
The coil needed at the base of the 3-m antenna described in the NEC model to produce resonance at 1.65 MHz needs have an inductance of about 326 µH. Using the coil calculator linked below, it was found that about 116 turns of 16AWG copper-annealed wire wound in a helical form of 75mm OD and 200mm length has an inductance of about 326 µH at 1.65 MHz. Its series resistance at 1.65 MHz is about 5.7 ohms. Probably other coil designs could be found giving lower a-c resistance while producing 326 µH inductance, but I didn't try very hard to find one.
Returning to my NEC model with no ground rods, I changed the R of the loading coil to 5.7 ohms. The maximum field at 1 km went from 178 µV/m to 276 µV/m. This is an indication that the r-f power radiated by the system increased by a factor of 2.4 (240%). That is a worthwhile improvement.
Higher radiated fields would result if the internal, ferrite-core loading coil in a Part 15 AM transmitter was bypassed, and the system used an external loading coil such as described above.
BUT, the r-f bandwidth of the system using a low-loss loading coil is reduced, and system tuning and tuning stability become more critical. The 3dB r-f bandwidth of the NEC system model using no ground rods and the 5.7-ohm loading coil is about 4.2 kHz, which means that the audio response of a perfect AM receiver is 3dB down at 2.1 kHz. That response is not even as good as some telephone circuits.
So a tradeoff is required between "hi-fi" audio and producing a higher field at, and within a few kilohertz of the carrier frequency, at a given distance, as well as how touchy the system is to set up and maintain at maximum performance.
http://hamwaves.com/antennas/inductance.html
The coil I use for base loading has a nominal inductance of 260 uH. It is 59 turns of #16 enamel copper wire close wound for 55 turns with a 4 turn trim coil in series which can be slid axially to trim the coil to resonance when installed in the antenna system. The dimensions are 76mm D by 70mm L.
It was found by experiment that 260 uH was required to resonate the antenna system at 1670 kHz and the coil was designed for this inductance while maintaining a L to D ratio near 1. A bench measurement of the coil Q using Ben Tongue's method using an air variable capacitor to resonate at 1670 kHz showed a Q of 147 which gave a calculated coil RF resistance (Rc) of 18.6 ohms.
What is striking is the calculated Rc for the coil in Rich's example above which can call to question both my measurement and the calculation cited. My coil is different in dimension and inductance but being shorter and having less inductance with fewer turns I would expect the Rc to be smaller than that of the coil used in the example.
An attempt to use the linked calculator with my coil parameters to check this was thwarted by my not knowing some of the required parameters. Rich, could you share these so I can run the calculation for my coil and compare what I measured with the calculated values?
It would be great to get a Rc as low as 5 ohms but I am skeptical.
Neil