Conclusion #1 
Buried radials are not tuned. Burying them detunes them significantly. Don't bother making them a length based on the frequency.  

Conclusion #2 
Always use a balun. It should be a 1:1 current balun unless a different ratio is required to match the rig to an antenna with an impedance different than 36 to 72 Ohms.  
Conclusion #3 
If your SWR (without the antenna tuner in the circuit) is not 1.4:1, something is WRONG. A vertical antenna has 36 Ohms of impedance (radiation resistance). Your rig and coax have 50 Ohms of impedance. 50 / 36 = 1.4, which is an SWR of 1.4 to 1. This is also true if you are using a dipole. A dipole has an impedance (radiation resistance) of 72 Ohms. Your rig and coax have an impedance of 50 Ohms. The SWR is 72 / 50 = 1.4, which is 1.4 to 1 SWR. If the SWR is less or more than 1.4 to 1, you have a problem. You can probably live with the problem, but you need to know the antenna is not as good as it can be.  
Conclusion #4 
Always use an antenna tuner when operating. Use it to get the lowest SWR you can. Please read the SWR section in this web site.  
Conclusion #5 
You need lots of radials. To keep the efficiency of the antenna high, which reduces losses, and stops keeping worms warm, you need to use at least (an ugly minimum of) 12 radials that are slightly more than a quarter wave long at the lowest frequency you will be using. 50 radials is not too many. Avoid keeping worms warm.  
Conclusion #6 
If you are forced into a situation where only 1 radial is used, the pattern will not have a big null in it. It will be essentially circular just like a normal vertical antenna. The pattern may have a 1 dB difference, but that is hardly noticable by you or the recieving station. (Got EZNEC ? Try it and see for yourself.) A 1 radial vertical antenna is very inefficient, with lots of dB loss compaired to a multiradial antenna, but the pattern is still nearly circular. It is this kind of antenna that is "weak in all directions." 
Actually, you might want to know the value of efficiency VS. decibels so you can make a better estimation of how many radials you really need. This information comes from the same two graphs at the Stepper web site on another page in this web site.
Efficiency  Loss in dB  # of Radials  Wavelength  SWR 













































