|
TANT 1.2 (April 23, 2006)
Computes Antenna Temperature and G/T Ratio for elevations from 0 to 90 degrees in steps of 5 degrees. (Uniform Sky and Earth temperatures assumed.)
Installation
No installation is required. Place TANT.EXE file in the same folder with EZNEC pattern files and run it.
TANT is a DOS program and can be run from either the genuine DOS or from MS-DOS Prompt / Command Prompt. X87 coprocessor is required.
Using the program
1. Model your antenna in free space using EZNEC 3 or 4.
a. Maximum gain must occur at elevation = 0 and azimuth = 0.
b. In Options, Far Field Table Units: select dBi.
c. In Desc Options, Plot, Fields: select Total field for 3D.
d. Set Plot Type to 3D, and Step Size to 1, 2 or 3 degrees.
e. Click FF Tab button, wait, select Azimuth Slices, check Full Range.
f. Save the Far Field Pattern data to a folder using 8+3 DOS filename.
2. Run TANT from the same folder and follow instructions.
Reasons to use TANT instead of Yagi Analysis 3.54
* You've already done modelling in EZNEC 3 or 4.
* To cross-check results of YA 3.54 algorithm against those of NEC.
* To take advantage of EZNEC's greater accuracy, particularly when using double-precision along with the sufficient number of segments (the best match between YA 3.54 and EZNEC occurs with ~13 segments per element, which is hardly accurate enough).
* To analyse stacked antennas accurately (YA 3.54 does not compute the actual stacked antenna coupling, so the values and the pattern shown are but a hint).
* To analyse non-Yagi antennas.
* To analyse antennas having arbitrary asymmetric radiation patterns (no shortcuts taken in integration).
* To look at temperature and G/T values at elevations other than 30 degrees (YA 3.54 allows no choice of elevation).
Notes
* For introduction to antenna temperature and G/T ratio see, for example, the well known article by DJ9BV: "Effective noise temperature of 4 yagi arrays for 432 EME" http://www.mrs.bt.co.uk/dubus/8704-1.pdf (598k)
* Antenna temperature and G/T are not parameters of the antenna itself. Instead, they are parameters of the entire "antenna + ambient" system.
* Results for elevation of 30 degrees are shown in red for easier comparison with YA 3.54. This elevation was chosen by DJ9BV as a typical for EME. Of course, there is nothing magical about 30 degrees. Depending on application, some other elevation may be of more use to you.
* At 0 degrees elevation the Tpattern is always the average of Tsky and Tearth for any antenna having a symmetrical pattern relative to the horizontal plane, regardless of the actual pattern, gain, losses, F/B, F/S or sidelobe level.
* Tpattern = effective noise temperature considering the pattern alone. This is not exactly the same as having the same dimensions with zero loss conductors, as the current distribution and antenna pattern will change slightly when the loss is removed.
* Tpattern is exactly the same as TA' in DJ9BV's article (the sum of "forward", "sidelobe" and "backward" temperatures), which is likely NOT the same as the sum of similar-named values in YA 3.54.
* Ttotal = Tloss + (Tpattern attenuated by the loss), exactly the same as TA in DJ9BV's article.
* Whenever Tpattern > Tref (290 K), any antenna loss will LOWER the Ttotal, i.e. it will DECREASE the total noise power towards the thermal noise power at Tref. Of course, this cannot be used to improve G/T or S/N ratios.
* Please note that terms "forward", "sidelobe" and "backward" in DJ9BV's article (and likely in YA 3.54) have nothing to do with your antenna's pattern. They are just unfortunately chosen names for different parts of the integration surface, related to the 30 degrees elevation used, and having no relation whatsoever to the actual antenna pattern.
* Please note that there is an error in DJ9BV's article. In section 4 the integration surface is partitioned incorrectly to "forward", "sidelobe" and "backward" parts. As stated there, the partitioning is correct only for azimuth values of 0 and 180 degrees.
The noise temperature and G/T treatment in YA 3.54 seems to be based on DJ9BV's article. However, with our antenna samples YA 3.54 results were in substantial agreement with TANT results, although it is not clear whether it was so because the error was small in the samples tested, or because YA 3.54 performed the integration correctly.
* Note the average and maximum gain values on the top of the screen. These values must be the same as those displayed by EZNEC. Otherwise would indicate that pattern data has not been transferred correctly.
* Either "Compass Bearing" or "CCW From X Axis" may be selected for Angle Convention. In either case, the maximum gain MUST occur at elevation = 0 and azimuth = 0 for TANT to work as expected.
* You can keep TANT running while you work in EZNEC. When you're ready, save the pattern file and select Compute in TANT.
* Bear in mind that 0.1 K accuracy in Tpattern or Ttotal would demand less than 0.002 dB error in pattern, which cannot expected from NEC modelling. Even the resolution of 0.01 dB in the EZNEC pattern file is far better than the actual accuracy.
Acknowledgements
* To YU1CF for numerous suggestions, testing and providing test samples and EZNEC pattern files.
* To YT1NP for testing with EZNEC Demo versions.
* To YU7EF and YU7XL for most stimulating discussion leading to the development of TANT.
Disclaimer
* No guarantee whatsoever, use TANT at your own risk.
* This is freeware, use TANT for any purpose you like.
|
