10/26/09

The AntennasDirect DB-8 8-Bay

42.6” wide, 28.5” high, 5” deep

Antenna weight: 6 lb.

All the 8-bay makers seem to be copying each other’s mistakes. The DB-8 has some of the same harness errors as the 4228HD. Replacing the harness with two baluns and a combiner would make it the same as the old DB-8. But the DB-8 is not fixable. The DB-8 dipole elements are only 6.2 inches long, compared to 8.0 inches for the 4228. This biases the DB-8 toward the higher channels. The DB-8 was always a bit weak below channel 40. Now that channels above 51 are gone, it is no longer a reasonable antenna, even with the harness fixed. Some day AntennasDirect will figure out that they have to rescale this antenna.

DB8 Net Gain:

Plot A is the new DB8. The harness and balun are included in the simulation.

Plot B is the old DB8, but excludes the losses in the combiner and baluns. The author has not measured these losses. Combiners often lose 1 or 2 dB at the highest channels, and transformer balun losses are similar. So this plot overstates the gain of the old DB8 by maybe 3 dB at the high end

Plot C is the raw gain of the DB8 with the combiner and balun removed. The net gain would be close to this if the combiner and balun were low loss devices.

The simulations show a problem below channel 20 for the DB-2, DB-4, and DB-8. This is the result of a resonance in the vees of each dipole. (The wire of each vee, if straightened out, is 12.4 inches long. This is a half-wave at 490 MHz, and results in a standing wave at this frequency. In other words, each leg of the vee has the opposite current from the other leg when the antenna is viewed from the front. This results in much radiation in the vertical direction, which saps the gain in the forward direction.) These dipoles are simply too short.

The right and left dipoles are set farther apart (by 3.4 inches) on the DB-8, compared to the 4228. This results in larger minor lobes in the azimuth (view from overhead) pattern, but it also narrows the main lobe. As a result the DB-8 is probably better at rejecting multi-path than the 4228. The major null is at about 25° for the DB-8, 30° for the 4228. The following table shows the beam widths (measured to the half-power points):

ch 20: ch 30: ch 40: ch 50: ch 60: average:

DB-8 25.3° 23.2° 21.0° 19.0° 17.7° 21.2°

4228 32.2° 26.2° 25.7° 22.3° 19.4° 25.2°

DAT-75 44.3° 39.2° 34.8° 30.5° 26.6° 35.0°

AntennasDirect calls the DB-2, DB-4, and DB-8 “Multi-directional antennas”. There is good justification for calling the DB-2 and DB-4 multi-directional, but the DB-8 is one of the most directional antennas you can buy. AntennasDirect is simply flat wrong about this. You must keep in mind that an 8-bay is very different from its kin.

High gain antennas like this one are big, hard to aim, and hard to keep up in bad weather. The author does not normally recommend this antenna inside 25 miles. But there are two exceptions to this:

1. A very high gain antenna might be necessary behind a hill or in a valley.

2. When an obstruction is causing multi-path, a very directional antenna such as this one can reject signal from the wrong directions.

The DB8 balun

The diagram comes from the ARRL Antenna Book. The DB8 balun is the same except the U-section has been divided into two quarter-wave transformers having characteristic impedances of 54 and 75 ohms, velocity factor 0.60. It will match 200 ohms with 75 ohms, making it a 2.66:1 balun.

This balun is a good impedance transformer but not a very good balun. A balun is supposed to block unbalanced currents. This one creates them. Also it does not prevent radiation from the coaxial shield. A balun of the 4228HD general type would have been a little better.

This balun will mostly filter out VHF, so don’t even think of buying this antenna for VHF.

This page is part of “An HDTV Primer”, which starts at www.hdtvprimer.com