Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!casbah.acns.nwu.edu!accuvax.nwu.edu!nucsrl!telecom-request From: asuvax!godzilla!dalyb@ncar.ucar.edu (Brian Daly) Newsgroups: comp.dcom.telecom Subject: Re: Information Needed About Cellular Antennas Message-ID: <15517@accuvax.nwu.edu> Date: 18 Dec 90 17:14:02 GMT Sender: news@accuvax.nwu.edu Organization: gte Lines: 110 Approved: Telecom@eecs.nwu.edu X-Submissions-To: telecom@eecs.nwu.edu X-Administrivia-To: telecom-request@eecs.nwu.edu X-Telecom-Digest: Volume 10, Issue 891, Message 1 of 7 In article <15427@accuvax.nwu.edu>, srm@dimacs.rutgers.edu (Scott R. Myers) writes: > I need an education on antennas. I'm purchasing a transportable and > want to install an antenna on my car for greater signal strength when > driving. I have a fair understanding of the different types (roof > mount, trunk mount, glass mount, etc.). What I want to know is the > layman's explanation of Db's. I see ads for 3db trunk mounts and 5db > glass mounts. The impression I'm given is the 5db glass mounts are > very acceptable for signal strength. How do they compare to roof and > trunk mounts? Just what are Db's anyhow? > I have read about the cabling > used in antenna installs (RG-58). Based on what I'm reading, there is > significant signal loss with that type of coax run long distances (ie. > trunk to front passenger side seat.). Would using another guage of > coax reduce the amount of signal loss from the cable? First, let's tackle the dB question: Say I have two radio transmitters. Transmitter A has an output power of ten watts, and transmitter B has an output power of one watt. One way of representing the ratio of transmitter A 's power to transmitter B's is using the decibel. The decibel is defined as: db = 10 log (P2/P1) So, in my example, the ratio of transmitter A's power to B's is: db = 10 log (10/1) = 10 db (note that the log is to the base 10) So, a 10db ratio represents a 10:1 power ratio. Now suppose transmitter A had a power output of two watts; now the db power ratio is 10 log (2/1) or 3db. thus a 2:1 power ratio represents a 3db power ratio. If I do this for other values of transmitter A power, I find that every time I double transmitter A's power (keeping B constant), I increase the db ratio by 3 db. Thus, doubling the power is the same as a 3db increase. Now, how does this apply to antennas? A measure of antenna efficiency is the gain (which is a power quantity). To identify the gain of a particular antenna, you have to have something to compare it to; as the above example demonstrated, the db is a measure of power ratios. Usually we refer to the gain of an antenna as compared to an isotropic source (an isotropic source is a lossless antenna that radiates power uniformly in all directions). The isotropic source has a unity power gain, or 0db. This is the reference. Looking at the two antennas you mentioned, a 3 db trunk mount and a 5 db window mount -- these measurements are against the same reference. From the above discussion, the 3 db antenna has twice the gain as the reference, and the 5 db has 3.16 times the gain. How are db's used? Let's say your transmitter (cellular phone) has a 5db power output, and you have 2 db loss in the coax running from the phone to the antenna. With the 3 db antenna, your overall system performance is: 5db - 2db + 3 db = 6 db. With the 5 db antenna: 5 db - 2 db + 5 db = 8 db. So, keeping everything else constant, the 5 db antenna will give better performance than the 3 db. Another important factor with the antenna is the placement of the antenna on the car. This will have an effect on the antenna pattern -- the pattern is a "picture" of how the electromagnetic energy is transmitted. An antenna will not radiate uniformly in all directions; the antenna might transmit better in one direction, and worse in another. This pattern is affected by the metal body of a car. To get the best pattern, I've usually had success with placing the antenna directly in the center of the roof. Placing the antenna on the rear window or trunk will change the pattern. On the subject of coax, you need 50 ohm cable for your cellular system. RG58 is by far the most common, and least expensive. It has an attenuation of about 20db per 100 feet, which is not great. However, the distance from the trunk to the front passanger side of a car is less that ten to fifteen feet, so you should be OK. There are better cables available, but these are usually larger in diameter which might not be good for automotive installations. I'd recommend RG-58A/U (specify this type -- it contains not only a copper braid, but an aluminum foil shield). At 900MHz, it has a loss of 13.8 db per 100 feet. There is one possible flaw in this however. I assume that both antennas in question were measured against an isotropic source. However, as one of my colleagues here at AGCS pointed out, you need to make certain of that fact.... > Looking at the two antennas you mentioned, a 3 db trunk mount and a > 5 db window mount -- these measurements are against the same > reference. From the above discussion, the 3 db antenna has twice the > gain as the reference, and the 5 db has 3.16 times the gain. Ah, but therein lies the rub - sometimes these AREN'T measured against the same reference. Sometimes dBi (dB gain relative to an isotropic point source) are used, and sometimes it's dBd (dB gain relative to a dipole at that frequency). I believe a dipole has about 1 dB gain over a point source. So it's more attractive for manufacturers to quote dBi if others quote dBd. I've seen this with ham and CB antennas, don't know if the cellular folks are doing it or not. Thus, when you read antenna specifications, make certain you are comparing apples to apples! Brian K. Daly WB7OML @ AG Communication Systems, Phoenix, Arizona UUCP: {...!ames!ncar!noao!asuvax | uunet!zardoz!hrc | att}!gtephx!dalyb Phone: (602) 582-7644 FAX: (602) 582-7111