Paul Holm wrote: >I'm wondering how a person can estimate the amount of rejection that would >be required from a duplexer given the amount of power coming out of the >associated tx. I realize that this also is dependent upon the amount of >sideband noise on the tx and the sensitivity of the rx. I'm focusing mainly >on the 2M ham band using Bp/Br type duplexers and MastrII radios. Are there >any rules of thumb to determine the minimum amount of rejection required >when using 'x' amount of power? Thanks. >
I'm going to take a slightly different approach to answering your question, one that might help others that wonder about the magic numbers in general. Within the text are references to web pages to further help with the numbers. The examples below are strictly general in nature. Every situation is obviously different and the figures below aren't meant to be used as concrete testimonial to determine duplexer requirements, but they are a good rule of thumb considering a few types of equipment (like Hamtronics, GE, and Motorola solid state) that are commonly used in the world of building repeaters. The easiest way to to understand the numbers involved with power level (in transmitting) or signal level (in receiving), is to use a common reference. In two way radio, this reference is dBm - Decibels relative to 1 mW into 50 ohms. dBm is calculated by using the ratio of some power (expressed in mW) to 1 mW. For example, 1 mW is 0 dBm, 10 mW is +10 dBm and 100 W is +50 dBm. For more examples of power levels related to dBm, see (Ref 1) For this example, we'll consider a 2M repeater with 100 watts (+50 dBm) and a basic sensitivity of .35 uV (-116 dBm) for 12 dB SINAD. 12 dB SINAD is a fairly noisy but very copyable signal level. For more examples of signal level related to dBm, see (Ref 2) First, lets talk about the transmitter. In general, the transmitter must be fairly clean, not producing unwanted spurious, harmonic, or noise emissions on or around the desired carrier. The noise produced beside the carrier is referred to as 'transmitter sideband noise' or "phase noise". This noise *can* and does fall directly on the receiver frequency. In repeater type duplex operation, if this noise is not reduced in level, the receiver would be desensitized. Obviously, it needs to be reduced below the level in which the repeater receiver can hear. From the specifications of these radios, this type of unwanted emissions *generally* fall about 80 to 85 dBm below the level of the desired carrier frequency [transmitter operating frequency]. In the case of our 100 watt (+50 dBm) transmitter, these unwanted signals are at -35 dBm. Next, lets talk about the receiver. A typical repeater receiver without a preamp will receive a fairly noisy but copyable signal at about .35 uV or -116 dBm. Since the receiver will actually open squelch below this point, we must provide isolation that will reduce the noise of the transmitter at, or preferably below, this open squelch level. This is referred to as the Minimal Squelch Sensitivity. The minimal squelch sensitivity of a receiver will be several dB below the 12 dB SINAD point. In the case of the very sensitive Micor Squelch, this level can approach -130 dBm. From the example numbers we have shown that a 100 watt transmitter will produce unwanted noise on the repeater receiver frequency at a level of -35 dBm. It has also been shown that our receiver will open squelch at -130 dBm. The difference must be made up by the duplexer. In this case the duplexer must provide *at least* 95 dB of rejection for zero desense. If the transmitter is better than specification, the minimum required level of rejection would be less by the amount better than the specification, but one should never design right to the edge. This part of the example is one half of the function necessary for the duplexer to provide for. The other function of the duplexer is to reduce the shear power of the repeater transmitter to a level that won't desense the repeater receiver. In a repeater receiver, unwanted signals must be reduced to a level at or below which intermod will occur. If a strong signal off frequency is forced into a receiver, the preamplifier or mixer will be overdriven and the result will be desensitization or squeals heard along with the desired message. In a typical receiver, this level is 80 dB higher than its rated sensitivity. In the case of our repeater receiver with a sensitivity of -116 dBm, desense or intermod will occur at a level of -36 dBm or greater. From the example numbers we have shown that a 100 watt (+50 dBm) transmitter will need its shear power reduced to a level no greater than -36 dBm. The difference must be made up by the duplexer. In this case, the duplexer must provide *at least* 86 dB of rejection for no desense. If the receiver hears better than the specified receiver in this example (like by adding a preamp) one must afford additional rejection to insure the repeater will operate with no desense. To summarize, a *typical* 100 watt 2M repeater with a decent receiver will require a duplexer providing at least 95 dB of transmitter side-band noise rejection and at least 86 dB of shear power rejection to properly duplex. It should be noted that a well tuned and maintained transmitter will typically exhibit slightly better than the manufacturers specifications where sideband noise level is concerned. A typical tube type transmitter or power amplifier will exhibit much less transmitter sideband noise than it solid state counterpart, and will usually require less transmitter side-band noise rejection from the duplexer. A GE Mastr II PLL exciter will produce about 20 dB less transmitter side-band noise than it multiplier counterpart, so it requires at least 20 dB less transmitter sideband noise rejection from the duplexer. (Ref 3) From this you can also see why there are duplexers made for certain applications that don't provide the same amount of rejection on each side, and care must be taken to apply these types of duplexer in other applications. Ref 1 - <http://www.minicircuits.com/dg03-110.pdf> Ref 2 - <http://www.repeater-builder.com/pdf/dbm-to-microvolts.pdf> Ref 3 - <http://www.ka9fur.net/geduplex/duplex.html> Hope this helps... Comments welcome, reply to the list. Kevin Custer W3KKC Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/Repeater-Builder/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
