RE: [Repeater-Builder] measuring coax for duplexer
Cables between cavities include the length of the coupling loops in the cavities. They are measured from the bottom of one loop, the cable, to the bottom of the loop in the next cavity. You need to figure the velocity factor of the cable and the velocity factor of each loop separately as the cable has a specific velocity factor and the loop will have a velocity factor of air. The cables that go to the T are measured from the bottom of the loop again as above to the center of the T connector. A notch cavity will provide about 30 to 35 db of notch depth. Each interconnecting cable provides an additional 10 db or so of notch depth when its length is right. Don't forget that the transmit side cables are tuned to the receive frequency just like the transmit notches are. The cable on the transmit side to the T provides a high impedance at the receive frequency because the notch in the transmit cavity is a short circuit at the receive frequency. That reflects a high impedance or open circuit at the center of the T connector thru the quarter wave length at the receive frequency. The cable on the receive side provides a high impedance at the transmit frequency at the T because the notch in the receive cavity is tuned to the transmit frequency and provides a short there. The quarter wave length cable reflects that short to a high impedance or open circuit at the T to the transmit frequency. Ideally you could disconnect the receive cable at the T and it should not effect the transmit power flow to the antenna. When the transmit energy gets to the T connector it sees a high impedance going to the receiver and a lower impedance (50 ohms) going out to the antenna line so it takes that path. When a receive signal comes in it sees a high impedance in the transmit path at the T and a low impedance (50 ohms) to the receiver path. The cables that connect to the T connector are necessary to provide transmitter/receiver isolation so power flows in the right direction. If they were not properly tuned there would be a lot of suck out of transmit power and receive energy going to the wrong path. The same thing is done on a transmit combiner where each transmitter has a pass band cavity (it could be done with notch cavities instead). The cavity isolation between transmitters needs to be around 10 db on each side in order to provide the isolation so that the power goes to the antenna and not to the other transmitter. So the other transmitter must be 10 db down on the skirt of the opposite transmitters pass band skirt. You can think of these cavities as switches. The same type of thing goes on at the T connection of a duplexer. 73 Gary K4FMX > -Original Message- > From: Repeater-Builder@yahoogroups.com [mailto:Repeater- > [EMAIL PROTECTED] On Behalf Of w6nct > Sent: Sunday, April 01, 2007 7:26 PM > To: Repeater-Builder@yahoogroups.com > Subject: [Repeater-Builder] measuring coax for duplexer > > PRIMARY CONCERN: I am in the process of building a 70cm repeater and > need to know how to measure and test the physical and electrical > length of the short coax cables linking the (band-pass/notch) duplexer > cavities. > > I read somewhere that these cables should be an odd multiple of 1/4 > wavelength at the operational frequency. In my case, a 1/4 wave cable > won't easily reach between cavities; so I am using one that is > approximately 3/4 wavelength. > > QUESTION #1: Due to the Tx/Rx split, I would expect these cable > lengths might need to be different between the Tx cavities and the Rx > cavities, assuming that I can make the cables this close to a target > length. However, I am now wondering how much this small of a > difference might matter. Should I be concerned about this small of a > difference in the cable lengths (i.e., approximately 0.2")? > > > QUESTION #2: My next problem came when I tried calculating the > "physical length" for these cables. I did the calculations for my > operational frequency on the test pair, even trying to allow for the > velocity factor (Vf = 0.695) for the RG-400 cable; but I'm not sure > where to measure the cable length with respect to the type-N > connectors. At 70cm even a 1/4" will impact the effective wavelength > of these jumper cables; so this measurement seems relativel critical. > > CALCULATIONS: > wavelength(ft) = 983.6 / [frequency(MHz) * Vf] > > for 3/4 wavelength cables,... > Tx @ 446.860MHz (L = 3 * 4.5893" = 13.7679") > Rx @ 441.860MHz (L = 3 * 4.6413" = 13.9239") > > If I install the type-N connector on one end of a 14" piece of RG-400, > at what point should I measure and cut it for the other type-N connector? > > > QUESTION #3: When I prototyped up a couple cables, I tried to confirm > that I was in the right ball-park for their lengt
[Repeater-Builder] measuring coax for duplexer
PRIMARY CONCERN: I am in the process of building a 70cm repeater and need to know how to measure and test the physical and electrical length of the short coax cables linking the (band-pass/notch) duplexer cavities. I read somewhere that these cables should be an odd multiple of 1/4 wavelength at the operational frequency. In my case, a 1/4 wave cable won't easily reach between cavities; so I am using one that is approximately 3/4 wavelength. QUESTION #1: Due to the Tx/Rx split, I would expect these cable lengths might need to be different between the Tx cavities and the Rx cavities, assuming that I can make the cables this close to a target length. However, I am now wondering how much this small of a difference might matter. Should I be concerned about this small of a difference in the cable lengths (i.e., approximately 0.2")? QUESTION #2: My next problem came when I tried calculating the "physical length" for these cables. I did the calculations for my operational frequency on the test pair, even trying to allow for the velocity factor (Vf = 0.695) for the RG-400 cable; but I'm not sure where to measure the cable length with respect to the type-N connectors. At 70cm even a 1/4" will impact the effective wavelength of these jumper cables; so this measurement seems relativel critical. CALCULATIONS: wavelength(ft) = 983.6 / [frequency(MHz) * Vf] for 3/4 wavelength cables,... Tx @ 446.860MHz (L = 3 * 4.5893" = 13.7679") Rx @ 441.860MHz (L = 3 * 4.6413" = 13.9239") If I install the type-N connector on one end of a 14" piece of RG-400, at what point should I measure and cut it for the other type-N connector? QUESTION #3: When I prototyped up a couple cables, I tried to confirm that I was in the right ball-park for their length. I tried measuring their electrical using my MFJ-269 analyzer. For reference, I set the analyzer for a 270 degree (i.e., 3/4 wavelength) cable length, and tried to measure the electrical cable length of the initial (approximately 14" long) cables; before I cut and installed the second type-N connectors. Adjusting the frequency even closer, I came out with a resonant frequency (i.e., lowest SWR) at around 420 MHz, presumably indicating that the cables were long. The analyzer also indicated a 1.2' length for the initial setting of 270 degrees; which seemed about right. Assuming a target frequency at around 444 MHz (i.e., mid-way between Tx and Rx frequencies for test case), a 420 MHz resonance would indicated that the cables should be cut down to approximately 95% of their length. If this is correct, I should cut the cables down to somewhere around (14" * 0.95) 13.3". This seemed quite a bit shorter than the caculated length (around 13.8" (+/- 0.1")); so wanted to validate my test methodology as well. Am I doing this correctly? Thank-you in advance for your assistance.
