Electrical Quarter and Half Wave coax length?
Displaying 1-6 of 6
| Jul 26th 2019, 17:11 | |
KR7CMTotal Topics: 0 Total Posts: 0 |
Calculating the electrical length of a quarter or half wave seems pretty straight forward. Or so it initially seems. Say, for example, I wish to make an electrical quarter wave section of coax to convert a 100 Z ohm antenna feed point to 50 Z ohms to match my feed line. I take a length of RG-59, calculate the length (in feet) using 983.6 (speed of light) divided by the frequency (145 MHz) and multiply that by the cable velocity factor (in this case 0.78). This gives me a length of ~1.3227 feet. OK great, right? Not Really... My question here is, where is this calculated length actually measured? The Shield length? The Center conductor length? How about when connectors are installed on both ends? Any clarification would be appreciated. |
| Jul 26th 2019, 22:01 | |
W1VTSuper Moderator Total Topics: 0 Total Posts: 0 |
You need to accurately model or account for the terminations on each end of the coax. UHF connectors aren't a constant impedance. I've modeled them as short transmission lines in series. I sawed one in half to and measured the conductor diameters. In practice, it may be easiest to measure the electrical length, or vary the length and choose the one that provides the lowest measured SWR. One idea I had for using RG-6 is to use F connectors but to solder the center conductors directly to connection points. RG-6 has an aluminum shield. RG-59B/U may have a copper braid, but these days you can't be too sure unless you have an actual sample to look at. Zak W1VT ARRL Senior Lab Engineer |
| Jul 27th 2019, 00:05 | |
KR7CMTotal Topics: 0 Total Posts: 0 |
Zak, First, Thank you for the response. However, the fundamental answer to the question was not provided. Not that I disagree with with anything you've stated. In a perfect world all of these factors would be taken into consideration. It's just that I am looking for a more basic answer here. Assume I don't have access to a TDR or analyzer, although I do agree one could resort to basic VSWR, something most amateurs should be capable of doing. Further assume, for the sake of practicality, that a given piece of cable does meet specifications. Given the above, before one gets to verifying/analyzing a cable, one needs to have a starting point, ergo the calculated electrical length. Variances in coaxial cable and connectors notwithstanding, but set aside for the moment... once I've calculated a given electrical length, where/how do I measure to apply that calculated length when cutting the coaxial cable? Again, I am looking for basic, fundamental, answer here as it pertains to a given calculated electrical length. Where is the calculated length measured? It's logical to assume that there will be some trimming of the cable to accommodate soldering to either another cable or some end point. Further assume that the trimming to accommodate a connection is as minimal as possible. Therefore, does not determining where/how to cut, include knowing where to measure... before cutting? Do I measure the calculated electrical length referenced to length of the cable shield or, do I measure the calculated electrical length referenced to the length of the inner conductor? :D |
| Jul 27th 2019, 19:43 | |
W1VTSuper Moderator Total Topics: 0 Total Posts: 0 |
One approximation is to use the shield length plus the center conductor as a 150 ohm transmission line. It is likely that using the shield length is too short and using the center conductor length is too long. There was a fellow who made 432 MHz EME rat race combiners and found that he was better off not using connectors. http://kl6m.com/misc/KL7WE-hybrid.pdf Zak W1VT ARRL Senior Lab Engineer |
| Jul 27th 2019, 22:44 | |
KR7CMTotal Topics: 0 Total Posts: 0 |
Zak, Thank You! That answer was helpful as was the link! |
| Jul 29th 2019, 10:22 | |
W1VTSuper Moderator Total Topics: 0 Total Posts: 0 |
A simple rule of thumb. You can probably get away with either at 6M or the HF bands. At 2M you need to experiment, or copy what someone else has found to work. At 432 you need to be lucky to get it to work with connectors. Zak W1VT ARRL Senior Lab Engineer |
Displaying 1-6 of 6
