Best coax length

[freecell]

dissipation is the release of electrical energy in the form of heat.

reflection loss or swr loss is a NON-DISSIPATIVE type of loss.
since it is NON-DISSIPATIVE THERE IS NO HEAT.

 

[W5LZ]

freecell,
I see what you are saying. But does that mean that there is never any heat in the feed line when there's SWR?
- 'Doc

 

[Marconi]

FC, I have a few questions?

After the resonant 1/2 wavelength line of 14.083' gives you a true mirror look at the line with an input Z = 36.042, you wish to get that match closer to Z=50 ohms by adjusting the line length a bit. You then determined that line length will be 19.1846' and would result in that input, Z=50.

Even so, we still have some inductive reactance that needs to be removed. Am, I alright to this point :?:

After this point I get lost, but I'll give it a try.

Do you use the new 19.1846' line as the feed line to the load, placing it into a port on a T-connector that is attached to the output of the Tx'r, and then use the shorted 5.1013' line as a shunt to the remaining open port of the T-connector, in order to cancel out the remaining inductive reactance :?:

Also, right after the results of 089.gif you asked, "look familiar?" What do you mean looks familiar, the load values for all of the tests are the same in every case, meaning the SWR does not change :?:

You also noted that later you would get back to whether the SWR changed or not. Did I miss that part :?:

Eddie

 

[freecell]

the 5.1013' line produces nearly the same results as the 19.1846' line, take another look at the input Z, X and R and the context surrounding this part of the post.

"there will not be just one spot in the line where this will occur, there will be several depending on the length of the line."

there are other spots in the line sample where Z=50 with -j values of capacitive reactance at the tx input that i haven't even mentioned yet. i made no comments in reference to the exact shorted length of the feedline used to cancel the +j values remaining at the tx input in my example, just that it was an option.

 

[Marconi]

Yes I see your point about the shorter line being one of several locations on the line that produce results, but how does the T-connector come into play here?

I know it was not your point in the piece to explain the software, but how does the software work to figure the lengths. What variables do you enter and where in order to produce the new line lengths?

 

[freecell]

the t-connector allows the transmitter to connect to the line while making it possible to add a shorted or open length of line in parallel with both of them (inductive or capacitive) to cancel any reactance present at tne transmitter.

in the software the line length, load resistance, load reactance, feedline type, velocity factor, design frequency and cable length are all inputted variables. the length of the line is also outputted in wavelengths and degrees and as you might well guess, changes in any of them result in changes in the values of Z, R and X seen at the transmitter input. the program can also be used in reverse beginning with the input variables and sampling the range of load variables as the other parameters are adjusted.

 

[trash]

Excellent post freecell.
Yes, it's correct that the length of coax does not matter.
On a perfect or good transmission line you will see a flat response across the band. like this one
http://www.austech.info/attachment.php?attachmentid=52140
But it's not a perfect world and there is always some reflection, the poorer the quality cable, the more you will notice it. Here is an excellent example of a poor cable.
http://www.austech.info/attachment.php?attachmentid=52243
The load is good, you can compare it to other related tests I did on other cables.
The top is insertion loss, the bottom return loss. (VSWR).
You can see that the VSWR varies with frequency, and not just any frequency, but those releated to the length of the cable (5m) Doh ! But it is only an illusion, the real VSWR is the peaks, not the troughs. Since the load in this case is esstially ideal, it is the actual return loss of the cable your seeing.
The troughs are caused by the standing waves of the bad mismatched cable, not a mismatched load as would be the case with an antenna. But it is still the same kind of result.

 

[Marconi]

Trash, your links will not work unless one has access to the site you post. I did not try to register, but maybe they will work then, not sure though.

 

[Sonwatcher]

Interesting article on coax and coax length-

http://www.signalengineering.com/ultimate/coax_basics.html

 

[freecell]

"Yes, it's correct that the length of coax does not matter".

the length is of no consequence only when Zl, Zc and Zs are all equal. when any one of these values deviates from 50 ohms for our example here the length of the line plays an integral role in the outcome. the link is appreciated but we're way beyond what scott covers in his coax basics.

here are the images referenced above, in order........

http://www.firecommunications.com/images/rg11.5.jpg
http://www.firecommunications.com/images/t.jpg

while i can see the relevance of the graphs linked to the phase angle relationships that we're dealing with here pile up very closely together on the line at frequencies between 500 mhz. and 2 ghz. i believe the context which these graphs were drawn from was satellite related. let's keep it in the hf range and continue.