Thanks @abcd567,
It’s been a long night thru the calculations and estimations of the values. I still have some issues on how much to cut and its based, sometimes, in the length of the antenna connector, that I must take in consideration. Aniway, I am trying to get a standard cable and calculate the cut based on the phase of the reflected wave, just to make sure that the same cable leads to the same measure with different conectors over the phase related value. I will keep you posted
In nearly 40 years of playing radio, I am yet to encounter any single instance where coax length makes the slightest difference and that’s working with frequencies from kHz up into the GHz.
Adjusting the coax length cannot change the SWR of the aerial attached to the end of it, it’s just not possible.
If you changed the coax and your reception improved, you most likely had bad coax, a poorly installed connector or even just a poor connection somewhere.
This happened because you used a coax of same impedance as the impedance of Antenna (say both 50 ohms)
Please see the formulae below. If you make Zant = Zcoax = 50, then the formulae simplifies to Zin = Zcoax, i.e length of coax will no longer change the impedance seen Zin
Bingo! Another reason to use the correct coax for the job.
Your tests above with different length cables, mismatched from the impedance of the aerial just demonstrate even further that one should always use the right impedance coax.
That is why “on the air TV” antennas (Yagis) with folded dipole active element (impedance 300 ohm) used a 300 ohm flat ribbon cable (twin-lead, similar to ladder line).
Similarly Satellite Dish antennas use 75 ohm coax, as the TV receiver’s coaxial input, and LNB at dish are both 75 ohms.
Exactly. You’re agreeing with me and this is why I get frustrated when I see posts from yourself and others insisting that the coax doesn’t matter, that it’s perfectly OK to mix and match 75 ohm coax with a 50 ohm aerial. You agree with me that one should be consistent but then post this sort of thing.
You can’t ignore the problem of mixing different impedance parts, the aerial system is just that, it’s a system, it’s a combination of the feeder and the radiating (or receiving) element.
It’s a waste of time and effort trying to find the ‘right’ length of coax. Just use the correct impedance feeder and not only will you have lower loss in the system but you can then use any length you like.
I’ve been saying this for years.
The formulae I presented here is general and applies to mixing of systems of any impedance values. In general the bigger the difference between two impedance values, the bigger the variation / degradation.
The degradation will be pronounced and unacceptable if a 50 ohm system is mixed with a 150, or 200, or 300 ohm system.
In the specific case of mixing 50 ohm system with 75 ohm system, the difference in impedance values is low enough to cause minimal degradation effect, and for all practical purposes it is acceptable.
You’ve demonstrated yourself in the pictures above that there’s quite a large impedance difference depending on cable length with it varying from 32.46 ohms to 77.26 ohms. I can’t be arsed to calculate the percentage but it’s not insignificant. @wcostrino seems to confirm this by problems he’s having and having to change the cable length to match the entire system.
I’ve said my piece, I really have been saying her for years that one should use the correct coax impedance. From what I can see, you’re agreeing with me but yet you’re still arguing against it.
I’m done, I really am. Forget it.
In that demonstration, I used dummy load resistors. I purchased these from china, with different values of 50 &75 ohms, but these were unmarked with their values. This unexpected result seems to be due to using wrong valued dummy load resistor. If both the dummy load resistor and coax had same impedance, there would be no variation of measured value with length. Similarly if it was combination of 75 & 50, variation will not be as great as shown by measured value.
This experiment need to be redone with dummy load resistors of good quality and value marked.
Even though the results of that experiment are inaccurate, that experiment proves one thing: if impedance of coax and load (or antenna) are not same, the impedance measured/seen at feed end of coax varies with length of coax.
It depends on what you’re trying to accomplish. If you just want to casually listen to shortwave radio for example, the impedance mismatch between 50 and 75 ohm coax has a negligible effect (that’s offset by the higher quality and better shielding of satellite TV RG6.) Even at microwave frequencies like ADS-B’s the slight impedance mismatch isn’t really that important unless you’re trying to squeeze every last nautical mile of range and every last message possible out of your setup, where every dB and fraction of a dB counts. For a newbie’s first ADS-B setup RG6 would be OK and let him or her start seeing airplanes. It’s way better than the RG58 or RG174 that might be what his antenna comes with or that he may unknowingly buy “because it’s 50 ohms so it’s better.”
It also depends on the impedance of the antenna. Back to my SWL example, an end-fed randomwire antenna’s impedance is somewhere around 450 ohms. You would want to use a 1:9 balun to match the impedance to that of the coax and the SDR, but again 75 ohm coax would be fine (it’s actually 72 ohms and the Z of 50 ohm coax is really 52 so the difference is even less.) A center-fed horizontal dipole’s Z is around 50 to 75 ohms depending on the antenna’s height above the ground so a 1:1 balun would be desired for best reception as it acts as an isolation transformer that supresses noise.
Even if you’re transmitting at low power (<10 watts), you don’t need a perfect 1:1 SWR in order to be heard. Using RG6 would yield roughly 1.5:1 which is just fine. In fact, in real-world transmitter installations it would be excellent. (SWR figures assume the antenna has been properly designed, properly installed, tuned to be resonant at the intended frequency and that you have a good ground.)
Newbies who read this thread might takeaway that 50 ohm coax = good, 75 ohm coax = bad, so they’ll run out to Wallmart or Home Depot and buy RG58 because it’s 50 ohm when they would be much better off buying the roll of 75 ohm satellite TV RG6 down the aisle but they don’t because “the experts on FlightAware said I had to use 50 ohm coax or it won’t work.”
Don’t you have a multimeter and a pen?
Without disagreeing with you, the “system” includes the receiver.
Most of my sites use the $10 receivers with either PAL or F connectors. For these systems, it would seem silly to over-capitalise on low-loss 50Ω coax when there will still be a miss-match.
For anyone with: terrain limited range, indoor antenna, bent coat-hanger antenna, cheap/basic receiver etc, then using (quality) RG6 won’t be the major limiting factor.
As soon as you have invested in a better-than-ref-design receiver, then it becomes worthwhile to upgrade the other components.
One thing not often discussed is the connectors. I see no point using $5/m coax and terminating it with no-name $1 connectors.
I agree. When you decide to move beyond the beginner stage and invest in a better antenna and SDR, and start adding filters and LNAs, that’s also the time to replace the coax with a good low-loss (at 1090MHz) one like LMR400 or at least LMR240 like I use. I always buy mine with connectors preinstalled because they’re installed better than what I could do myself. Unfortunately this means I’m limited to using whatever pre-assembled lengths that are available. In my case I have one 10 meter and one 2 meter length connected together via an SMA adapter which introduces another 0.2dB of loss (you get 0.1dB of loss for each pair of connectors and I imagine also with each cable adapter) but I compensate by having a filtered LNA just below the antenna which makes up for all the coax and connector attenuation, and also improves the signal-to-noise ratio and linearity.
This experiment I performed 5 years ago in Feb 2017. I cant find those dummy load resistors now.
At that time the purpose of this experiment was to demonstrate that impedance of a load (antenna) seen at other end of coax varies with length of coax, which the experiment demonstrated successfully.
As the experiment was not intended to determine precise values of mesured impedance across coax vs directly measured value, I did not bother to measure dummy load’s value by a multimeter.
I see:
What makes you think your measurements can be applied to other components?
All test setup use connectors, or is there a way to connect coax without using any connector?
Also this setup used same connectors with different lengths of coax. The connector’s contribution in all three test pieces was therefore constant, and the only variable was length of coax.
Please note this setup was NOT meant for precice measurement of impedance. It was meant to show that impedance seen varies with length of coax while all other things remain same.
If you start with a 50~75 transition, it rather invalidates applying the observations to other setups.
Similarly, you need to have confidence in the components you are using.
eg. you have two terminators that ‘look’ that same. Regardless of how good/bad they are, a basic test is to run a scan with one then repeat with the other being careful not to change anything else. It won’t tell you anything about the quality of the components, but if the two tests show different results, you’ll know there is a problem.
That is an unnecessarily flippant response.
Yes you you need connectors, but you don’t have to use the cheapest nastiest connectors you can find.
There is a reason quality connectors cost more the junk connectors.
If you just junk connectors on junk cable after an impedance mismatch - what exactly are you measuring?
The theory / formulae show two things very clearly:
(1) If Coax and load (antenna) impedance have same value, then there will be no variation in value of impedance measured at receiver end irrespective of the length of coax. It will always be Zin = Zcoax = Zant
(2) If Coax and load (antenna) impedance are different, then measured impedance at receiver end Zin will vary in a cyclic manner with length of coax.
The above phenomenon is independent of quality of coax and connectors, hence the arguments that this phenomenon occurs because of poor quality coax and connectors, and will not occur if good quality coax and connectors are used, is not valid.
As regards quality of coax and connectors used in the demo setup, I have already said that this setup was NOT meant for precise measurement of impedance. It was meant to demonstrate cyclic variation of receiver end impedance with variation of length of coax. Raising issue of accuracy of measurements and quality of material used is simply side tracking the objective of the experiment.
If you do not believe the experiment is valid, please forget about it, and search the Internet about the subject phenomenon.
You’ve missed my point entirely.
If you use junk components, you can not know what is dominating the results.
No I did not miss your point.
As you were not trusting experimental demonstration of phenomenon due to low quality coax and connectors, I took another approach. I explained the phenomenon by formulae which is independent of quality of coax & connectors. I then advised you to disregard the experiment as the formulae alone could prove the phenomenon.
Repeating yourself does not make it right.
I was pointing out the limitations for other peoples benefit as much as yours.
By using an SMA-F adapter, you have spiked your own experiment.
Try using your terminator direct on the adapter and see what result you get.
