Can anyone help me out with how to electrically measure coax lengths? Rather than using a tape measure.
If I am hooking 4 yagis together using a power divider I want my coaxial lengths to be all exactly the same. I currently have used a tape measure. But I would rather be a little more accurate taking into account velicity factors etc of the cable to be used. I would want to cut the first cable to the length needed, then be able to measure that electrically and then replicate that first measure trimming up another 3 lengths to be exactly the same.
Any help?
Electrically measuring coax length
-
VK2GOM
Re: Electrically measuring coax length
Leigh,
Assuming you already have a good fast oscilloscope - you need to build a TDR - Time Domain Reflectometer. Easy to build. Info here:
http://www.epanorama.net/circuits/tdr.html
I used one at Uni, a good few years ago
73 - Rob VK2GOM / G0MOH
Assuming you already have a good fast oscilloscope - you need to build a TDR - Time Domain Reflectometer. Easy to build. Info here:
http://www.epanorama.net/circuits/tdr.html
I used one at Uni, a good few years ago
73 - Rob VK2GOM / G0MOH
-
VK2FLAW
Re: Electrically measuring coax length
Leigh,
Why not just use the Pulse Echo Tester from work?
It's configured for balanced testing, but it should give the same reference figures for comparison of the separate cables.
You aren't overly concerned with physical length, only the electrical length.
Why not just use the Pulse Echo Tester from work?
It's configured for balanced testing, but it should give the same reference figures for comparison of the separate cables.
You aren't overly concerned with physical length, only the electrical length.
Re: Electrically measuring coax length
I've used the following for cutting phasing lines to equal electrical lengths on many bands from 80M to 2M ...
An open circuit "phasing line cable under test" of a half wavelength (or multiple thereof) will present an open circuit at it's port on the Tee connector, thus the SWR meter will only "see" the dummy load and display a low SWR/low reflected power. Any deviation in the transmitter frequency away from a half wave (or multiple thereof) in the phasing line will introduce some reactance from the phasing line across the dummy load and the SWR will rise.
Normally I'm using phasing lines with a multiple of a half wave length on the ham band in question for no good reason other than using the above measuring method with a ham band transmitter ... but it can be done for any length if you have a rig that has had the frequency range "expanded" and it can transmit on any frequency to find an SWR dip and thus a multiple of a halfwave length.
To get an accurate ("sensitive") reading, use about 25W of RF power and a Bird 43 wattmeter with a 25W slug "switched" in the reverse position for the initial cuts until the reflected power is getting really low, then insert a 1w slug for the final trimming of the phasing line to the target frequency. With care it is an easy matter to get several phasing lines cut with electrical lengths matched so the SWR meter dip is within 10 or 20kHz on the 2M band (for example) ... to my way of thinking this is around 0.014% accuracy (20kHz/144MHz).
The physical lengths usually turn out being slightly different. On longer phasing lines (e.g. around 4 meters long at 144Mhz for an array of yagis) this difference can sometimes be several cm! 9913 is particularly bad in this respect (it is also bad coax in other respects at VHF
)
With a little imagination, measurement and calculation the same method can be used to determine the velocity factor of unknown coax cable.
Note: for cutting short lengths e.g. one half wave length of coax for a 4:1 coax balun ... or one quarter wave length for a transmission line transformer (where the respective coax length needs to be INSIDE the coax braid with extra length for termination leads at each end), it is better to determine the VF of the actual coax you are going to cut from by electrically measuring a longer length (several half waves) using the depicted method, then physically measure the long length of coax with a ruler/tape measure and do a little maths to determine the actual velocity factor. Once the VF is accurately determined, measure out the calculated electrical half wave or quarter wave length with a ruler and add the necessary millimeters for the termination ends before cutting.
Cheers, Bob ZL1RS
An open circuit "phasing line cable under test" of a half wavelength (or multiple thereof) will present an open circuit at it's port on the Tee connector, thus the SWR meter will only "see" the dummy load and display a low SWR/low reflected power. Any deviation in the transmitter frequency away from a half wave (or multiple thereof) in the phasing line will introduce some reactance from the phasing line across the dummy load and the SWR will rise.
Normally I'm using phasing lines with a multiple of a half wave length on the ham band in question for no good reason other than using the above measuring method with a ham band transmitter ... but it can be done for any length if you have a rig that has had the frequency range "expanded" and it can transmit on any frequency to find an SWR dip and thus a multiple of a halfwave length.
To get an accurate ("sensitive") reading, use about 25W of RF power and a Bird 43 wattmeter with a 25W slug "switched" in the reverse position for the initial cuts until the reflected power is getting really low, then insert a 1w slug for the final trimming of the phasing line to the target frequency. With care it is an easy matter to get several phasing lines cut with electrical lengths matched so the SWR meter dip is within 10 or 20kHz on the 2M band (for example) ... to my way of thinking this is around 0.014% accuracy (20kHz/144MHz).
The physical lengths usually turn out being slightly different. On longer phasing lines (e.g. around 4 meters long at 144Mhz for an array of yagis) this difference can sometimes be several cm! 9913 is particularly bad in this respect (it is also bad coax in other respects at VHF
)With a little imagination, measurement and calculation the same method can be used to determine the velocity factor of unknown coax cable.
Note: for cutting short lengths e.g. one half wave length of coax for a 4:1 coax balun ... or one quarter wave length for a transmission line transformer (where the respective coax length needs to be INSIDE the coax braid with extra length for termination leads at each end), it is better to determine the VF of the actual coax you are going to cut from by electrically measuring a longer length (several half waves) using the depicted method, then physically measure the long length of coax with a ruler/tape measure and do a little maths to determine the actual velocity factor. Once the VF is accurately determined, measure out the calculated electrical half wave or quarter wave length with a ruler and add the necessary millimeters for the termination ends before cutting.
Cheers, Bob ZL1RS
Bob, ZL1RS in the Bay of Islands at RF64vs
-
VK2TDN
Re: Electrically measuring coax length
Bob, ZL1RS ...
Thats
BRILLIANT I have copied and pasted that into a file on its own for easy future reference
thanks so much
Dave
VK2TDN
Thats
BRILLIANT I have copied and pasted that into a file on its own for easy future referencethanks so much
Dave
VK2TDN
-
VK2KRR
Re: Electrically measuring coax length
Thanks for the replies everyone. I'll look into the suggestions a bit further and maybe try some tests.
-
VK2KRR
Re: Electrically measuring coax length
Well Ive just tried to measure a small 1m RG223/u patch cable with a pulse echo tester but its virtually impossible to read how long it is due to the weird short then open type curve and where do you pick is where the open starts?
I think it needs to be something that gives a numerical reading would be more accurate.
Also no I dont have an oscilliscope.
I was sure I had tried the pulse echo tester (time domain reflectometer) before and it was not at all practical and that just proved why.
I think it needs to be something that gives a numerical reading would be more accurate.
Also no I dont have an oscilliscope.
I was sure I had tried the pulse echo tester (time domain reflectometer) before and it was not at all practical and that just proved why.
Re: Electrically measuring coax length
I have had trouble measuring "short" cables too ... inconclusive results like you report.
Try a longer cable e.g. a few wavelengths (like the phasing lines you want to build) and see what happens.
If you still have trouble, the SWR meter/dummy load/T connector method described above works a treat ... I've built phasing harnesses for at least ten 2-bay and 4-bay EME arrays on 2m and 70cm using that method and never had reason to doubt the results. The frequency of the narrow dip on the SWR meter as indicated on your radio's display is your "numerical" reading. It is quite repeatable.
73, Bob
Try a longer cable e.g. a few wavelengths (like the phasing lines you want to build) and see what happens.
If you still have trouble, the SWR meter/dummy load/T connector method described above works a treat ... I've built phasing harnesses for at least ten 2-bay and 4-bay EME arrays on 2m and 70cm using that method and never had reason to doubt the results. The frequency of the narrow dip on the SWR meter as indicated on your radio's display is your "numerical" reading. It is quite repeatable.
73, Bob
Bob, ZL1RS in the Bay of Islands at RF64vs
-
VK2KRR
Re: Electrically measuring coax length
Thanks Bob I will look into this one next when I get a chance.
-
VK2XSO
Re: Electrically measuring coax length
There are quite a number of TDR's floating around on the surplus market. (Time Domain Reflectometers).
They're ex-telecom/telstra. I've used them at work and they once quite common. But they got rid of them and sold them all off, which is why there are
plenty on the surplus market.
I bought mine at wyong. The guy selling them had three and didn't know what they were. He was asking $20, but I told him I'd buy one for $5 and tell him what they were AND how to use them. He agreed.
I now use it ... for work (telstra) duh ! While one of the other techs uses a measuring wheel, I use the TDR. I wanted to see him use the wheel for a 5km cable run. But it can measure almost anything conductive. I've measured coaxes to find faults and even fences. Electric fences are really easy to find faults.
Just connect up the TDR and it can measure faults out to 20km. I can tell if it is open circuit or short circuit. Such a handy device if you know how to drive them.
As mentioned above, you can use a CRO and a pulse generator to do the same thing. The simplest TDR is nothing more than a CRO with a relay.
A 4016 as the gate and a 555 timer works just as well. Send a DC pulse down the wire and have it trigger the sweep on the CRO.
For short measurements you need to be able to produce very short pulses. Tunnel diodes are good for the task, but I had false pulses from 'ringing' of my circuit. I've tried to use TTL bit slicing to get short asymmetrical pulses, but I had some problems with the TTL logic and never finished the experiment.
Adjust the sweep time on the CRO against a reference piece of cable. 10m is a good length to calibrate against and most higher frequency oscilloscopes can manage 100nS divisions, even my old 15MHz CRO can manage 500nS/div.
They're ex-telecom/telstra. I've used them at work and they once quite common. But they got rid of them and sold them all off, which is why there are
plenty on the surplus market.
I bought mine at wyong. The guy selling them had three and didn't know what they were. He was asking $20, but I told him I'd buy one for $5 and tell him what they were AND how to use them. He agreed.
I now use it ... for work (telstra) duh ! While one of the other techs uses a measuring wheel, I use the TDR. I wanted to see him use the wheel for a 5km cable run. But it can measure almost anything conductive. I've measured coaxes to find faults and even fences. Electric fences are really easy to find faults.
Just connect up the TDR and it can measure faults out to 20km. I can tell if it is open circuit or short circuit. Such a handy device if you know how to drive them.
As mentioned above, you can use a CRO and a pulse generator to do the same thing. The simplest TDR is nothing more than a CRO with a relay.
A 4016 as the gate and a 555 timer works just as well. Send a DC pulse down the wire and have it trigger the sweep on the CRO.
For short measurements you need to be able to produce very short pulses. Tunnel diodes are good for the task, but I had false pulses from 'ringing' of my circuit. I've tried to use TTL bit slicing to get short asymmetrical pulses, but I had some problems with the TTL logic and never finished the experiment.
Adjust the sweep time on the CRO against a reference piece of cable. 10m is a good length to calibrate against and most higher frequency oscilloscopes can manage 100nS divisions, even my old 15MHz CRO can manage 500nS/div.