Coax Fault Finding: Locate Breaks, Shorts and Water With a NanoVNA
Coax fault finding starts with a symptom. The signal is weaker than it was last month, the SWR has crept up, the noise crackles when the wind blows, or the antenna that worked well in the summer is dead after a storm. The cause is often in the one part of the station that you cannot see all at once: the coaxial cable. It runs through a wall, across a roof, down a mast and along a fence, and a fault can be anywhere on it. Replacing the entire run is expensive and a lot of work, and you may replace it for nothing if the fault is in a connector. Finding the fault is the art of locating the problem with a few measurements. This guide starts with the simple tests that need only a multimeter and a dummy load, and then shows how a time domain reflectometer, or TDR, including the time domain function of a NanoVNA, tells you how far away the fault is.
Common Faults and Their Signs
Coax cable fails in a limited number of ways, and the symptoms point to the type of fault. The table lists the usual cases.
| Fault | Typical cause | Symptom |
|---|---|---|
| Open circuit | Broken centre conductor, pulled-out connector, cut cable | Very high SWR, little power reaches the antenna |
| Short circuit | Braid strands touching the centre pin, crushed cable | High SWR, a protection circuit may cut the power |
| Water in the cable | Cracked jacket or unsealed connector | Gradually rising loss and SWR, worse after rain |
| Kink or crush | Sharp bend, a pinch at a window or a clamp | A bump in the SWR and extra loss |
| Corroded or loose connector | Weather, poor soldering or crimping | Intermittent noise and changing SWR |
| Degraded cable | Years of sun and heat, cheap braid | High loss on the upper bands |
The first useful observation is when the problem appears. A fault that comes and goes with the wind suggests a loose connector or a partly broken conductor. A fault that grows after rain suggests water. A cable that has always had a high loss on VHF may be the wrong type or a poor one, which our guide to coaxial cable types helps you judge.
Step One: Simple Tests
Do the easy things first, with the antenna disconnected and the transmitter off.
- Inspect. Walk the whole cable and look for cracks in the jacket, flattened sections, bites by animals, the drip loops and the connectors. Open the weather seal of each connector if you suspect water, and look for green corrosion.
- Measure the resistance. At the shack end, with the other end free, a multimeter should show an open circuit between the centre pin and the shield. If it shows a low resistance, there is a short somewhere. Then make a temporary short at the far end, and you should read a very low value, a fraction of an ohm to a few ohms, between the pin and the shield. If you read an open, the centre conductor is broken.
- Test with a dummy load. Put a 50 ohm dummy load at the antenna end of the cable, and measure the SWR at the shack. A good cable shows a low SWR, close to 1.0 on the low bands. If the SWR is high with the load connected at the far end, the fault is in the cable, and not in the antenna. If the SWR is good with the load but bad with the antenna, the antenna is the problem. Remember that a low SWR does not prove that a cable is healthy, as a lossy cable also hides a poor antenna.
- Flex the cable gently. While you watch the SWR or the resistance, move the connectors and the cable. If the reading jumps, the fault is where you moved it.
Often these steps are enough. A connector with a loose centre pin or a bad solder joint shows up in the continuity test, and a re-made connector solves the problem. The method for making a good connection is described in our guide to soldering a PL-259.
How a TDR Works
If the simple tests tell you there is a fault but not where, the next tool is the time domain reflectometer. It sends a fast pulse into the cable and watches for echoes. A pulse that travels along a cable of uniform impedance is not reflected. Wherever the impedance changes, a part of the pulse is reflected back, and the instrument measures the time between the pulse and the echo. Since the pulse moves at a known speed, the time gives the distance. The speed is the speed of light multiplied by the velocity factor of the cable, which is explained in our article on the velocity factor. A solid polyethylene cable such as RG-58 or RG-213 has a velocity factor of about 0.66, a foam cable about 0.8 and many low-loss types about 0.85.
The distance to the fault is half of the echo time multiplied by the speed in the cable. Because the speed of light is about 0.3 metres per nanosecond, a cable with a velocity factor of 0.66 gives about 0.2 metres per nanosecond in one direction, so each nanosecond of echo time corresponds to about 0.1 metre of cable. An echo that returns after 100 nanoseconds indicates a fault about 10 metres away.
The shape of the echo shows the type of fault. An open circuit reflects the pulse with the same polarity, so the trace goes up. A short circuit reflects it with the opposite polarity, so the trace goes down. A section where the impedance is lower, for example because water has raised the capacitance, appears as a dip that has a width, and a kink or a crushed section as a small bump. The characteristic impedance of the cable is what the pulse sees along the way, and any change from 50 ohms makes an echo.
Using a NanoVNA as a TDR
You do not need to buy a dedicated TDR. A vector network analyzer measures the reflection over a range of frequencies, and a mathematical transform converts it into a function of time, which gives the same picture. Our guide to the NanoVNA shows how to calibrate the instrument. To use the time domain mode:
- Calibrate at the end of the test cable, or at the end of the adapter that you will use, with the open, short and load standards.
- Set the velocity factor of the cable in the transform menu. If you do not know it, measure a short piece of the same cable of known length, or use the manufacturer’s value.
- Select the low-pass step or impulse transform of the S11 trace and read the distance to each feature on the horizontal axis.
- Choose the sweep range to fit the cable. A wide span gives fine resolution but a short range, and a narrow span reaches farther but with coarser detail.
The relation is simple. The resolution is the speed in the cable divided by twice the span, and the maximum range is the resolution multiplied by the number of points in the sweep. The table gives the numbers for a velocity factor of 0.66 and a sweep of 101 points.
- Span of 900 MHz: resolution about 0.11 m, maximum range about 11 m.
- Span of 300 MHz: resolution about 0.33 m, maximum range about 33 m.
- Span of 100 MHz: resolution about 1 m, maximum range about 100 m.
- Span of 30 MHz: resolution about 3.3 m, maximum range about 330 m.
For a typical run of 15 to 30 metres a span of 100 to 300 MHz is a good compromise. More points in the sweep extend the range at the same resolution, and a larger version of the instrument can offer 401 points or more. If the display shows nothing at the expected distance, widen the range or reduce the span.
Reading the Result
Start by measuring a good cable of the same type to learn what a healthy trace looks like. You will see a feature at the zero point, due to the connector and the adapters, and then a flat trace until the far end, where an open or a short makes a large step. If the far end is the antenna, you will see the feed point as a smaller reflection. Compare your cable with this model and look for extra features. A step near the middle shows a damaged section or a splice, and a gradual change in the trace shows water in the cable. A dip with a width of a metre or more near a connector is the classic sign of water that has entered and spread along the cable. Check the distance to the feature, then walk to that point along the cable and inspect it. A measurement to an accuracy of about one metre is usually enough to find a fault, especially if the cable runs through a few places where it is easily damaged.
Check the measurement with a second method when you can. Measure the length of the open cable to test the velocity factor: if the instrument shows 21.2 metres for a cable that you know to be 20 metres, the velocity factor was set about 6 percent too high, and you should correct it before you trust the distance to the fault.
Fixing the Fault
Once you know where the problem is, the repair depends on the cause. A faulty connector is cut off and replaced, and the new one must be sealed with tape and mastic as explained in the guide to weatherproof coax connectors. A cable that has taken in water is rarely worth saving, because the water spreads along the braid and corrodes it, and a short stretch of dry cable that you cut off is not a permanent solution. A crushed or kinked section is best cut out and replaced with a splice using a good double female connector, or a new length. If the loss has risen on all the bands, the cable may simply be old, and a thicker low-loss cable is the better choice for the new run. Finally, record the result of a good measurement: the SWR, the length and the trace of the healthy cable are a valuable reference for later.
Preventing the Next Fault
Most coax failures are failures of installation, not of the cable itself. Leave a drip loop where the cable enters the house, so that water runs off the cable instead of following it inside. Support the cable at intervals so that the weight does not pull on the connector, and use a UV-resistant jacket where it is exposed to the sun. Avoid bends tighter than the manufacturer’s minimum radius, which is usually about ten times the diameter of the cable, and do not squeeze the cable in a window frame or under a door. Seal every outdoor connector with self-amalgamating tape and a layer of vinyl tape, and check it once a year as part of the seasonal inspection. A short log with the date, the SWR and the length of each run, together with a saved trace of the healthy cable from the NanoVNA, makes the next diagnosis a ten-minute job.
The Bottom Line
Coax fault finding begins with inspection, a continuity test with a multimeter and a dummy load at the far end to decide whether the cable or the antenna is at fault. A TDR, or the time domain mode of a NanoVNA, then locates the fault by timing the echo: distance equals half of the echo time multiplied by the speed of the signal in the cable, which is the velocity factor times the speed of light. An open shows as an upward step, a short as a downward step, and water or a crush as a dip or a bump. Choose the sweep span to match the cable length, check the velocity factor on a known length and walk the cable to the indicated distance.
Frequently Asked Questions
How do I know if my coax is bad or my antenna is bad?
Put a dummy load at the far end of the cable and measure the SWR at the shack. If it is high, the cable has a fault. If it is low, check the antenna.
What is a TDR?
A time domain reflectometer sends a pulse into the cable and measures the time for the echo from a fault. The time gives the distance to the fault.
Can a NanoVNA find a break in a cable?
Yes. Its time domain transform of the reflection measurement shows the distance to an open, a short or a change of impedance, if the velocity factor is set correctly.
What velocity factor should I use?
Use the value from the data sheet. Solid polyethylene cables such as RG-58 are about 0.66, foam cables about 0.8, and many low-loss cables about 0.85. Measure a known length to check.
Can water in coax be repaired?
Seldom. Water spreads along the braid and corrodes it, so the cable is usually replaced. Cutting off a wet end may help only if the water has not spread far.
How accurate is the distance to the fault?
With a suitable span, within about a metre or better. The error comes mainly from the velocity factor, so check it on a cable of known length.