العربيةCable
Minimum Bend Radius: The Coaxial Cable Spec Everyone Ignores
Minimum Bend Radius: The Coaxial Cable Spec Everyone Ignores
Published by Bravosatcom · Cable Engineering Series
Every coaxial cable datasheet lists a minimum bend radius, usually in a table near the mechanical specifications. It is one of the most commonly ignored figures in RF installation — and one of the few whose violation causes permanent, invisible damage that will not show up until the link degrades months later.
This guide explains what the spec protects, what actually happens when you exceed it, the limits for common LMR sizes, and how to route cable correctly in the field.
What the Spec Protects
A coaxial cable is a precision transmission line. Its characteristic impedance — 50 ohm for RF work, 75 ohm for video and some satellite receive paths — depends on the ratio of the outer conductor’s inner diameter to the centre conductor’s outer diameter, and on the dielectric constant of the material between them.
Bend the cable too tightly and three things happen at once:
- The dielectric deforms. Foam PE compresses on the inside of the bend and stretches on the outside. The conductor spacing is no longer uniform, so the impedance changes locally.
- The outer conductor distorts. The braid bunches on the inside of the bend and opens on the outside, reducing shield coverage exactly where it matters. Corrugated and solid outer conductors can kink permanently.
- The centre conductor migrates. In severe cases it moves off-axis toward the inside of the bend, which is the worst case for impedance uniformity.
The result is an impedance discontinuity. Part of the forward signal reflects back toward the source instead of continuing to the load. On a transmit path that reflected energy shows up as elevated VSWR at the amplifier; on a receive path it shows up as ripple and reduced sensitivity.
Single-Bend vs Repeated-Bend Limits
Datasheets list two figures, and they are often confused:
- Single-bend (installation) radius — the tightest the cable may be bent once, permanently, during installation. This is the figure that applies to a fixed route with a corner.
- Repeated-bend radius — the tightest the cable may be flexed repeatedly over its life without fatigue. This applies to anything that moves: a service loop on a moving antenna, a flyaway system that is packed and unpacked, or a cable that is regularly disconnected.
The repeated figure is roughly four times the single-bend figure. Using the single-bend limit on a cable that will be flexed is a common and expensive error.
Reference Table
| Cable | Outer diameter | Single bend | Repeated bend |
|---|---|---|---|
| LMR-195 | 4.95 mm | 12.7 mm | 50.8 mm |
| LMR-240 | 6.1 mm | 19.0 mm | 63.5 mm |
| LMR-400 | 10.3 mm | 25.4 mm | 101.6 mm |
| LMR-400-UF (UltraFlex) | 10.3 mm | 19.0 mm | 76.2 mm |
| LMR-600 | 15.8 mm | 38.1 mm | 152.4 mm |
| LMR-900 | 23.1 mm | 57.2 mm | 228.6 mm |
Note the UltraFlex variant. LMR-400-UF uses a stranded centre conductor instead of a solid one, which lowers the bend radius meaningfully at the cost of roughly 0.2 to 0.4 dB per 100 ft of additional attenuation at L-band. Where a tight route is unavoidable, that trade is usually worth making — a correctly routed UF cable outperforms a kinked standard cable by a wide margin.
What Over-Bending Actually Costs You
The critical point is that this damage is permanent and invisible. Straightening the cable afterwards does not restore the foam dielectric — it has already taken a compression set. The jacket may show no mark at all. The only way to find it is to measure: a TDR will locate the discontinuity along the cable, and a return loss sweep will show the degradation.
On a VSAT IFL run this typically presents as a link that closes but never quite achieves the expected Eb/No, or one that sits a decibel or two below the commissioning baseline for no obvious reason.
Routing Practice
The single most common violation on real installations is not the cable route at all — it is surplus cable. An installer finishes a run, finds three metres of slack, coils it into a neat 150 mm loop, and cable-ties it to the antenna mount. On LMR-400 that loop is a 75 mm radius against a 101.6 mm repeated-bend limit, and the cable is now permanently compressed at six points around the coil.
If slack must be stored, coil it loosely at or above the repeated-bend radius. The better answer is to cut the run to length and terminate it properly.
Frequently Asked Questions
Is bend radius measured to the inside or the centreline of the cable?
To the centreline of the cable. If a datasheet specifies a 101.6 mm minimum bend radius, that is the radius of the arc traced by the cable’s axis. Measuring to the inside surface will make the bend appear larger than it is and can lead you to under-size the corner. When in doubt, measure the diameter of the circle the cable would complete and halve it.
I bent the cable too tightly during installation but the link works. Is there a problem?
Probably yes, even though the link came up. A single over-bend typically costs a fraction of a decibel and raises VSWR — enough to eat into your rain fade margin without preventing the link from closing in clear sky. The consequence shows up later, during a weather event, when that margin is the difference between a working link and an outage. Sweep-test the run; if return loss at the bend is materially worse than the rest of the cable, replace that section.
Does bend radius matter more at higher frequencies?
Yes. An impedance discontinuity reflects more energy as its physical length becomes a larger fraction of a wavelength. At L-band IFL frequencies of 950 to 2150 MHz a wavelength in foam-dielectric cable is roughly 150 to 270 mm, so a bend of a few centimetres is already electrically significant. The same bend at HF would be almost undetectable. For Ku- and Ka-band systems where the IFL runs at the top of L-band, treat the spec as a hard limit.
How do I find a bend radius violation on an existing installation?
A TDR (time domain reflectometer) is the right tool — it shows the location of an impedance discontinuity as a distance along the cable, so you can walk to the exact point. A cable analyser or VNA sweep of return loss will tell you a problem exists and roughly how bad it is, but locating it requires the TDR. Failing both, physically inspect the run at every corner, every tie point, and immediately behind each connector.
Can I use a tighter radius if I bend the cable slowly and carefully?
No. The limit is a material property, not a technique. The foam dielectric takes a compression set past a certain strain regardless of how gradually it is applied, and the braid distorts the same way. Bending slowly avoids the acute kinking that happens when cable is yanked around a corner, which is worth doing — but it does not change the number on the datasheet.
Coaxial Cable Cut and Terminated to Length
Bravosatcom supplies Times Microwave LMR-400, LMR-600, and UltraFlex variants with N-type connectors fitted, so runs arrive at the correct length with no surplus to coil.