The oscilloscope showed it clearly: the received pulse from channel 64 had a visibly slower rise time than channel 1. Same transducer elements, same electronics, same cable length—2.3 meters of 42 AWG micro coaxial cable . The only difference was the cable routing. Channel 1 ran straight. Channel 64 was routed along the outer layer of the bundle where it picked up about 8 pF of stray capacitance from adjacent shield-to-shield coupling. That extra 8 pF per meter added roughly 18 pF total to the channel's capacitive load—enough to roll off the 15 MHz harmonic content that the system was trying to image with.
Low capacitance micro coaxial cable isn't an abstract spec line on a data sheet. It directly determines how much bandwidth your cable preserves between the transducer and the receiver—and in medical imaging, bandwidth is image quality. Every picofarad per meter adds to the RC time constant that filters your signal, and on high-frequency systems operating above 10 MHz, the difference between 72 pF/m and 55 pF/m cable is visible on the display.
The Physics: Cable as Low-Pass Filter
A coaxial cable has distributed capacitance between the center conductor and the shield—it's inherent in the geometry. This capacitance, combined with the source impedance driving the cable, forms an RC low-pass filter. The −3 dB bandwidth of this filter is f₋₃dB = 1/(2π × R_s × C_total), where R_s is the source impedance and C_total is the cable's total capacitance (capacitance per meter × length).
Let's work through a real example. A 42 AWG FEP-insulated cable at 72 pF/m, 2 meters long, driven by a 50Ω source:
C_total = 72 pF/m × 2 m = 144 pF. f₋₃dB = 1/(2π × 50 × 144×10⁻¹²) = 22.1 MHz.
That's the frequency at which the cable attenuates the signal by 3 dB—half the power. At 15 MHz (common harmonic imaging frequency), the attenuation is about 1.2 dB. Not catastrophic, but not nothing either.
Now the same cable with ePTFE dielectric at 55 pF/m:
C_total = 55 pF/m × 2 m = 110 pF. f₋₃dB = 1/(2π × 50 × 110×10⁻¹²) = 28.9 MHz.
That 17 pF/m reduction in capacitance bought you 7 MHz of additional bandwidth. At 15 MHz, the attenuation drops to about 0.6 dB. In ultrasound terms, that 0.6 dB improvement translates to roughly 3–5% better sensitivity at harmonic frequencies—the frequencies that carry the best contrast resolution in tissue harmonic imaging.
A Customer's Bandwidth Problem—and How Cable Capacitance Was the Root Cause
About two years ago, an ultrasound OEM came to us with a puzzle. Their new 15 MHz linear array probe was meeting all specifications on a 1-meter test cable, but when they installed the production 2.5-meter cable assembly, harmonic image quality dropped measurably. Sensitivity was down by about 2 dB at 12 MHz receive, and the lateral resolution at 2 cm depth degraded from their target of 0.3 mm to 0.45 mm.
The cable was electrically correct—50Ω impedance, proper phase matching, good shielding. But the cable's capacitance was 74 pF/m (FEP dielectric), giving a total cable capacitance of 185 pF over 2.5 meters. Combined with the transducer's source impedance (which was closer to 100Ω per element, not 50Ω—an important detail), the RC bandwidth was:
f₋₃dB = 1/(2π × 100 × 185×10⁻¹²) = 8.6 MHz.
At 12 MHz receive, the cable was already 4 dB down. At 15 MHz harmonic, it was almost 7 dB down. The cable was acting as a significant low-pass filter, and the engineer had overlooked the source impedance detail—using 50Ω in the calculation instead of the actual 100Ω element impedance.
The fix was switching to ePTFE-insulated cable at 54 pF/m . New total capacitance: 135 pF. New bandwidth: 11.8 MHz. The system recovered to within 0.5 dB of the 1-meter test cable performance. Not perfect—the cable still has some attenuation effect at 15 MHz—but within the system's link budget and clinically acceptable.
The lesson: always calculate RC bandwidth using the actual source impedance, not the cable's characteristic impedance. Transducer elements can have source impedances of 50–200Ω depending on the element design, backing, and matching layer. Higher source impedance makes the cable capacitance effect worse.
What Determines Cable Capacitance?
Cable capacitance is fundamentally set by two things: the dielectric constant (Dk) and the dielectric geometry (thickness and concentricity). The formula for coaxial cable capacitance is C = (2π × ε₀ × Dk) / ln(D/d), where D and d are the outer and inner conductor diameters.
For a fixed impedance (say 50Ω), the D/d ratio is locked by the impedance equation. That means the only lever you have for reducing capacitance is the dielectric constant. Lower Dk = lower capacitance, period. This is why every low-capacitance micro coaxial cable uses either ePTFE or foamed dielectric—both achieve lower effective Dk by introducing air into the dielectric structure.
| Dielectric Material | Dk (typical) | Capacitance (pF/m) | VoP | Best Application Fit |
|---|---|---|---|---|
| Solid FEP | 2.15 | ~72 | ~69% | General purpose, cost-optimized, best VoP consistency for phase matching |
| Solid PFA | 2.06 | ~69 | ~70% | High-temperature Applications , slightly lower capacitance than FEP |
| Foamed FEP | 1.5–1.7 | ~52–58 | ~77–82% | Low-capacitance needs with cost sensitivity; good bandwidth improvement |
| ePTFE (moderate expansion) | 1.5–1.6 | ~53–57 | ~79–82% | Premium low-capacitance; excellent flex life |
| ePTFE (high expansion) | 1.3–1.4 | ~45–50 | ~84–88% | Maximum bandwidth; limited availability, highest cost |
| Solid PTFE (reference) | 2.1 | ~71 | ~69% | Legacy designs; limited use in new micro coaxial Applications |
The Trade-Off Nobody Warns You About
Lower capacitance means lower Dk, which means ePTFE or foamed dielectric. Both introduce manufacturing trade-offs that solid FEP doesn't have.
ePTFE's microporous structure is inherently less uniform than melt-extruded FEP. The Dk varies with porosity, and porosity varies with processing conditions. In a 128-channel bundle where every channel needs ±1% VoP matching, ePTFE's wider VoP distribution means more channels fall outside tolerance, more sorting is required, and first-pass yield is lower. We've covered this in detail in our phase matching guide —the short version is that ePTFE gives you better capacitance but worse phase matching consistency per unit of manufacturing effort.
Foamed dielectrics have their own issue: crush resistance. The air cells that lower the Dk also make the dielectric mechanically weaker. Under the radial pressure of shield braiding and cable bundling, foamed dielectric can compress locally, raising the effective Dk and increasing capacitance at those points. The TDR trace of a poorly processed foamed-dielectric cable shows impedance bumps at every point where the braid tension was uneven—each bump is a localized capacitance increase.
We've seen engineers specify ePTFE low-capacitance cable for Applications where the bandwidth improvement isn't actually needed—they saw "lower capacitance" on the data sheet and assumed it was universally better. For a 2 MHz cardiac ultrasound probe with 2-meter cables, the FEP cable's RC bandwidth (22 MHz) is already 11× the operating frequency. Switching to ePTFE at higher cost and tighter manufacturing tolerance for a bandwidth improvement the system can't use doesn't make engineering sense.
When Low Capacitance Actually Matters
Based on working through hundreds of cable specifications, here's when investing in low-capacitance micro coaxial cable pays off versus when it's wasted spend:
Worth it: Operating frequency above 10 MHz, cable length above 1.5 meters, transducer source impedance above 75Ω (higher impedance makes the RC filter worse), or harmonic imaging modes where receive frequency is 2× the fundamental. These conditions—especially in combination—create scenarios where standard FEP cable's capacitance measurably limits the system.
Not worth it: Operating frequency below 5 MHz, cable length under 1 meter, or applications where the receiver bandwidth is already narrower than the cable's RC bandwidth (meaning the electronics are the bottleneck, not the cable). In these cases, standard FEP gives you equivalent system performance at lower cost and with better phase matching consistency.
Edge case: Portable/POCUS ultrasound probes operating at 5–12 MHz with 1.5-meter cables. FEP cable is marginal but usually acceptable. ePTFE provides measurable improvement but the system-level impact depends on the specific electronics design. This is the application zone where a prototype comparison test—building two identical probes with FEP vs. ePTFE cables and comparing image quality back-to-back—saves more time than calculations.
If you're designing a medical imaging cable and aren't sure whether the bandwidth improvement from low-capacitance cable justifies the cost premium, share your operating frequency, cable length, and transducer source impedance —we'll run the RC bandwidth calculation and tell you whether standard or low-capacitance cable is the right choice for your specific system. It's a 10-minute analysis that can save weeks of prototype iteration.
So when a cable supplier quotes "low capacitance" without giving you the pF/m number and the dielectric type—ask. And when they give you the number, plug it into the RC bandwidth formula with your actual source impedance and cable length before deciding it matters. The answer might surprise you in either direction.
Related Products
FRS Technology supplies these constructions to medical, NDT and industrial OEMs worldwide, with engineering support at the design stage. Related products:
Not sure which construction fits? Talk to our engineers - share your channel count, frequency and space constraints.
