开云体育

Hello,

I'm trying to measure a 40 degree delay line on 1.83 MHz for a 75 Ohm cable.
Would the S21 Phase reading be reliable given that the in/out of the NanoVNA is 50 Ohm?

I cut the cable for S21 Phase of 40.3 degr which is close enough.
The issue is that if I measure the same cable using the L/4 method ( min(|Z|) I do see a different VF than what the S21 Phase would imply (about 5% difference in VF from S21Phase (VF = 0.82), min(|Z|) - L/4 (VF = 0.84) and cable datasheet (VF = 0.87)).
The cable length is 14.89 m. Frequency for min(|Z|) is 4.26 MHz. S21 Phase is -40.3 degree.

Thanks,
Dani

Have you tried using the TDR method to find the VF of the cable?

Roger

Hey Roger, thanks for your answer.
Running a TDR on the cable between 5 and 100 MHz (505 points) shows the correct length of cable (14.89m) at 0.859 VF. (which is a 4th VF value..)

Dani

On Wed, Dec 30, 2020 at 12:54 PM, Dani YO5LD wrote:

Hey Roger, thanks for your answer.
Running a TDR on the cable between 5 and 100 MHz (505 points) shows the
correct length of cable (14.89m) at 0.859 VF. (which is a 4th VF value..)

Dani

Velocity factor does change with frequency. Here is a plot of one type of Belden 75 ohm coax

Roger

I did some Velocity Factor (VF) measurements today on 9.4M of Belden RG-58A/U. The Belden specs give the VF as .66 or 66% of the speed of light in free space for this cable type.

The first step was to measure the round trip delay using the Transform Function in the NanoVNA-H4. It was 97.6 nanoseconds or 48.8 nanoseconds one-way. In free space it takes 31.34801 nanoseconds to go 9.4M. So the VF is 31.34801/48.8 = .6424 or 64.24%

I connected the NanoVNA-H4 to NanoVNA Saver and the VF was measured at 64.37%.

The next step was to use a signal generator and a 2 channel scope to measure the differential arrival delay in nanoseconds due to the cable over a range of frequencies - 1 to 20 MHz. The results are attached as a graph and show VF increasing with frequency.

The second attachment is a plot for the same cable type using the excellent TLDetails program by AC6LA


You can clearly see that my results have the same overall shape as the the TLDetails plot but are slightly lower. This can be due to manufacturing differences or that my cable which did not have the same Belden cable code was slightly different.

The measurements done by the NanoVNA or traditional fast rise time-short pulse methods both use very high frequency stimulus so they yield VF numbers that are higher than those at low frequencies of only a few Megahertz.

Roger