time domain reflectometry alternative

[yefj]

Hello, i have read that TDR is problemtaic because if we have many transmission lines in series which cause didcontnuety to the sourse resistanse then the closer TL's will mask the reading of the far TL from the source,so is when the rise time of the source is larger the TL delay.



So what is the alternative to the TDR? what mehtod is used to overcome those problems and see the impedance at every location of the system?
UPDATE:

An example is shown in the link bellow
we have 80,30 ,50 TL's and 80,30 create masking problem for reading the 50ohm TL

1 hosted at ImgBB

Image 1 hosted in ImgBB

ibb.co

Thanks.

 

[FvM]

Masking doesn't exactly hit the point, I think. Starting with the second step, when the 30 ohms line appears in the response, the response is produced by superposition of multiple reflections. You can still determine the individual line segment impedances by solving a system of equations.

Although we can solve the problem this way, I doubt that the setup is a useful TDR application. Typical TDR applications are e.g. fault finding and performance check in a basically impedance matched cable system.

 

[ferdem]

I do not know the subject well but I would ask how do you find the impedance profile using such a simple formula, Ro=zo*(1+rho)/(1-rho)? Extraction of the impedance profile must not be that simple. There are multiple reflections making the problem complicated. Yes, that formula can be used to calculate load impedance by using rho and Z0. However, extraction of an impedance profile is a much different problem. Some kind of inversion or deconvolution methods must be employed rather than a simple formula.

 

[FvM]

You are right, the voltage Vdr observed at the left side of Line 1 is generated by multiple superimposed reflections.

You can calculate a reflection factor and a corresponding impedance for Vdr, but it's not equal the impedance profile along the transmission line system, except for the very first 80 ohms segment.

--- Updated Mar 16, 2021 ---

While the first rho value is caused only by a single reflection rho1 = r1 = (80-50)/(80+50) = 0.231, the second level is already the sum of the first reflection plus a reflection at line 2, modified by two transmission factors
rho2 = r1 + t1*r2*t1r = r1 + (2*80/(80+50)*(30-80)/(30+80)*2*50/(80+50)) = 0.231 - 0.430 = -0.200

 

[biff44]

You are right, multiple discontinuities along a transmission line will reflect and cause confusion. Make sure the signal source itself is well matched to 50 ohms, so you do not get re-reflections at the source!

BUT you might be able to gate out the spurious reflections, and concentrate only on the one reflection you want.

Other than that, things get complicated in a hurry, as you would need to dump it all into some DSP board, and employ multichannel cancelation techniques, possibly with a more complicated waveform such as a frequency ramp, or OFDM signal, etc.