50 ohm cable and a non-50ohm source

[exe]

Hi!

Can I use coax cable with a non-50ohm source? If yes, why? I always thought that a transmission line should have a proper source, and a proper termination, or signal will be distorted due to reflections.  But in this video Dave has only termination, or only 50 ohm resistor at source: . And it seems to be working, which puzzles me.

[TimFox]

The three ideal cases above give frequency-independent amplitude and delay from source to load.
If the frequencies of interest are very low (compared to the reciprocal of the cable’s delay time), with arbitrary source and load resistances, then the cable acts as a capacitance in parallel with the load resistance, and produces a R-C low pass filter governed by that capacitance and the parallel combination of the source and load resistances.  For example, 3 feet of RG-58/U 50 ohm cable has approximately 5 nsec delay and 75 pF capacitance.  At 20 kHz audio, with 600R source and high resistance load, the signal sees a filter down by 3 dB at 3.5 MHz.

[David Hess]

Yes if your source impedance is very low. But you have to terminate it with 50R on load side (and your source should be able to drive 50R in the first place).

A 50R cable can be drive from a 0R source with 50R sink termination, and you will get 100% amplitude.

That also works if the source impedance is very high which is the case with a current output.  At high frequencies the current sees the 50 ohm transmission line impedance and at low frequencies, the current sees the 50 ohm load resistance.

The cable only needs to be terminated at one end assuming that only the ends see the signal.  At midpoint however the reflection will be apparent so termination is required on both ends in party line applications.

[hsn93]

Yes if your source impedance is very low. But you have to terminate it with 50R on load side (and your source should be able to drive 50R in the first place).

A 50R cable can be drive from a 0R source with 50R sink termination, and you will get 100% amplitude.

A 50R cable can be driven from a 50R source with no sink termination, and you will get 100% amplitude.

A 50R cable can be driven from a 50R source with 50R sink termination, and you will get 50% amplitude.

50R cable - 0R source - 50R inline termination


amplitude 50% for first 2 exmaples
amplitude 33.3% for last one ?

what is my mistake?

[Dave]

You mistake is assuming that a 50R cable has 50 ohms of series resistance.
Cable impedance has to do with characteristic impedance, which is a ratio between the series inductance per unit length and parallel capacitance per unit length.

\$Z_0 \approx \sqrt{\frac{L}{C}} \$

[magic]

I prefer to define it as:

The apparent resistance you see when you apply voltage to the cable and wait for its capacitance to charge up, until the charge reaches the other end and all hell breaks loose

Or we can describe "characteristic admittance" as the product of capacitance per meter times propagation velocity. I think it makes sense and shows what's going on.
edit
Or maybe not completely so, by ignoring current

[hsn93]

You mistake is assuming that a 50R cable has 50 ohms of series resistance.
Cable impedance has to do with characteristic impedance, which is a ratio between the series inductance per unit length and parallel capacitance per unit length.

\$Z_0 \approx \sqrt{\frac{L}{C}} \$

no, i assumed the cable impedance = ZL .. 50ohm parralel..

50ohm series is the termination resistor