How bad is a 50/75 ohm impermanence mismatch?

[rwgast_lowlevellogicdesin]

I have a cheap scope and function generator right now, but im waiting in anticipation for my back ordered rigol 1054z. I bought a few bnc cables from radio shack not realizing they were RG59 75ohm. I also got a a 50ohm termination pass through off ebay. When I directly connect my scope and function generator through a BNC using the 50 ohm pass through it cleans up the signal a whole bunch! But this is only at very low frequency's like 1mhz. When I get my rigol .I would like to do some more experimentation, and Id also like to use my cables for SDR stuff. is a 50/75 ohm mismatch going to cause alot of problems especially in higher frequency's?

Also when a cable is rated at a certain impeadence I know I know impeadence is derived from both capacitive recatance and inductive reactance whos formulas require a frequency to be solved. Exactly what frequency do they use to rate cable impeadence?

[ruffy91]

The impedance only changes for very low frequencies where the resistance is big compared to the capacitence and inductance. As soon as you have a few kHz upwards it's only the ratio between capacitance and reactance that gives the impedance. Z0=sqrt((R+L*2*pi*f))/(G+C*2*pi*f)), where Z0 is the characteristic impedance and G is the conductance between inner and outer conductor.

[babysitter]

Yes, this mismatch is there, but how important it is depends strongly on your requirements.

You will have power losses with a big contribution that depends on the frequency:electrical length ratio.
There are frequencies where it will be almost losless and frequencies where the loss is more visible.
If the electrical cable length (about 0.66 times physical length) is 1/2 of the vacuum wavelength of the signal, this mismatch loss will almost disappear and mostly connector and cable losses remain.
Also, the interface of 50 Ohm to 75 Ohm will be visible in TDR applications.

Anyway, use it and replace it when you are in doubt that it is suitable for your needs. I consider it OK for SDR (broadcast is 75 ohm standardized anyway!) and everything else - try it, and if something that you use the cable for looks fishy, consider its the cables fault.

[Dago]

You also need to remember that you should not connect a 50 ohm BNC to a 75 ohm BNC. The middle pin is thicker on the 50ohm one and it will damage the contact on the 75 ohm one. The other way around should be fine (not sure if you get a good connection though).

[Howardlong]

One comment: purpose-built SDRs are 50 ohm. If you are using a device that was primarily designed for broadcast reception as an SDR, then it will be 75 ohm.

Will you notice the difference? Probably not for receive-only, unless you're into weak signal stuff.

Having said that, as a general rule, I do tend to avoid introducing uncertainty into anything if I can, especially if I don't fully appreciate its effects. Although in this case it's very likely to be benign, you might find you become blasé about it and then one day it will make a difference.

[TimFox]

The impedance mismatch (e.g., 50 ohm source to 75 ohm cable to 50 ohm load) becomes important at high frequency, when the length of the cable is substantial compared with the wavelength of the signal.  When calculating the wavelength, the velocity of the cable is somewhat slow compared to vacuum:  in solid polyethylene the velocity is 2/3 c, or 8 inches/nsec.
For pulses, you will see distortion in the shape when the cable electrical length (in nsec) is substantial compared with the pulse rise time.
This is caused by partial reflection of the waveform at the source and load.  With these impedance values, the reflection is roughly 20%.
See  http://www.belden.com/docs/upload/Part-3-Reflection-Coefficient.pdf  for a thorough explanation.

[macboy]

You also need to remember that you should not connect a 50 ohm BNC to a 75 ohm BNC. The middle pin is thicker on the 50ohm one and it will damage the contact on the 75 ohm one. The other way around should be fine (not sure if you get a good connection though).

completely untrue.
This is true to RF (N type) connectors, but absolutely not for BNC. The specs for mating surfaces of both 50 and 75 ohm BNC are identical. Be much more concerned about cheap and poorly made Chinese garbage that is not to spec and can in fact cause issues.

[katzohki]

Well, it depends on what you're going to do how much difference it makes.

If I were you I'd put standardizing on 50 ohms on my list of things to do, but not consider it an emergency.