Characteristic impedence matching

[PurdueRy]

Ok, here's the deal:

I am taking a class at Purdue University about electromagetics and such and we are learning about transmission lines. The topic of impedence matching came up and I was reminded about a common cabling topic here.

Correct me if I am wrong, but it is generally accepted that a video cable works great in audio purposes too, but a audio cable should not always be used as a video cable. For instance, it may not be good to use a composite+stereo cable as a component cable due to some cables having the audio cables be 50 ohm cables and 75 ohm video cables.

I have always taken this as fact but in class we are discussing reflection of electric and magnetic waves at a impendence mismatch. I asked and the professor stated that is did not matter if the characteristic impedence was higher or lower, either way the amount of signal transferred would be lower than with an characteristic impedence match.

Am I missing something here? Why is it said that using audio cables as video cables is not good but the other way around is just fine?

NOTE: Copied from AVSforum thread...sorry if there are references made that regard that forum

 

[TuxDave]

I have no real concrete information to back this up BUT, I'll throw out my guess. Impedence is a function of frequency SO perhaps in the audio frequency band, the audio and video cable have roughly the same impedence. However at the higher frequency bands for video, the impedance of the audio cable is significantly higher than that of the video cable and then you get an impedence mismatch.

 

[PurdueRy]

Originally posted by: TuxDave
I have no real concrete information to back this up BUT, I'll throw out my guess. Impedence is a function of frequency SO perhaps in the audio frequency band, the audio and video cable have roughly the same impedence. However at the higher frequency bands for video, the impedance of the audio cable is significantly higher than that of the video cable and then you get an impedence mismatch.

Ah but characteristic impedence is not dependent on frequency...they are two different things(impedence and characteristic impedence)

But good guess! Thanks :thumbsup:

 

[TuxDave]

Originally posted by: PurdueRy

Originally posted by: TuxDave
I have no real concrete information to back this up BUT, I'll throw out my guess. Impedence is a function of frequency SO perhaps in the audio frequency band, the audio and video cable have roughly the same impedence. However at the higher frequency bands for video, the impedance of the audio cable is significantly higher than that of the video cable and then you get an impedence mismatch.

Ah but characteristic impedence is not dependent on frequency...they are two different things(impedence and characteristic impedence)

But good guess! Thanks :thumbsup:

Nuts... no wonder I didn't do well in my high speed links class. Stupid transmission lines.

 

[f95toli]

Audio cables do not have a well defined impedance, since audio only covers 20-20 000 Hz (roughly) the charachteristic impedance is irrelevant (since the charachteristic wavelenghts are usually much longer than the cable you don't have problems with e.g reflections). Hencem there are many types of audio cables, some are made from plain 75 ohm video cable and should work well for video as well, some are e.g twisted wires, various braides etc, strange insulation (e.g water) so the impedance varies widely.
There is one exception and that is digital audio (S/PDIF), here a 75 Ohm cable should be used since the frequencies are much higher (it is a digital signal after all).

Hence for video you want 75 ohm because the frequencies involved are high, for audio you don't need to care.

Btw, discussing technical details about cables on a hifi forum is usually not a good idea, at least not if you expect answers that make sense. There is usually more faith than fact in the answers you get.

 

[Born2bwire]

Using an audio cable in place of a 75 Ohm video cable is not going to make that much of an impact. I do not know offhand what the frequency of an analog video signal would be, but it cannot be very high in frequency. Considering that a TV refresh rate is 60 Hz, you really do not need a very high frequency in your analog signal (i.e.: it is probably in or below the Kilohertz range). With this in mind, the time that it takes for reflections that arise from impedence mismatch to settle is insignificant to the frequency of the wave. If we have a 30KHz signal, the wavelength is on the order of 10 Km in length in freespace. So over the length of your cable, the signal is basically going to be DC and the reflections are not going to introduce a significant noise. If you start venturing into the Megahertz region and above, then you really need to start taking note of impedence matching, because at that stage the wavelengths of your signal are more comparable to the length of your transmission line and the change in your signal is going to be more comparable to the settling time of your reflections. So if you are handling digital cables, you will want to take note of your impedence matchins.

 

[PurdueRy]

Originally posted by: Born2bwire
Using an audio cable in place of a 75 Ohm video cable is not going to make that much of an impact. I do not know offhand what the frequency of an analog video signal would be, but it cannot be very high in frequency. Considering that a TV refresh rate is 60 Hz, you really do not need a very high frequency in your analog signal (i.e.: it is probably in or below the Kilohertz range). With this in mind, the time that it takes for reflections that arise from impedence mismatch to settle is insignificant to the frequency of the wave. If we have a 30KHz signal, the wavelength is on the order of 10 Km in length in freespace. So over the length of your cable, the signal is basically going to be DC and the reflections are not going to introduce a significant noise. If you start venturing into the Megahertz region and above, then you really need to start taking note of impedence matching, because at that stage the wavelengths of your signal are more comparable to the length of your transmission line and the change in your signal is going to be more comparable to the settling time of your reflections. So if you are handling digital cables, you will want to take note of your impedence matchins.

Video signals easily surpass 4 MHz

 

[PowerEngineer]

Obviously my expertise lies with much lower frequencies, but we do actually consider the propagation of high frequency waves on power lines that result from lightning strikes and switching events. It's been a while since I studied the theories, but I do remember that the surge impedance of a conductor the square root of the ratio of its inductance to its capacitance [ (L/C)^1/2 ]. It is true that any change in this impedance as a signal passes along a cable will set up reflections at the point of change that essentially attenuate the signal continuing down the cable. (For us power engineers, these reflections back on top of the incoming surge can add to the peak voltage experienced and so we need higher insulation levels.)

What you say about video cables being suitable for both audio and video, while audio cables aren't very good for video makes sense to me. But if they are of different impedances, then you still have the reflections and attenuation problem caused by the impedance change to deal with. Ideally you'd use some sort of impedance matching connector (which isn't much different in theory than the splitters commonly used for video signals).

 

[Born2bwire]

Originally posted by: PurdueRy

Originally posted by: Born2bwire
Using an audio cable in place of a 75 Ohm video cable is not going to make that much of an impact. I do not know offhand what the frequency of an analog video signal would be, but it cannot be very high in frequency. Considering that a TV refresh rate is 60 Hz, you really do not need a very high frequency in your analog signal (i.e.: it is probably in or below the Kilohertz range). With this in mind, the time that it takes for reflections that arise from impedence mismatch to settle is insignificant to the frequency of the wave. If we have a 30KHz signal, the wavelength is on the order of 10 Km in length in freespace. So over the length of your cable, the signal is basically going to be DC and the reflections are not going to introduce a significant noise. If you start venturing into the Megahertz region and above, then you really need to start taking note of impedence matching, because at that stage the wavelengths of your signal are more comparable to the length of your transmission line and the change in your signal is going to be more comparable to the settling time of your reflections. So if you are handling digital cables, you will want to take note of your impedence matchins.

Video signals easily surpass 4 MHz

Ok, finally found a site that stated what the frequency of the unmodulated signal is, Text, which is 0 - 4.2 MHz. Before I could only find out the bandwidth of the modulated video/sound signal that is broadcasted. Anyway, 3 MHz is still 100 m in wavelength so the concept is still the same.

EDIT: Hell, why not just hook it up and find out?