Tweakers' Asylum

Posted by Maxamillion (A) on May 13, 2010 at 04:15:22

In Reply to: RE: Filters Done - Pictures posted by JimBop on May 12, 2010 at 16:54:55

Well, first you need to determine what the characteristic impedance of your cables is. Perhaps your cable supplier can tell you. If not, you can calculate it if the supplier will tell you the series inductance (L) and shunt capacitance (C) of the cables. Then, the characteristic impedance (Z) may be calculated according to this formula:

Z=sqrt(L/C)

In my case, Zu supplies the L and C values for their Ibis cables, so I was able to calculate it.

If you cannot get that information, the characteristic impedance may be measured, but you will need specialized high-frequency equipment for that, as we are now talking about the RF behavior of cables, which is governed by transmission line theory above roughly 100kHz. The key tenet of all transmission line theory is that the source impedance must be equal to the load impedance in order to achieve maximum power transfer and minimum signal reflection at the destination. In real world case this generally means that the source impedance is the same as cable impedance and the value of the receiver in another end of the cable has also the same impedance. What you are attempting to do here with the matching is to ensure that the maximum amount of RF is transferred into the filter and dissipated by the resistor, instead of being reflected back down the cable due to an impedance mismatch. What's cool (in my nerdy world!) is that when you match the impedance properly, the signal thinks it's in an infinitely long cable, because it never encounters anything that will reflect it.

Here's a nice little summary of the theory as applied to coaxial cables; the theory still applies to our speaker cables at RF frequencies, which is what we are attempting to filter out:

• http://www.epanorama.net/documents/wiring/cable_impedance.ht   (Open in New Window)

Edits: 05/13/10