Use REW to measure cable differences?

[edbayarea]

I've been reading quite a few posts in the REW forum. and all the Help pages as well. I think I can use REW to determine if ICs or speaker cables have any differences in response. Obviously REQ is measuring the room response but if a cable is changed this should be noticeable in the graphs?

Is this a crazy idea?
Which Charts would work best?
I was thinking of only using a single main speaker to test (to minimize the room effects.) Bad/Good idea?


I am fairly prepared: program downloaded, tripod, RS meter, cables.

Thanks the REW gurus for their hard work.

ed

[drakan]

Not crazy, not exactly. The cables will almost certainly measure flat and you will essentially be measuring the response of your sound card. But it will prove, to you at least, that the cables are fine and the no cable is, within its limits, better then another.

[brucek]

Is this a crazy idea?
Which Charts would work best?
I was thinking of only using a single main speaker to test (to minimize the room effects.) Bad/Good idea?

Certainly it will work. The soundcards response influence is removed after the soundcard calibration routine, so a loopback cable will subsequently measure flat. Using the frequency response graph with an exaggerated vertical axis will reveal differences between interconnect cables. No speakers are involved. You simple connect the cable from line-out to line-in of the soundcard.

You would have to accept that the results would only be valid as a comparison against another cable. The only meaningful electrical characteristic that would seperate one interconect cable from another is capacitive reactance. Not knowing the output impedance of the soundcard, nor the capacitance of the setup cable among other things would stop you from concluding that a specific cable was good, but you could certainly say one cable was better or worse than the one used to create the calibration file.

REW is also useful in measuring the response of any line level device.

brucek

[edbayarea]

Thanks much guys for the replies.

Brucek, that makes sense with the loopback cable. I guess I utimately want to test the speaker cable combination.

I am on a macbook pro and after a few hours last night I got as far as calibrating the card. but no luck with an actual test too many error messages. I think I confused the two different approaches: 1. Soundcard calibration file method. vs. 2. Left Channel Calibrate method --I was following the steps On the Help page.

Going to start all over. and probably just use #2. Left Channel Calibrate method

Regards, Ed

[brucek]

I guess I utimately want to test the speaker cable combination.

Well, there's an issue there. From the line-out to the line-in of a soundcard, it is considered a high impedance interface, also known as a voltage bridge. The ouput impedance of the line-out will be quite low and the line-in will be quite high. The only effects from interconnect cables would be capacitive reactance (and even then it would be quite low).

Generally, with speaker cables, the concern would be inductance and its associated reactance that would become a filter in a low impedance connection from a power amp to a speaker. Capacitance is not important in a speaker cable. We want to know its inductance. So, to introduce a speaker cable across a soundcards interface would produce non-meaningful results. Inductive reactance would have no effect on this connection and would be unmeasureable.

Going to start all over. and probably just use #2. Left Channel Calibrate method

I strongly suggest using method #1.

brucek

[thewire]

Here is a measurement using ETF5 to determine why my front left and right speakers kept using their built in protection. This is weeks old. When they would go into protect mode my amplifiers would click on and off very fast. It was done using two bridged MA-500 mono amps on each channel for 360W @ 8ohm and a total of 720W and 1000W peak for the two (I think). The two speakers were Polk RTi70's and measured with a Digital RS meter @ 9 feet. The setup was a Y-cable to a pair of RCA cables into the soundcard and the calc file was the analog RS meter calc file. The test was to done to determine if I needed to use my extra wires in the ceiling and from the walls that were left over from passive bi-amplification for any reason. They were there so I wanted to use them. The test was of one speaker. I later needed to better strip the wires to get each side to match more closely. I also removed the extra wire connections from the amps after the test.

First graph - Pair 12AWG wire from amps - pair of in-wall wire to wall plates - pair of 12AWG wire to the speakers.



Second graph - Pair 12AWG wire from amps - single in-wall wire to wall plates - single 12AWG to speaker.



After I removed the useless wire my amplifiers and the speakers they quit using protection. Ignore the 3K area.

This proved to me that bi wiring is useless and can actually harm my setup. So I archived my goal.