Home Theater Forum and Systems banner

1 - 5 of 5 Posts

·
Registered
Joined
·
32 Posts
Discussion Starter #1
Hello,

please take a look at this graph:



The light green line is a normal measurement from 0 to 25kHz where I directly connected the in- and output together with a correction file which gives this traight line.

The black one is the RTA measurement.

What you can see here is that instead of a also straight line the dB's are increasing with the frequency.

Isn't that strange? I would expect a straight line also....

What you can not see (I would have to make a movie for that) that especially in the low regions the black line is moving going up and down.

I would expect also here that the RTA line would be stable and straight.

Why isn't this so?

I think it's of no use using RTA with a mic as long as a direct connection isn't stable and straight.

Am I correct or what should I do for a correct RTA?

Thank you!
 

·
Registered
Joined
·
678 Posts
From your description, I infer that you are using the same loopback wire connection, and taking an RTA measurement immediately after verifying the soundcard calibration. What a strange idea, I had not thought of trying that.

Measuring RTA on a loopback connection, I believe you are seeing the random noise of your computer itself, somewhere on the output or the input side. As you can see from your chart, this is 100dB lower than your -12dBFS test signal. As this is a logarithmic scale, different signals do not add linearly -- two equal level signals added together increase the level by 6dB -- so this noise 100dB down will not affect your measurement results. Unless you are measuring the background noise in a highly soundproofed acoustic isolation chamber, I expect the noise floor in your room will be much higher than this. As I recall, John has indicated that a noise floor in an impulse curve at -80dB is good, at your loopback noise floor is much lower than this.

Bill
 

·
Registered
Joined
·
32 Posts
Discussion Starter #4
Thank you both for your (quick) reply.

Well.....i made a BIG mistake....(I'll admit..... :) )

When you press RTA there is no signal coming out of REW!!! (this IS the case on my Behringer, that's why I didn't think of it....)

So, my graph above only shows the noise from the computer (well thought of Bill!)

So, below are some pic's with RTA, loopback, correction file in place AND WITH A SIGNALGENERATION (Shame on me :R)

This is the 'pink noise' setting. I have no clue why you would like to use this shape but here it is.....



This is 'pink PN'. I think this one is the one to use if you want to set-up your room environment as a line as straight as possible:



And this is 'white PN'. Also no clue why you want to use this but, again, here it is....



Thanks again!
 

·
Premium Member
Joined
·
6,711 Posts
That first pink noise trace you showed is the "speaker cal" signal, it is band limited centred around 1kHz, hence the shape, the full range Pink Noise does (eventually) average to a flat trace. Pink PN is, however, the correct signal to use, with the RTA window set to Rectangular. It requires little or no averaging. If you would like to see the RTA plot shown with levels that are easily compared to sweep measurements tick the "Adjust RTA Levels" box.
 
1 - 5 of 5 Posts
Top