problems with acoustic measurements

[wintermute]

Re: Re: Re: Started a new thread with this stuff

Hey, any number are welcome. Always.

No, no way. I know this setting, and I've never switched it on. Moreover, see what I've posted about 8KHz and 10KHz.... I'm getting absolutely uniform amplitude at all those frequencies. From 1KHz to 10KHz, the same amp volume gets me the same recorded amplitude.

OK was worth a shot, note that the pre-emphasis is only used with the MLS signals so won't affect the sine wave results

No, I didn't try 200ms, but I did try -100ms. I am now desperate enough to take an exported dump of the file and look at the actual text values to see where the high readings go. In software development, we say, "When all else fails, look at the source."

good idea

Yes, but then how do I get clean sine waves at 10KHz?

you would be surprised how difficult it is to see spme distortion in a sine wave I was getting 1.5% THD on my amp when I had a problem with it (measured with RMAA) and a sine wave looked perfect.

but 2M is probably ok for lenght of the mic cable, this would be one area that having a preamp can improve, becuase you can have a longer line level cable going from the preamp to the computer, rather than having to have the mic cable long.

Tony.

 

[tcpip]

In-room freq analysis using RMAA

I ran RMAA on the speaker I was testing, using the same sound card and mic that I've been struggling with. I wanted to see whether I'd get any rolloff at the high-freq end in this case. This is what I got:

An externally hosted image should be here but it was not working when we last tested it.

It's clear that my earlier SPL measurements which rolled off from 3.5K upwards were definitely due to some problems with the measurements, not due to any limitations of the sound card, mic, mic cable, cable capacitance, mic preamp, audio device drivers, etc, etc, etc.

Well, it's now one less set of variables to worry about, but we're back where we were. Why is the damn thing not working in Speaker Workshop?

 

[tcpip]

Looked at the ASCII text for the impulse reponses

I exported the impulse responses for both these sampling rates, and looked at the ASCII text numbers in the exported files. The actual files contain a long line of a single column of numbers, in floating point format. There are 4096 numbers in each file.

I removed the first and last few lines from each file, and then pulled out the largest positive and largest negative numbers from each file, to see what the range of values was.

I discovered that for the 24khz sampling rate, the values ranged from
0.001453261
0.000958403
:
:
-0.00031666
-0.000459182

And for the 44.1Khz sampling rate, the values ranged from
4.35194E-05
4.08338E-05
:
:
-4.68154E-05
-4.90432E-05

Therefore, it's pretty clear that the raw numbers in the 24KHz sampling rate impulse are ranging across a very wide range of values, while the raw numbers for the 44.1KHz sampling rate are ranging across a very small range of values.

This tallies with what I see in the graphs. The 44.1kHz sampling rate graph shows only noise... no real signal at all.

 

[wintermute]

Hi TCP/IP,

well I'm pretty well out of ideas for now... maybe you should put it down for a while, it may just come to you when you aren't thinking about it.

I haven't had either of the problems you are experiencing now, and I'm at a bit of a loss as to what is causing them. have you tried doing the measurements using your spl meter and line in jack instead of the microphone jack??

Tony.

 

[tcpip]

It's working now, but I don't know why

Hi guys,

I've been away for about a week now, trying to get things to work. It worked. I can now get measurements at 48KHz which look believable. But I am still not clear why it's working now.

See my full report here. I thought I'd not waste bandwidth by posting the same report again here.

Any comments you may have on my experiences will be most welcome. Please let me know what you think, and what else I should watch out for. And if you can give me any explanations about why it's working now, I'll be most interested.

 

[wintermute]

Well done The graphs look good. seems you are getting some sort of interference or diffraction effect on your tweeter, but that is more than likely the speaker not the measurement.

I hate it when something just starts working and you don't know why!! have you downloaded quickmix?? I have saved the various settings I use for measurements (it allows you to save your current config of your sound card) and always load them before doing measurements of various types.

sorry can't shed any light on what changed.... One thing though (which you probably already know/realise) It is VERY important when trouble shooting to only change ONE thing at a time!! it takes longer but IMO is the only way to go.

Tony.