Measuring, again...

Maybe a dumb question, but I'm not sure how to measure my control room response. First thing I'm confused is can I measure close response of my Tannoy AMS8 (already in soffit )? I did -6db decompensation calculated from Barefoot Wall Bounce Calculator, and here's the graph of my filter. Now, I have to trim my HF crossover (adjustment of the relative level on the HF channel, shelving +/-3db, having break F at 2khz), but I'm not sure do I have to measure with Pink noise or something else :? BTW, I have NTI MiniSPL calibrated mic and Minirator, Minilyser, and of course some software to help about. After that measuring I will do the rest of measuring and post my results to finally finish topic about my studio building :D

Have you watched the videos at realtraps.com? The one on treating the control room is awesome. http://www.realtraps.com/videos.htm the first one on that page is great.

Thanks, unfortunately I have really poor net, so can't dl so big files... :cry: Hope that will find some way to DL. In the meantime will be nice to hear (read) some comment about :wink:

Well the basic concept of the video is this: Set a mic at the mixing position (where your head goes) and play a sweep of frequencies. At the bottom of the page I linked for you there's a Sonar file with that. It goes from 40hz up to 300hz, increasing by 1hz every second. You play the audio at a loud volume while recording to a new track from the mic. Once it's done playing you look at the wave file created from the recording. You can actually see peaks and dips in the frequency response of your room. Do that before you treat the room, and after you treat the room to see how much it improved. Ethan says in the video that if you can get your room to within +/- 6db, you're in good shape. It's a really good video. It's quite an easy technique for testing your room's response.

Or, you can take the original signal and deconvolve it with the recorded versin of that signal, turn it into the impulse response, and garner any and all information from it with the help of a friend (me) ;) To do this, you ill need a test signal that I provide, a test omni, and the ability to play the tone back through your speakers and record it without monitoring the recorded signal. If you're interested, let me know.

Thanks guys :D

It goes from 40hz up to 300hz

That part of measuring I understood and there's no magic :wink: But, can I use same method for "balancing" Lf and HF channel of my active crossover? Generating the sweep from lets say 100hz to 20khz. Seems reasonable for me, but have to ask.

Or, you can take the original signal and deconvolve it with the recorded versin of that signal, turn it into the impulse response, and garner any and all information from it with the help of a friend (me) To do this, you ill need a test signal that I provide, a test omni, and the ability to play the tone back through your speakers and record it without monitoring the recorded signal. If you're interested, let me know.

Only thing I understood is that you became my friend, what I really appreciate. :wink: Yes, I'm interested a lot, let me know about more :roll:

Hey Moby! I'm guessing that the software you have is capable of doing what I had in mind, but if not, or if you'd ratehr have my help, that's cool with me. I see you have a suitable mic already. If you'd like me to help you collect the raw data, all you have to do is play a test tone that I'll give you through your speakers, record it, get it back to me, and I'll do the rest. Let me know.

No problem, I have a lot of software available, Lynx, RME and Digigram cards, and the pocket analyser, but not sure what you have in mind. So, will be nice to try. What type of test tone you are talking about? I have a lot of software based and standalone type of generators available, but if test tone you are talking about is some special one, please let me know where I can DL , so I will post you the recorded one soon. Thanks for helping :D

The tone I had in mind was just a 20 Hz - 20 kHz exponentially swept sine wave. Let me know if you need one.

Thanks David. Sure, Sweep 20hz-20khz shouldn't be problem to me. Just let me know about duration of the sweep and maybe about usual db acoustical pressure suitable for this kind of measuring. BTW, if your method of analysing require original and recorded response file, maybe will be easier to use sweep generated by you, so when I send you response you already have the original. Also tell me does bit depth and s. rate of response matters, or not? Thanks :D

Email me and I'll send you a sweep. The sweep will be 12 seconds long. Just get it playing back at about 85 dB SPL or so. Really, you should just play it back as loud as you can while still having zero distortion. Record it at 16/44.1. Leave plenty of silence at the end after the tone has stopped playing - this helps me to estimate the SNR. Be sure to document the exact dimensions of the room and the exact positions of the speaker and microphone. This will be fun!

A little more. Place the mic at the listening positon, at the point of your (hopefully) equilateral litening triangle. Measure left and right separately. I remember that you have slow internet. Sending wave files may be dumbersome, so you should take a look at FLAC if you don't know about it already. It's a lossless compression for audio that can reduce wave files to about 30% of their size, much better than the best general purpose data compressors like ZIP, RAR, etc.

Thanks David :D Check the mail please

Hi Moby, here's some data from the files you sent me.

Great David, only I don't know to understand multicolor files. I presume that the last one i reverberation, but not sure on which spectrum. Will be nice if you can add some comment about. Not just you, anyone who know a little aboutt :wink:

WTF?!? I swear I typed comments into each of these attachments! Oh well. The ones that start with 'LF' show the decay in time of the low end of your speakers. The red line represents the response from about 0-300 msec, and the blue line represents time from about 200-500 msec, so basically, it's the response plot updated every 20 msec. The cooler the color, the later in time. This is useful for identifying resonances such as modes. I'm sure you know 1/3 octave plots. The last one is called the ETC (Energy Time Curve) It is another way of looking at the impulse response in the time domain (the other graphs were in the frequency domain). It shows that all the energy is at -60 dB afetr about 150 msec... a very short time.

Cool :D Of course I did this to check and improve my acoustics, so have to ask some questions about.

The ones that start with 'LF' show the decay in time of the low end of your speakers. The red line represents the response from about 0-300 msec, and the blue line represents time from about 200-500 msec, so basically, it's the response plot updated every 20 msec. The cooler the color, the later in time. This is useful for identifying resonances such as modes.

I see that some bass traps will be nice... But, what I have too look for? Is that 110.2hz fundamental freq. I have to treat or I have to go for every "null" and "peak? Not sure, is that 55.1 hz or maybe some other :?:

I'm sure you know 1/3 octave plots.

Thats cool, I see that will be nice to adjust my HF response to "few db more". BTW, HF close measuring (from about 10cm) of my Tannoy's looks pretty close to the hot spot measuring. How the Tannoy guys measured hf response flat to 20khz +/-3db :roll:

The last one is called the ETC (Energy Time Curve) It is another way of looking at the impulse response in the time domain (the other graphs were in the frequency domain). It shows that all the energy is at -60 dB afetr about 150 msec

It's about rev. time?

It shows that all the energy is at -60 dB after about 150 msec... a very short time.

Is that good or bad? :oops:

The bass in your room, with the exception of the 110 Hz slice, looks fantastic. From just looking at the graph, I assume that it sounds incredibly tight. Is that the case? I can't imagine that you need any bass trapping. I think that the slice is due to a single reflection somewhere, but it would be hard, at least for me, find the exact cause. Yes, the ETC is related to reverb time, adn yes, it is IMO very good that your reverb time is as short as it is.

The bass in your room, with the exception of the 110 Hz slice, looks fantastic

Glad to hear that :D

From just looking at the graph, I assume that it sounds incredibly tight. Is that the case?

Yes, sounds very tight, except I not too happy with Tannoy AMS8 low end, but that's out of topic.

I think that the slice is due to a single reflection somewhere, but it would be hard, at least for me, find the exact cause.

I tried to "hide" the back of listening position with massive absorber, and that slice at 110.2 hz greatly improved, but when I move that "mass" a 1m into direction of rear wall slice comes agin. To be honest, will be very happy to solve this 110.2hz problem :D Here's the pic what I'm talking about.

Yes, the ETC is related to reverb time, adn yes, it is IMO very good that your reverb time is as short as it is.

Great 8) Thanks a lot for helping :D

Bump :)

I don't see any question marks in your last post. :D I suppose you're wanting help with the 110 Hz slice. I'm not sure just from looking. Older more experienced chaps might be. I will think more on this.

I suppose you're wanting help with the 110 Hz slice

Yee, I'm happy with my new studio, but will be more happy if i can do some fine tuning. Start can be to kill that 110hz slice... Unfortunately I don't have any experience about :?

What's the exact distance from your front wall, center, to the center of your rear wall? Also, where is the chair/mic in relation to these two points? Steve

Steve, I presume that you are asking about distance fom one brick wall to another... Here's the pic. Distance is 5.20m and Mic was placed at hot spot, (chair) at 2.50m from front wall. Thanks :)

Guys, is my description and drawing skimpy or you just in the middle of some business :roll: