PC Spectrum Analyser?

Great question. I asked this some time ago and was refered to the "audio toolbox", which was about $1000. More than I wanted to spend. But to answer your question there are inexpensive spectrum analyzers. HP works FFT spectrum analyzer at: http://www.hpw-works.com/ it's between $69 and $300 depending on options. Audua Speaker Workshop: www.audua.com. You can download an alpha product for free. Sample Champion is another that you can download a trial version for free at www.purebits.com. I wanted to do the same thing you are trying to do, but I can't find a linear microphone. Some of these programs have methods to calibrate the microphone for non-linearity, but if you don't have a flat source to do the calibration it's basically impossible. Hopefully, someone else will post some ideas on microphones or methods to calibrate them. If you get a set-up like this working, please post it--I'm sure many audiogoners would be interested.

Being unsure of the exact use you intend...you might want to check out the ETF 5 software package, which alows you to do MLS measurements on speakers, and on listening room acoustics, etc. You can find more info at their web-site http://www.etfacoustic.com/. It may do some of things you're hoping to accomplish. Good luck!

There is a program called "Spectra-Lab" or something of that sorts. I have used it briefly and it is quite versatile. Can't speak for how much it costs though as it was someone else's computer. As to finding a microphone that is quite excellent for the money, try contacting Old Colony Sound Labs. They have what is called "Mitey Mic" and they are on the second version of this. It is very linear and was designed by Joe D'Appolito if memory serves me right. Once you've gotten a good mic, you'll have to invest in a good sound card. If you've ever looked at the frequency response chart or specs for a typical "sound blaster" type of card, you'll realize that it is FAR, FAR below what you would want to do the job. Sean
>

The street price for the Terrasonde Audio Toolbox is $750; it's not cheap, but still a very good value. It has a high quality built in microphone, test signal generator, and the RTA is selectable for 1, 1/3, 1/6 and 1/12 octave accuracy. Plus the Toolbox performs a number of other functions. Visit their website and see if it makes sense for you.

Sean, that's good to know about the mic. You are also correct about the sound card. However, the programs I pointed out (at least 2 of them) have sound card calibration, where you plug the input directly to the output and it tests and stores a correction file for the sound card. This is very useful, because I want to use this with my portable computer, where I really can't put in a full size soundcard. But you are right--there must be some correction if you aren't using a high quality sound card.

Where or how can one contact Old Colony Sound Labs, for this microphone?

Another option to consider is renting a professional rig. Though I have not done it yet, I have found a sources. AudioClassics.com. I f you look under "Test Equipment", you'll see they will rent an Audio Control SA3050-A with all the trimmmings for $ 90/week. Since this is not something I would use every day (in fact, only when making a MAJOR change to louspeakers type/position, room, room treatments and/or listener position), this may be worthwhile. On another website, someone also suggested hiring an acoustic engineer for a day to come in and do the measurements, etc. I'd love to this, but there don't seem to be any AE's for hire in Cincinnati (the one I found in the yellow pages was very polite, but said he only did large venuses - as in stadiums and churches!)

Well, I hate to admit it but I gave up on the idea of the PC Spectrum Analyser using my soundcard for now. Running a 1kHz test signal from CD through an osciloscope gave a nearly perfect sine wave. Running it through the spectrum analysis software on the PC showed TONS of harmonics at as little as 10dB below signal level... If those were really there, I should have seen them on the o'scope as well. Besides, signals above about 10kHz just did not get digitized very well at all. So my conclusion was: Forget about the sound card.

So I looked at a few solutions that have external attachments, like digital osciloscopes. Some looked like they might do the job, but still not really what I needed. Still, there was the issue of the linear mic and calibration.

So the next best thing I coudl find was the Audio Toolbox mentioned above. It is independant, but it is connectable to a PC for reading out results. We will see how well it works...

Niels.