John Warren Posted December 22, 2007 Share Posted December 22, 2007 k Quote Link to comment Share on other sites More sharing options...
mas Posted December 22, 2007 Share Posted December 22, 2007 John, May I suggest that you literally lay the mic on the floor. It will act as a poor man's PZM and minimize the FX of the floor bounce reflection. Quote Link to comment Share on other sites More sharing options...
jwc Posted December 22, 2007 Share Posted December 22, 2007 John. You suggested I buy this setup in the past. Can you give me a link to this product. jc Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 . Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 jwc- The distributor is link below. I have the CLIO lite unit. The Lite version has most of the same functionality as standard. When I bought mine last year, it included the mic. They are not saying that now(?). If the mic is not included add the cost of the mic (~$400). I use a K&M mic stand and a generic mic adaptor. Need some cables for impedance and mics. Note for anechoic FR measurements, the LF limit is limited by the distance between the mic and nearest reflection surface (floor/ceiling). So if you want to measure the base response (<300Hz) you can't. http://estore.websitepros.com/1736754/Categories.bok?category=CLIO Revision- I checkled my invoice and you will have to buy the mic too, so the cost is about $1100 for CLIO lite. The dollar has taken a pounding against the euro. Quote Link to comment Share on other sites More sharing options...
mas Posted December 23, 2007 Share Posted December 23, 2007 Without getting into all of the math surrounding the swept frequency impulse, your gating does not completely solve this issue. What you are trying to do is what TDS (Time delay spectrometry) does inherently,and that is done via the use of agile time and frequency tracking filters to isolate the response relative to the impulse...But that is a much deeper subject than either of us want to jump into right now. Including the fact that all frequencies are not necessarily being generated simultaneously depending on the nature of the stimulus utilized. Whatever method you are using, lay the mic on the floor in front of the speaker. This minimizes the floor reflection to a nominal value relative to the direct signal. This technique was developed by Don Keele too many years ago (~20) for exactly such a test. It even works with the TEF using TDS. You are discovering one of the interesting and inherent facets of MLS. 4.5 x 1.13 - 5.08500 ft (ignoring proper precision!) you will still catch these reflections. Additionally, time window of 4.5 ms will result in a frequency resolution of 1/T= 222Hz. But you are coming close to discovering what advantages TDS has over MLS, not to mention the additional ambient noise immunity of TDS.[] ...and in anticipation of the subject, no, I have no interest in debating MLS vs TDS (that straw horse has been beat to death in other venues!). If you understand the limitations, both are great tools. Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 gg Quote Link to comment Share on other sites More sharing options...
mas Posted December 23, 2007 Share Posted December 23, 2007 John. I would refrain from trying to literally interpret the impulse response - especially for reflections! That is not its practical purpose. (And note that there is a fundamental glitch in the IR! The apparent peak is the second peak in the IR response....very curious...but this may also have to do with the windowing function... note the attached expansion) You will want to look at this in the ETC. This could be caused by a number of issues...But having the second peak have a greater intensity is rather fascinating as that rules out diffraction, internal resonances, etc.) Instead, use the Envelope Time Curve (ETC) to evaluate reflections - both in regards to intensity and time of arrival. This measurement, depending upon your windowing (time and duration of the pulse, thus affecting the resolution of the results and the frequency response), is a most powerful tool for signal alignment and reflection identification and control. MLS offers a powerful environment, but be aware that a primary weakness is noise immunity. Extraneous noise, including the ambient noise floor effects the measurements, sometimes destructively. One site you might want to visit that may prove useful is the EASERA forum and tutorial site linked from the Renkus Heinz site. The primary module of EASERA utilizes MLS as well. (In fact, EASERA is the product developed due to the failure of MLSSA to be actively undergo further development... I love seeing you play with this! Products such as this and ETF offer a significant opportunity for folks to look at system performance with a much greater accuity in many insightful additional and more powerful ways! Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 Mas- Below are the ETC and the MLS impulse response. On the ETC, the marker is set at 5.5695ms which corresponds to a distance of 1.9159m (6.37ft = 6ft 4.5inches). The corresponding marker is also shown in the MLS impulse response (below). Note that it does correlate with a *indication* of some type but,as you state, the ETC is the correct way to examine reflectiions. Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 Here's the MLS Impulse Quote Link to comment Share on other sites More sharing options...
DrWho Posted December 23, 2007 Share Posted December 23, 2007 Interesting plots John! Would you be apposed to providing some off-axis measurements too? Like maybe 15, 30, and 45 degrees? I'm mostly just interested in the horizontal polars, but the verticals might be interesting too. There seems to be fairly good correlation to the JBL spec sheet too:http://www.jblpro.com/pages/pub/components/2404.pdf Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 23, 2007 Author Share Posted December 23, 2007 Mike - I'll measure both the K77M and 2404H off-axis at a few angles and post results. Note also that the 2404H tweeter response is connected to my filter so it is not the *raw* driver response that the PDF shows in the JBL manual. jw Quote Link to comment Share on other sites More sharing options...
mas Posted December 23, 2007 Share Posted December 23, 2007 I have a question about the IR, and I wonder if the speaker may be connected out of polarity. The Nyquist would expose this very quickly. Can CLIO display the Nyquist plot? (I don't think it can, but I may be incorrect...) The large initial negative response may be non-casual (negative time)...it is hard to tell from just the graph. Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 24, 2007 Author Share Posted December 24, 2007 Mas-Yes, the speaker is out of polarity wrt to the amp. The polarity is inverted to get the correct phase response (the mic inverts polarity). I'll redo the IR with the polarity reversed to see if a difference is observed. jw Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 24, 2007 Author Share Posted December 24, 2007 mas- CLIO cannot do a Nyquist Plot. The first spike positive is associated with polarity. Both in and out of phase wiring IRs are shown below. First, inverted polarity wrt mic polarity Quote Link to comment Share on other sites More sharing options...
John Warren Posted December 24, 2007 Author Share Posted December 24, 2007 Now with speaker leads switched at the network Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.