D-MAN Posted February 4, 2005 Share Posted February 4, 2005 Have you read the article? That is what it says. I am taking it at its word. If that makes me wrong, then fine. You are correcting nothing by trying to make me the bad guy here. DM Link to comment Share on other sites More sharing options...
John Warren Posted February 4, 2005 Share Posted February 4, 2005 My point is that the Klipschorn bass unit, when loaded with the K33E, is approximately an 8 Ohm loudspeaker between 50-400Hz, give or take. BTW, the minimum impedance (the trough after the peak) in the Klipsch plot is 6.5 Ohms, consistent with the minimum I've measured. Link to comment Share on other sites More sharing options...
D-MAN Posted February 5, 2005 Share Posted February 5, 2005 Your test does indicate quite a difference between what I expected based on what I read and what the electrical impedance actually is. I've done some research and I come up with this: PWK was saying that the driver becomes 16 Ohmns when loaded in the Khorn starting in 1946 and as late as 1986. The earlier reference (probably got and accepted from PWK) is 1966 in "How to Build SPeaker Enclosures", Howard W. Sams and Co., Indianapolis, IN, 1966 (13th reprint 1978, page 91, which I quote: "The 15 inch woofer is specially designed: it has a 3.2 Ohm (dc-resistance) voice coil. However, installed in the horn, its impedance becomes 16 Ohms because of the excellent loading and high efficiency of the horn." This is expectedly what PWK said in the later article, more or less. The laws that he cites in his earlier patents is Berenek's Impedance curves for calculating the resistance and reactance resulting from horn-loading a driver in a particular horn in 2 separate curves plotted by Ohms and frequency. Dinsdale also quotes the Berenek formulas for calculating the same thing for calculating throat impedance. It's clear that this is something other than electrical impedance "seen" by the amplifier, but then, what is PWK talking about?... I still am confused by PWK's and others statements as to what driver he wanted for the Khorn and why. This doesn't make any sense! DM Link to comment Share on other sites More sharing options...
D-MAN Posted February 7, 2005 Share Posted February 7, 2005 Here is the PWK comment from "A Visit to the Klipsch Kingdom", from article in SpeakerBuilder magazine, April, 1989 by Bruce Edgar, quote: SB: Was that the 15WK? PWK: Yes, but the original was 16 Ohms. I wanted an 8 Ohm driver which in my horn came out to 16 Ohms... I am at a loss to explain any of this, except that perhaps the impedance test itself is "different" with modern electronics? DM Link to comment Share on other sites More sharing options...
John Warren Posted February 7, 2005 Share Posted February 7, 2005 ---------------- On 2/7/2005 8:17:23 PM D-MAN wrote: Here is the PWK comment from "A Visit to the Klipsch Kingdom", from article in SpeakerBuilder magazine, April, 1989 by Bruce Edgar, quote: SB: Was that the 15WK? PWK: Yes, but the original was 16 Ohms. I wanted an 8 Ohm driver which in my horn came out to 16 Ohms... >>Soooooo, if I put a 4 Ohm driver in there it will be.......................... I am at a loss to explain any of this, except that perhaps the impedance test itself is "different" with modern electronics? >>That must be it. DM ---------------- Link to comment Share on other sites More sharing options...
JohnA Posted February 8, 2005 Share Posted February 8, 2005 D-man, You have misinterpreted the article (or incorrectly remembered it). The reason the K-33-E is a 4 ohm nominal woofer is so that the woofer impedance IS NOT 16 ohms when in the bass horn. Though called 8 ohms for the public, it should realy be considered a 6 ohm bass horn as the corner bass horn obviously doesn't double the impedance of the driver inside it. If a 10 ohm woofer, much like a Heresy K-22, could be found with the correct T-S parameters, you could have a 16 ohm K-horn. Link to comment Share on other sites More sharing options...
Al Klappenberger Posted February 8, 2005 Share Posted February 8, 2005 Guys, What all you guys are missing is that impedance is a complex number. To say the impedance is 4 Ohms or 8 Ohms is completely missing the boat! Here's an actual impedance plot of one of my Belles. The Impedance is actually 6 Ohms in series with 1 mHy inductance. The marker on the plot is located at 700 Hz. The data reads 7.41 Ohms to a phase angle of 34.3 Degrees. The top plot is the phase angle of the polar impedance. If you convert from "polar" ro "rectangular" impedance you have: 6.12 +j4.18 Ohms. If you solve the inductive reactance formula for L (Xl = 2 * Pi * F * L) you will see that it represents .95 mHy. So: The nominal 8 Ohm K33 woofer is actually 6 Ohms in series with 1 mHy. To totally ignore the fact that a woofer voice coil is an inductors is a gross oversimplification. Al K. Link to comment Share on other sites More sharing options...
JBryan Posted February 8, 2005 Author Share Posted February 8, 2005 Fellas, Thanks for all the information. Although most of this stuff is over my head, I now have a vague idea of how impedance works in regard to Khorns. The reason behind this topic is that I have a pr of Al's ALK xovers and I read that the..."network is a true multiplexer providing constant load impedance to the amplifier." The description went on to state that the load seen by the amp(s) will be a constant 8 ohms (+/- 1 ohm). If the impedance can be manipulated or controlled at the xover to provide a constant load, I was wondering if its possible to alter the load to 12 ohms (or thereabouts) at the xover or will it require a major overhaul like changing out the drivers. I'm only interested in temporarily changing the impedance to accomodate the SE-OTL amps - nothing permanent. Thanks again and have fun -Bryan Link to comment Share on other sites More sharing options...
John Warren Posted February 8, 2005 Share Posted February 8, 2005 ---------------- On 2/8/2005 7:15:09 AM Al Klappenberger wrote: Guys, What all you guys are missing is that impedance is a complex number. To say the impedance is 4 Ohms or 8 Ohms is completely missing the boat! Here's an actual impedance plot of one of my Belles. The Impedance is actually 6 Ohms in series with 1 mHy inductance. The marker on the plot is located at 700 Hz. The data reads 7.41 Ohms to a phase angle of 34.3 Degrees. The top plot is the phase angle of the polar impedance. If you convert from "polar" ro "rectangular" impedance you have: 6.12 +j4.18 Ohms. If you solve the inductive reactance formula for L (Xl = 2 * Pi * F * L) you will see that it represents .95 mHy. So: The nominal 8 Ohm K33 woofer is actually 6 Ohms in series with 1 mHy. To totally ignore the fact that a woofer voice coil is an inductors is a gross oversimplification. Al K. ---------------- I am aware of the fact that what is presented is not the total story. The data is from the calibrated voltage divider method which you are aware of. Link to comment Share on other sites More sharing options...
John Warren Posted February 8, 2005 Share Posted February 8, 2005 double post Link to comment Share on other sites More sharing options...
D-MAN Posted February 8, 2005 Share Posted February 8, 2005 ---------------- On 2/8/2005 12:12:39 AM John Albright wrote: D-man, You have misinterpreted the article (or incorrectly remembered it). The reason the K-33-E is a 4 ohm nominal woofer is so that the woofer impedance IS NOT 16 ohms when in the bass horn. Though called 8 ohms for the public, it should realy be considered a 6 ohm bass horn as the corner bass horn obviously doesn't double the impedance of the driver inside it. If a 10 ohm woofer, much like a Heresy K-22, could be found with the correct T-S parameters, you could have a 16 ohm K-horn. ---------------- I quoted the article and other reference EXACTLY ver batim. That's the crux of my problem! However, (as usual) I agree 100% with you. DM Link to comment Share on other sites More sharing options...
Al Klappenberger Posted February 8, 2005 Share Posted February 8, 2005 Bryan, It is possable to change the impedance of any load with a transformer, but to do it at woofer frequency requires a HUGE one. The output transformers of a McIntosh power amp is the size it takes. The "Zero" transforme should be able to do it. With a crossover network, the woofer therefore dictates the impedance of the entire network. In my "Universal Type A", the 40 uF cap compensates for the 2.4 mHy inductor and woofer voice coil dinductance. John's voltage divider method is ok so long as the reactive component is not a problem. You can get back to that simple measurement done by the divider from the rectangular Zo by the series impedance equation Zo = sqrt(Xl * Xl + R * R). The Xl (inductive reactance) is the +j component. When you are designing a filter it is definitley a factor if you want a resistive load to the amp. Al K. Link to comment Share on other sites More sharing options...
John Warren Posted February 9, 2005 Share Posted February 9, 2005 Al I picked up a Krohn-Hite phase meter to address extracting the Xl component from the impedance magnitude. Link to comment Share on other sites More sharing options...
Al Klappenberger Posted February 9, 2005 Share Posted February 9, 2005 John, I don't know about the K-H phase meter. I have no experience with it. I tried a Hewlett-Packard 4800 vector impedance meter for that purpose and found it totally unusable. The microphone characteristics of a loudspeaker overwhelmed it's readings with any ambient sound at all. It operated at too low a level. If you are using a signal generator with it and can actually hear the test signal over the room ambient noise it sould be ok. If the reading is stable when you thump on something nearby it should be right. I found that I needed the full output from the 70V line of a PA amp into my bridge to plot impedance with my amalyzer. With the 4800 meter even a car going up the street made the reading jump around! Al K. Afterthought: Does the KH meter have an analog output proportional to phase? If it does, an old Wavetek function generator and an X-Y plotter cound be used to make polar plots like I get out of my SD375 analyzer. Link to comment Share on other sites More sharing options...
John Warren Posted February 9, 2005 Share Posted February 9, 2005 Al-Stable under all operating conditions. It does include an analog output, equal to 10mV per degree phase, for use with an external meter or strip chart recorder, plus an optional BCD output. Link to comment Share on other sites More sharing options...
Al Klappenberger Posted February 9, 2005 Share Posted February 9, 2005 John, Ok.. You've got everything you need to make polar plots on an analog plotter. They are very cheap on eBay. I got one sitting in the attic. I was able to graduate to a digital plotter (HP 7475A) as the SD375 also has RS-232 and HP-IB outputs. I think you have a function generator. If not, there are plenty of those on eBay too. Al K. Link to comment Share on other sites More sharing options...
Recommended Posts