DrewBolce Posted January 13, 2011 Share Posted January 13, 2011 I am finally getting around to updating my ALK universal crossovers to Al's new(er) woofer filter so I thought I would once and for all get some good measurements on my woofers. The Speakerlab woofers are 8 ohm nominal so I expected to have to tweek Al's numbers. To my surprise they measured 1.15mH and 6 ohms! Al's design is based on 1.00mH and 6 ohms. This is a little suspect since the DCR of these is 6.8 ohms. I would have expected them to measure 10+ ohms in the khorn enclosure. My measuments were done using my 10mhz Tek scope, ST1000B for a signal generator and Recal-dana frecuency counter. My cap was a 27.1uf Solen. I used a 1/2 watt 220 ohm 1% resistor, my channel 2 trace was prettier than Al's noisy Carbon comp. I nulled at 902hz. Using the formula I came up with 1.15mh. Inserting a 10 ohm 1% resistor and calibrating to 5 divisions I was surprised to see 3 divisions when I reinserted the cap. The only caviot is that I was only able to generate about 1.5vrms, maybe that is too little for this measurement? Unless I did something wrong this is a surprising result and indicates that the woofer is a much closer clone to the K33E than I thought. Quote Link to comment Share on other sites More sharing options...
PrestonTom Posted January 13, 2011 Share Posted January 13, 2011 It is not clear what you are up to, but at the very least it looks like you are confusing resistance (DCR) and impedance (which is an AC measure composed of resistance and reactance and varies as a function of frequency). I can't speak for ALK, but I will assume that in his design process he will have the complex impedance measure (resistance and reactance) of the woofer (in the enclosure) as well as its frequency response. Again, I can't speak for ALK, but I assume he is most interested in the value of the impedance around the eventual crossover point Now, what is it that you are up to? Quote Link to comment Share on other sites More sharing options...
DrewBolce Posted January 13, 2011 Author Share Posted January 13, 2011 Oops, sorry I forgot to post a link to Al's measurement method. http://www.alkeng.com/le/le.html I am not confusing the two, it's just that I would expect the measurement (which is supposed to be the sum of the AC and DC components) to be larger than the DCR. Yes, I am looking to match my woofer with his crossover design. I just read his paper a little more thoroughly and realized I need to change the cap value until I get a frequency close to the 400hz crossover point. I'll try that and post the results. Thanks, your questions got me to wondering about the 902hz and its relationship to my measurements. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 14, 2011 Share Posted January 14, 2011 The impedance of the K33 driver does change a bit with different frequencies but it generally around 6 - 7 Ohms in series with about 1 mHy voice coil inductance. It's not extremely critical anyhow. The general rule of thumb is an error of 1.5:1 or less is good. That would mean a filter designed for 6 Ohms could be terminated by anything between 4 to 9 Ohms and perform ok. 158 uF is the correct value to resonate the K33 voice coil inductance with if you are looking to design for a 400 Hz. crossover. Al K. Quote Link to comment Share on other sites More sharing options...
DrewBolce Posted January 14, 2011 Author Share Posted January 14, 2011 Thanks AL. It just got weirder on the ohm measurement. My speaker required a 121uf cap at 397hz. This calculated out to 1.326mH, very consistent with the manufactures spec of 1.3mH. But the ohms measurement is getting really weird, now I am getting 5 ohms. I am using a 200 ohm series resistor, I assume it's value isn't that critical in this setup. I am using a 1% 10 ohm resistor for the calibration. Am I on the right track? Can I use Xover to calculate the new values? If so, are they 52.75uf cap and 1.33mH coil? Quote Link to comment Share on other sites More sharing options...
DrewBolce Posted January 14, 2011 Author Share Posted January 14, 2011 Oops, bad setup on the R measurement (I was leaving the speaker in circuit for the 10 ohm calibration). I am now getting 12 ohms. This is more like I expected. Now I know something will need to be changed since this a 2:1 error. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 14, 2011 Share Posted January 14, 2011 I may have just missed it, but is the driver supposed to be a K33 clone? Also, the 1st order filter in the Universal is just not effected much by value changes. This is why it's "universal"! I wouldn't fool with it unless you are going to go to at least a 2nd order filter (12 dB / octave). Al K. Quote Link to comment Share on other sites More sharing options...
DrewBolce Posted January 14, 2011 Author Share Posted January 14, 2011 Yes it is, but I have the 8 ohm version. They did offer a 4 ohm version which I suspect is a very close clone. I have read your "universal" thread and agree that it is quite adaptive. I am in the process of swapping my 39 + 1uf caps with a 25 + 25uf (as close to the 48uf as I have on hand) with 1.36mH coils (handwound 12 ga. and measured on a Sencore LC102) and was interested in the "optimum" values for my 8 ohm speaker. The main reason is I have other caps and coils at my disposal so matching optimum values is possible. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 14, 2011 Share Posted January 14, 2011 Why not go up to 2nd order? Al K. Quote Link to comment Share on other sites More sharing options...
DrewBolce Posted January 14, 2011 Author Share Posted January 14, 2011 Not a bad idea, however, I want to preserve the constant impedance of the Universal. I have fine tuned my measurements and have come up with 1.4mH, 10 ohms at 406hz. It took me awhile to get the setup and finesse your method. Now that I have it down I like it, it is solid and repeatable. I guess my next question is how to calculate the 2nd order values correctly. My calculations for the woofer filter is 28uFand 4.23mH. I assume the cap ( 48uf ) to the upper section would remain, but would the value be changed? Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 14, 2011 Share Posted January 14, 2011 Any correctly designed crossover, no matter the order, will have constant impedance! There is a design program downloadable from the same web site were the measurement procedure is described. As long as you use the same order on the highpass as on the lowpass, it will be constant impedance. The woofer Le counts as one of the elements to the impedance, but not to the filtering function. You should design for n=3 which will give you two parts in the lowpass and 3 in the highpass. This is the configuration I recommend. AL K. Quote Link to comment Share on other sites More sharing options...
WMcD Posted January 15, 2011 Share Posted January 15, 2011 For what it is worth. In this thread I posted the impedance curve of a Belle woofer in a zip file. http://community.klipsch.com/forums/t/94042.aspx?PageIndex=2 It may give you something to work with. Of course that is not really addressing your question. But better than nothing. Wm McD 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.