pzannucci Posted January 22, 2004 Share Posted January 22, 2004 Wanted opinions if a parallel notch filter put on the tweeter on '85 vintage khorns would mess up the crossover function on the tweeter. The filter would be put before the crossover connection for the tweeter and would be used to calm a 3/4db rise (annoys me) between 6k and 9k. The crossover is an AK-2. If anyone has a better suggestion, please let me know. Thanks, Quote Link to comment Share on other sites More sharing options...
ZAKO Posted January 22, 2004 Share Posted January 22, 2004 Is that a 3/4db or 3 to 4 db rise? Quote Link to comment Share on other sites More sharing options...
WMcD Posted January 22, 2004 Share Posted January 22, 2004 The conventional wisdom is that at least some midrange drivers need the P-trap to remove a signal at about 9 kHz. Gil Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 22, 2004 Author Share Posted January 22, 2004 Sorry about the 3/4 it should have been 3-4 db rise. Thanks, Quote Link to comment Share on other sites More sharing options...
ZAKO Posted January 22, 2004 Share Posted January 22, 2004 Peter Z.....How did you measure this rise On your tweeter? BEtween 6K & 9K ? Its usually the opposet. Is this on both tweeters? Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 22, 2004 Author Share Posted January 22, 2004 I measured the rise using an spl meter. In my room I have a rise around that range which I would like to smooth out (I am used to the smooth sound of a pair of MDT-33s in this room). I was trying to find out if anyone has done any measurements and would know if the notch filter would cause any problems with the tweeter crossover not working as designed if I put the filter before the crossover network for the tweeter (saw some discussions by Al K. about the slope and didn't want to affect that). I also would not like to burn out my tweeter too easily. Thanks, Quote Link to comment Share on other sites More sharing options...
itsmyforte Posted January 22, 2004 Share Posted January 22, 2004 maybe you should correct the room and not the speakers. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 22, 2004 Author Share Posted January 22, 2004 Unfortunately it is not the room. It would cost a lot more to fix the problem in the room (not counting WAF) than smooth out the reponse of the speakers. There were some previous threads about this rise in response and I was wondering about the technical merit and issues with the eliptical filter on the tweeters with the change in impedance introduced by the filter. Thanks, Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 22, 2004 Share Posted January 22, 2004 Peter, A notch filter could be inserted but would require careful design. It would be and R-L-C network. The ratio of L to C determines the width of the notch. The "procuct" of L and C determines the frequency of the notch and the resistor determines the notch depth. As you can see. it's not an easy thing to nail down. It WILL upset the impedance match a bit but not to a degree that will casue any problems. I suggest that you simply purchase a graphic equalizer instead though. This is by far the best way to solve a problem like that! Al K. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 22, 2004 Author Share Posted January 22, 2004 Al, Thanks... I had seen your posts about the tweeter filter and was worried that I might change the function significantly with the notch filter. I don't have an easy way to put an eq in the system so since I like most everything about the khorns in the room but wanted to tame that one peak, that I would persue the filter. I do believe that this problem was addressed in this in these forums before (I have been on it off and on for several years) but have not seen references to this issue in particular except for how to lower the output of the tweeter altogether which is not exactly what I want to do. Thanks, Quote Link to comment Share on other sites More sharing options...
3dzapper Posted January 22, 2004 Share Posted January 22, 2004 I'm listening to a pair of ALK networks right now in my Khorns. After trying many different taps on the autoformer I have found one that I like and is flat +/- 3db from 30 to 10Hz without the 3500Hz or 9Khz spikes in the stock AAs. although I had already removed them with my modded AAs. I'll be using the ALKs at least until I can build new AAs with my mods and better inductors. Since your networks are AK-2 I would recommend the type A, constant impedance ALKs for you. For adjustability they can't be beat the adustability. Also check out Al's new extreme slope networks optimized for the Khorn. Rick Quote Link to comment Share on other sites More sharing options...
moefasta Posted January 22, 2004 Share Posted January 22, 2004 I think that maybe the offender is not the tweeter, but instead the squaker. Al uses a low pass filter on the squaker in his ALK's. If you roll off the squaker around 6000hz, you might eliminate that bump from 6000-9000hz. Just a thought Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 23, 2004 Author Share Posted January 23, 2004 moefasta, I had been thinking of that change also but was going to try the tweeter change first. I was less likely to think that a 12db slope would correct it after reading about the actual slope and function of the tweeter network discussed elsewhere on this forum. I think the rise is so quick and steep on the tweeter crossover that it kicks in hard. I would like to smooth the frontend of the slope somewhat to make smoother transition to the tweeter. Don't know if that will cause other problems due to the offset of the mid and tweeter drivers. I'll experiment. I even have a pair of the Aurum Cantus G1 Ribbon tweeters that I might try and cross in at around 3 or 4 Khz for the fun of it. Thanks ALL for your thoughts. Quote Link to comment Share on other sites More sharing options...
ZAKO Posted January 23, 2004 Share Posted January 23, 2004 Peter Z ...Is the ribbon tweeter efficient enough? I cant find that kind of rise (3 to 4db) on my tweeters (between 6K & 9K) K77 or T35. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 23, 2004 Author Share Posted January 23, 2004 I don't believe it is the tweeter alone that's why I was referring to the quick rise of the tweeter around crossover. The combination of the slow rolloff of the mid horn and the quick rise (steep slope) of the tweeter might be causing the effect I am noticing. I think if I drop off the quick rise a little, the smoother transition between the mid and the tweeter will lower the rise and make for a better overall presentation. I hope the notch filter will lower the output at the frontend of the tweeter rise and lower that peak. Quote Link to comment Share on other sites More sharing options...
ZAKO Posted January 23, 2004 Share Posted January 23, 2004 Im using a electronic crossover & can change the slopes from 6, 12, 18, 24db on both sides of mids & tweets. I found keeping slopes the same at crossover always sounds the best & in phase. Hell with the tweeter 24" in front of mid driver things are bad enough. How close was the sound level meter to the speakers to minimize room reflections? Like AL mentioned a Rane EQ might help to narrow your problem. Quote Link to comment Share on other sites More sharing options...
Q-Man Posted January 23, 2004 Share Posted January 23, 2004 I know that the AK-2, AK-3, AL-3, AL, and the AA networks all peak at the 6000Hz crossover point. I don't know about the others. I also believe the ALK also has this same peak, I will have to check my notes. A notch filter may do the trick. John Warren put some in my networks for use with the 290/311 horn. One notch filter was needed for an 800Hz peak in the HORN it self. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted January 23, 2004 Share Posted January 23, 2004 Guys, I think all of you are barking up the wrong tree! This business of a peak at the 6000 Hz crossover be part of all Klipsch networks is new to me. The AA network has a slight peak in the tweeter filter at around 5500 Hz because it's a bad filter, but the so-called "elliptic filter" of the AK series networks are quite flat. I have posted the curves on both on the Ods and Mods forum under "extended highs for the AA" thread. If the phasing is not right you could get a theoretical boost of 3 dB at the crossover, but the wavelength at 6 KHz is so short the peak would only be a one spot in the room and would even become a notch if you moved to another place in the room! I would check very carefully where that peak is in frequency and location in the room. Make sure it is from BOTH speakers and not just one. I would also make sure it isn't just one of your tweeters out of wack! I really don't believe you have network trouble. If it turns out to be real, make every effort to use an electronic equalizer before you do anything else. Every stereo receiver has a tape monitor in and out. The equalizer can be put into this loop. Adding notch filters is very difficult thing to do properly! BTW: there is a constant impedance equalizer circuit, but it is complicated consisting of two inductors, two caps and two resistors. I have used it at radio frequency and it's a pain in the butt! Al K. Quote Link to comment Share on other sites More sharing options...
pzannucci Posted January 25, 2004 Author Share Posted January 25, 2004 I think I have figured what might be happening. I am not sitting anywhere close to the direct center line on the khorns (about 6 feet back from center axis w/horns about 16 feet apart) so it appears that the tweeter has better dispersion off axis than the midrange thus giving rise to the tweeter frequencies. Guess I'll have to weigh what I'll have to do about my positioning since I want to keep the horns locked directly in the corners. Thanks, Quote Link to comment Share on other sites More sharing options...
D-MAN Posted January 26, 2004 Share Posted January 26, 2004 There aren't any easy answers for this... I am suggesting the following as a possible solution on the basis of mid-and-upper horn alignment first followed by room treatments (they are inter-related) and have had the most success with it (also, it is the cheaper of all of the routes to go down, I think). At least, this is what I did... One of my gripes with the classic Khorn is the top cabinet having the horns being hard-mounted to a baffle board. When the top cabinet is aligned to the bottom, which is aesthetically pleasing of course, the horns probably are not pointing directly at the listening position (at least in my place as it ain't that big of a room). This caused me endless grief as all of the upper-end horns were off-axis and sounded pretty "peaky" and/or shrill at some times. So I fixed this by replacing the top cabinet so that the "new" cabinet remains aligned with the bottom but I can adjust the horns' orientation independently within it. I feel that they sound better on-axis, others may prefer off-axis. Could the upper horn axis alignment be the issue rather than a network problem? In the right space, this issue may not come up, but so far, I haven't had the right space... I would not recommend modifying the original Khorn top cabinet - I would say make a replacement cabinet to play with. That way no harm is done. DM 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.