That was a great explanation of why we would use a series notch or a parallel notch and how to adjust values to broaden or narrow the notch. Very informative and easy to understand. Well done.
Thanks, Danny for telling me that my 1987 Polk Audio SDA-1C speakers had very much lost their original great sound. I did not know that the crossovers were inadequate to begin with and had worn out through the years. Well, I undertook the crossover rebuild using Sonicap and Clarity capacitors with Mills resistors. I just copied the values of the old parts. Wow what a difference in the sound, clarity, and tight bass. Now, I am ready to do the same to my Klipsch RP-160M bookshelf speakers. Thanks for sharing your knowledge and expertise .
Thank you for sharing your knowledge with the world and us! Things Danny is teaching me - 1. Every speaker's response needs to be tested to "see" what the speaker is actually doing. 2. Sometimes adjustments could be made to make the response graph flatter and more true, but is it worth it. 3. The room the speakers exist in IS an important part of the "speaker system." 4. ALL brand names mean nothing to me now. I have heard $20,000 dollar speakers and was not impressed. Listen to what you like and it doesn't matter what name is on it!
It’s remarkable that large speaker manufacturers, with their huge budgets and “expert” engineers can’t do the same. Take the Klipsch rp600m. Even if they didn’t have the budget to design a proper crossover at first, they sold enough of them to offer a MkII version with a better crossover. It’s shameful, really.
Thanks for making this video. Cleared up a lot of what I was seeing experimenting with VituixCAD. I think a lot of confusion comes with these types of filters from many names others seem to use for them. Anyway very clear video you made of it.
I'd be curious what you think of paper cone tweeters and the things you might have to do to optimize them. Sony SS-MF550H has some kind of science paper tweeter, with the mid and low being Kevlar. They are remarkably pleasurable to listen to, no doubt you would measure some wacky stuff happening with the Kevlar Woofer playing full range. You can hear a lot of treble coming out of the woofer but it doesn't sound bad or grading somehow. The cabinets resonate very, very, very noticeably. My Sansui s820u's are my favorite. Paper cone tweeter, mid, and woofer. They image like the surround sound is turned on, I often have to check because I can't believe the receiver is in stereo mode.
Great video and explanation! What are you going to make a video on next? Do you have a video on the power system on your reference system? I want to better understand what you have and how you use it.
Thank you "Danny" for the in lightful lesson. Now I have a Polk audio monitor 70 series 2 loudspeakers . I'm looking to do an upgrade . I've ordered the parts . I been following you for sometime . Question what if on the Tweeter you grabbed a 14 awg inductor for it , is that a bad move? And or should I get a 18 awg inductor for the Tweeter in line with the caps an resister ? Thanks keep up great work .
Price points. Those bookshelf speaker are about $330 a pair. Would be over $500 with the mods. Also alot of times the upgrade crossovers can barely fit into the cabinet. There is forum that is going to take some of GR's upgrade speakers and do blind listening tests and measurements to see if there is a difference.
Very informative, thanks. Do you have a video on baffle step compensation? I'd be interested in how the baffle dimensions and driver placement, effect frequency response, as far as crossover design is concerned. I assume baffle step compensation might entail some sort of notch filter in some cases?
A Great Fix again. It’s nice to See and understand the changes theese Notch Filters do. But How about the big Dip ? What Kind of Parts did you use to flaten this Area ?
New to the channel. Love the vids. Im currently pulling apart.my homebrew loudspeaker. Will be completely upgrading the crossovers and using audyn copper max caps as bypass caps o. The mids and tweeters. My question is. What are your thoughts on L pad attenuators. I had I e on the mid and one on the tweeter. Over the last year in listening I kept finding myself setting both to 0 essentially. So im doing new backs o. The cabs and have to do some cab tuning i. General I was just thinking get rid of the L pads. I figure its less stuff I. The signal path anyway. But do you find that you like or should use an L pad or is using an L pad basically an admittance of poor crossover design? Thoughts?
hello there thanks allot for your efort and time to put these videos together,pure gold i have a Question about notch filter i`ve build an sb acoustic kit the eka kit listed on their webpage now in the crossover design,on the woofer side there is a notch filter to kill cone break up an an odd resistor in paralel with a second coil thath kill a big hump in the 2k region right before the crossover point i`ve bild the crossover like they listed in the schematic i saw in the 1k region a litle wide hump and no bothered but when i listen to music some female and man voice as wel as some electronic piano pierce my ear playing with an eq i find that the cause is that 1k hump do not want to use eq the question is : can a put there a notch filter to lower that 1 k hump ? i`ve played with some parts that i have around and did a sweep and there is a great atenuation on that region but i don`t know if there is a beter solution to that besides using a second notch filter i don`t have a dats to see the impedance an phase thanks!
what would be the appropriate gauge to make inductors. bass 22 , midle 22 , tweeter 24 ? .Of course, then there is the number of turns that we give it.
Hi. Wouldn't changing out the mid/sub driver to one that was a better match be a viable alternative? More expensive than what you did? Thanks for the video. I learned something today. : )
Is a notch filter necessary for 100hz notch or is it too low to affect the sound negatively. NB the notch is in the software curve and not the actual speaker measurement. Haven’t measured the speaker due to lack of calibrated microphone. Cabinet construction imperfections might have removed the notch and added other imperfections but I am happy with it’s sound. I just wanted to know so that in the future when the tinker bug bites then I can measure the speaker and create a notch filter for it.
9:19 ... It looks like that this was a designed feature to me. This seems to be common in many speakers. they are purposely attenuating the midrange up to 5 or 6 dBm which makes the speaker produce more bass and treble than midrange. this appeals to many people as a "more open" speaker and they believe it has greater clarity than a speaker designed with a flat response. However on some recordings especially a small vocal group with a female singer singing lead, her voice just seems to drop out of the mix. A good example would be Pozo Seco Singers LP "Time".
Thanks Danny, I was wondering the same thing. I'm curious about why you chose to have four woofers in your cc design. Sorry but I have a last request. Here in Canada there are very few places to audition electronics and I'd really appreciate if you could suggest a best bang for the buck max $3000. I will be buying some of you speaker kits but it adds up fast at 250 per flat pack when you want 6 to 11 units.
Great video as always! What is the main contributor to these large impedance spikes in individual drivers, i.e. is it the magnet structure, suspension or surround, etc? Thanks!
So one of the challenges of designing a 2 way is to have good speakers cover the larger bandwidth. I am using an Emenence kappalite 3012LF and need to get it to 2K for a crossover of 1600. using a notch filter and a lower crossover on the woofer I think this can be done.BUT does this whack out the phase at the critical crossover and am i kidding myself that the woofer will sound OK through this notch? TYVM
Hi Danny, can you explain the difference between the notch filter and a Zobel network. My understanding is that the Zobel corrects for impedance anomalies in the driver....so would you use both, combine the two or what?
In my experience notch filters and impedance correction filters are more of a "textbook formula" approach to crossover design. Rarely are they actually necessary except perhaps as a bandaid to "fix" a crossover that was not properly simulated or measured. If you use a textbook formula to make the crossover, then you'll probably need a textbook formula impedance correction circuit to make it even remotely useful. Better off to simulate the crossover components effect with response, phase, impedance, and diffraction data all incorporated. In that process, you can adjust component values and incorporate resistors in various parts of a 2-4th order network that will "tilt" the response and suppress peaks as part of the network design. This has to be done with care so as not to wind up with a nasty impedance dip, or strong circuit resonance, but there are indicators in the simulation to look out for to keep things reasonable (electric phase angle, impedance dips, sharp hooks or corners in response). In that process you might wind up with something that looks like a zobel network making up a "leg" of a 2-4th order network, as the final design may benefit from a resistor in series with a capacitor, but the value of the resistor or the capacitor is unlikely to be close to what a textbook formula would spit out for a correction circuit for a given driver impedance. For example, instead of building a 2nd order low-pass + zobel network for the midbass, (a 3+1 component network), going up to a 3rd order network with an additional series inductor, skipping the zobel network, and "tuning" those component values with a *possible* resistor incorporated *somewhere* in the circuit to fine tune the tilt/shape of the response (Still a 3+1 component cost), is likely to produce a steeper, more controlled rolloff and flatter response with proper tuning.
hi Danny, sorry this question is not about the filter, but the woofers response had that attenuation below 600 hertz. now my question is how come that attenuation is not present in the speaker measurement? thank you
Do you use formulas to determine the value of the crossover components; or variable resistors, etc, whose values can be changed while you're looking at freq response; or just a trial and error methodology?
@Lloyd Stout please spare me. How does the change in resistance equate to the peak, or bandwidth, or slope of the notch? Do you have a formula for that? How does changing the capacitance or inductance change the same properties?
Is there any negative unintended side effects of a LCR notch filter such as phase shifting distortion, damping, or ? around the intended bandwidth? You know, ELI the ICE man, if you know that saying. Thanks for sharing your knowledge.
Another great video Danny, thank you for explaining the difference in notch filters.
A handy little lesson for any speaker builder.
Always educational and eye opening. Thanks for the great video Danny.
That was a great explanation of why we would use a series notch or a parallel notch and how to adjust values to broaden or narrow the notch. Very informative and easy to understand. Well done.
Amazing how the spectral decay cleaned up.
Indeed, that was surprising how much it cleaned up.
Soooo much knowledge! Thank you so much for posting this.
My favorite audio "Elitist"! lol. Thanks for another great vid Danny!
Evidently Jr. didnt bother to look up the definition of Elitist.
Very informative as usual. Thanks Danny.
I love your explanation about how a higher x-over point increases the out of phase condition in the vertical axis due to the shorter wavelengths.
Thanks, Danny for telling me that my 1987 Polk Audio SDA-1C speakers had very much lost their original great sound. I did not know that the crossovers were inadequate to begin with and had worn out through the years. Well, I undertook the crossover rebuild using Sonicap and Clarity capacitors with Mills resistors. I just copied the values of the old parts. Wow what a difference in the sound, clarity, and tight bass. Now, I am ready to do the same to my Klipsch RP-160M bookshelf speakers. Thanks for sharing your knowledge and expertise .
Thank you for sharing your knowledge with the world and us! Things Danny is teaching me - 1. Every speaker's response needs to be tested to "see" what the speaker is actually doing. 2. Sometimes adjustments could be made to make the response graph flatter and more true, but is it worth it. 3. The room the speakers exist in IS an important part of the "speaker system." 4. ALL brand names mean nothing to me now. I have heard $20,000 dollar speakers and was not impressed. Listen to what you like and it doesn't matter what name is on it!
Looking forward to your future builds. Really looking forward to the new 8" driver.
me too. 3-way old school 8" would be really interesting
It’s remarkable that large speaker manufacturers, with their huge budgets and “expert” engineers can’t do the same. Take the Klipsch rp600m. Even if they didn’t have the budget to design a proper crossover at first, they sold enough of them to offer a MkII version with a better crossover. It’s shameful, really.
Very nice. I love this stuff! Cheers 🍻
Cool video very informative, good to see you back in your lane.
Thanks for making this video. Cleared up a lot of what I was seeing experimenting with VituixCAD. I think a lot of confusion comes with these types of filters from many names others seem to use for them. Anyway very clear video you made of it.
I really was wondering what a notch filter was! Awesome job, Sensei! My brain still struggles to understand circuits. Not your fault.
I'd be curious what you think of paper cone tweeters and the things you might have to do to optimize them. Sony SS-MF550H has some kind of science paper tweeter, with the mid and low being Kevlar. They are remarkably pleasurable to listen to, no doubt you would measure some wacky stuff happening with the Kevlar Woofer playing full range. You can hear a lot of treble coming out of the woofer but it doesn't sound bad or grading somehow. The cabinets resonate very, very, very noticeably. My Sansui s820u's are my favorite. Paper cone tweeter, mid, and woofer. They image like the surround sound is turned on, I often have to check because I can't believe the receiver is in stereo mode.
Great video and explanation! What are you going to make a video on next? Do you have a video on the power system on your reference system? I want to better understand what you have and how you use it.
Top notch video as always (sorry couldn't help myself).
Thank you "Danny" for the in lightful lesson. Now I have a Polk audio monitor 70 series 2 loudspeakers . I'm looking to do an upgrade . I've ordered the parts . I been following you for sometime . Question what if on the Tweeter you grabbed a 14 awg inductor for it , is that a bad move? And or should I get a 18 awg inductor for the Tweeter in line with the caps an resister ? Thanks keep up great work .
What you do to improve cross overs seems fairly easy to do. I’m mystified why the manufacturers don’t do that in the first place!
Price points. Those bookshelf speaker are about $330 a pair. Would be over $500 with the mods. Also alot of times the upgrade crossovers can barely fit into the cabinet. There is forum that is going to take some of GR's upgrade speakers and do blind listening tests and measurements to see if there is a difference.
My Model 5 is sounding great after your up grade, crisp, clean and clear. Also dynamic! Thanks
Typical person with extra Money,never happy with what they have .
What a bizarre comment. It's his money. If he wants to use it to improve the performance of speakers he owns, that's his business.
@@welshrarebit9238 Typical person with no content on their channel, going around talking ****.
Very informative, thanks. Do you have a video on baffle step compensation? I'd be interested in how the baffle dimensions and driver placement, effect frequency response, as far as crossover design is concerned. I assume baffle step compensation might entail some sort of notch filter in some cases?
great video
A Great Fix again. It’s nice to See and understand the changes theese Notch Filters do.
But How about the big Dip ? What Kind of Parts did you use to flaten this Area ?
May God bless Danny.
New to the channel. Love the vids. Im currently pulling apart.my homebrew loudspeaker. Will be completely upgrading the crossovers and using audyn copper max caps as bypass caps o. The mids and tweeters. My question is. What are your thoughts on L pad attenuators. I had I e on the mid and one on the tweeter. Over the last year in listening I kept finding myself setting both to 0 essentially. So im doing new backs o. The cabs and have to do some cab tuning i. General I was just thinking get rid of the L pads. I figure its less stuff I. The signal path anyway. But do you find that you like or should use an L pad or is using an L pad basically an admittance of poor crossover design? Thoughts?
Notch filters were used to get free HBO and spice channel back in the day!
hello there thanks allot for your efort and time to put these videos together,pure gold
i have a Question about notch filter
i`ve build an sb acoustic kit the eka kit listed on their webpage now in the crossover design,on the woofer side there is a notch filter to kill cone break up an an odd resistor in paralel with a second coil thath kill a big hump in the 2k region right before the crossover point
i`ve bild the crossover like they listed in the schematic
i saw in the 1k region a litle wide hump and no bothered but when i listen to music some female and man voice as wel as some electronic piano pierce my ear
playing with an eq i find that the cause is that 1k hump
do not want to use eq
the question is : can a put there a notch filter to lower that 1 k hump ? i`ve played with some parts that i have around and did a sweep and there is a great atenuation on that region but i don`t know if there is a beter solution to that besides using a second notch filter i don`t have a dats to see the impedance an phase
thanks!
I can see you letting a bit more details out of the bag. You are generous.
what would be the appropriate gauge to make inductors. bass 22 , midle 22 , tweeter 24 ? .Of course, then there is the number of turns that we give it.
Could you make a video about series crossovers, as used on Fritz speakers? Advantages vs disadvantages an so forth.
I have actually upgraded some of those to a standard parallel filter.
Hi. Wouldn't changing out the mid/sub driver to one that was a better match be a viable alternative? More expensive than what you did? Thanks for the video. I learned something today. : )
Is a notch filter necessary for 100hz notch or is it too low to affect the sound negatively. NB the notch is in the software curve and not the actual speaker measurement. Haven’t measured the speaker due to lack of calibrated microphone. Cabinet construction imperfections might have removed the notch and added other imperfections but I am happy with it’s sound. I just wanted to know so that in the future when the tinker bug bites then I can measure the speaker and create a notch filter for it.
9:19 ... It looks like that this was a designed feature to me. This seems to be common in many speakers. they are purposely attenuating the midrange up to 5 or 6 dBm which makes the speaker produce more bass and treble than midrange. this appeals to many people as a "more open" speaker and they believe it has greater clarity than a speaker designed with a flat response. However on some recordings especially a small vocal group with a female singer singing lead, her voice just seems to drop out of the mix. A good example would be Pozo Seco Singers LP "Time".
The mid-range is actually in the 300 to 500Hz region. So the dipped area is well up into the treble region.
Thanks Danny, I was wondering the same thing. I'm curious about why you chose to have four woofers in your cc design. Sorry but I have a last request. Here in Canada there are very few places to audition electronics and I'd really appreciate if you could suggest a best bang for the buck max $3000. I will be buying some of you speaker kits but it adds up fast at 250 per flat pack when you want 6 to 11 units.
Great video as always! What is the main contributor to these large impedance spikes in individual drivers, i.e. is it the magnet structure, suspension or surround, etc? Thanks!
Usually driver resonance(s), otherwise the broad upshift is caused by the inductance of the coil.
So one of the challenges of designing a 2 way is to have good speakers cover the larger bandwidth. I am using an Emenence kappalite 3012LF and need to get it to 2K for a crossover of 1600. using a notch filter and a lower crossover on the woofer I think this can be done.BUT does this whack out the phase at the critical crossover and am i kidding myself that the woofer will sound OK through this notch? TYVM
Fundamentally, a bandpass in reverse and over a narrow frequency range?
Hi Danny, can you explain the difference between the notch filter and a Zobel network. My understanding is that the Zobel corrects for impedance anomalies in the driver....so would you use both, combine the two or what?
In my experience notch filters and impedance correction filters are more of a "textbook formula" approach to crossover design. Rarely are they actually necessary except perhaps as a bandaid to "fix" a crossover that was not properly simulated or measured. If you use a textbook formula to make the crossover, then you'll probably need a textbook formula impedance correction circuit to make it even remotely useful.
Better off to simulate the crossover components effect with response, phase, impedance, and diffraction data all incorporated. In that process, you can adjust component values and incorporate resistors in various parts of a 2-4th order network that will "tilt" the response and suppress peaks as part of the network design. This has to be done with care so as not to wind up with a nasty impedance dip, or strong circuit resonance, but there are indicators in the simulation to look out for to keep things reasonable (electric phase angle, impedance dips, sharp hooks or corners in response).
In that process you might wind up with something that looks like a zobel network making up a "leg" of a 2-4th order network, as the final design may benefit from a resistor in series with a capacitor, but the value of the resistor or the capacitor is unlikely to be close to what a textbook formula would spit out for a correction circuit for a given driver impedance.
For example, instead of building a 2nd order low-pass + zobel network for the midbass, (a 3+1 component network), going up to a 3rd order network with an additional series inductor, skipping the zobel network, and "tuning" those component values with a *possible* resistor incorporated *somewhere* in the circuit to fine tune the tilt/shape of the response (Still a 3+1 component cost), is likely to produce a steeper, more controlled rolloff and flatter response with proper tuning.
@@mdocod Thanks for the comprehensive reply.
@@mdocod that was a good explanation also!.can you tell me a good software for the simulation of crossovers??
hi Danny, sorry this question is not about the filter, but the woofers response had that attenuation below 600 hertz. now my question is how come that attenuation is not present in the speaker measurement? thank you
Do you charge labor for reworking these speakers? I'd like to send something in, but it's entirely DIY.
Do you use formulas to determine the value of the crossover components; or variable resistors, etc, whose values can be changed while you're looking at freq response; or just a trial and error methodology?
I was just going to ask exactly the same question :)
@Lloyd Stout please spare me. How does the change in resistance equate to the peak, or bandwidth, or slope of the notch? Do you have a formula for that? How does changing the capacitance or inductance change the same properties?
@Lloyd Stout I'm familiar with capacitive reactance (Xc=1/2pifc), etc. I was asking what HE uses in designing his filters. Thanks.
@Lloyd Stout where in ohms law is frequency mentioned?
👍🙏
Why introduce more parts in the signal path, throw the driver away and start with a better driver, thats what you said as I was typing this!
Hi Danny, Have you ever played around with measuring speakers with different amplifiers hooked up? comparing the amplifiers that way?
Yes.
@@dannyrichie9743 would be cool to do a video on it, compare a few amps…
@@dannyrichie9743 Review with a Topping PA5 in the mix, 2nd highest rated SINAD on Audiosciencereview
The real question is, Why would they start with a woofer with such an uneven response curve? It's a mess.
Money
This offering is all about performance/price ratio.
C+ is good enough.
Do you put the notch filter on the tube connector?
Notch filters are built into the circuit.
220$ plus S&H, wait how much pair of good woofers cost ? Can be better mod.
👍 is that acoustically treated hair? 😂 keep up the great work👍🌈
What would you recommend to do to improve my JBL L100 's besides just the caps? Thanks.
Have you ever done a upgrade for any RSL speakers from Roger Sound Labs out of California?
Nice hair 🤠
Is there any negative unintended side effects of a LCR notch filter such as phase shifting distortion, damping, or ? around the intended bandwidth? You know, ELI the ICE man, if you know that saying. Thanks for sharing your knowledge.
Nope, no ill effects other than more parts in the signal path.
I'm from Russia, subtitles with translation are not available, I'm sad...(((