High pass filter
With the 3rd order high pass filter use MicroCap or one of the alternative provided in the Instructions to obtain the -3 db crossover point of 2.00 hertz. Deduct the 10kohn fixed resistor from the Microcap value to obtain the value for the trim post. Select the C value to scale the R so it fits within the range of the trim pots
Set the dip switches for 3rd order in the high pass section.
With the 3rd order high pass filter use MicroCap or one of the alternative provided in the Instructions to obtain the -3 db crossover point of 2.00 hertz. Deduct the 10kohn fixed resistor from the Microcap value to obtain the value for the trim post. Select the C value to scale the R so it fits within the range of the trim pots
Set the dip switches for 3rd order in the high pass section.
Last edited:
Can anyone suggest the size of capacitor to parallel across the 10uF Nichicon Muse.
I asked AI and it said start at 0.1uF and go up potentially to 0.47uF - does that sound about right?
Hello ZestClub,
I would start with the small 0.1µF- cap. If you do this experiment, then measure at the RCA-outs of your 6-24 AXO for no DC - voltage
(no input / no poweramp connected). There will be no one - but for safety reasons.
Let us know if you can hear a dfference. I think it will be a very small and difficult to hear difference? The Nichicon Muse-caps do a really good job. Especially at their price point. You can always try boutique caps. Let your ears decide.But it is a never ending story
about caps and their influence on the sound.
Cheers
Dirk
I would start with the small 0.1µF- cap. If you do this experiment, then measure at the RCA-outs of your 6-24 AXO for no DC - voltage
(no input / no poweramp connected). There will be no one - but for safety reasons.
Let us know if you can hear a dfference. I think it will be a very small and difficult to hear difference? The Nichicon Muse-caps do a really good job. Especially at their price point. You can always try boutique caps. Let your ears decide.But it is a never ending story
about caps and their influence on the sound.
Cheers
Dirk
I am sure he will once he sorts out what he should be doing.
An active crossover is not a preamp where you simply plug it in and it goes. It needs to be configured correctly or at least in the ballpark to get a desirable result. You unfortunately can’t assume too much with loudspeakers.
My recommendation is to first test your kit is functioning correctly by taking some measurements of the filters with REW using a usb interface like a focusrite 212.
That will give the user certainty at the outset.
Then start to explore what the passive loudspeaker or the drivers are doing on a baffle using REW.
There are a number of helpful guides online with getting started with REW. It’s not difficult. But you do need to be committed to taking that first step if your going to build this kit.
Many thanks Dirk for your advice on this topic and many others throughout this thread 🏆👍
The 10uF Nichicon UKZ are fitted already - any suggestions for a larger Cap to go in parallel?
I have 220uF ELNA caps in stock, or is it important to stay with Nichicon level quality ones?
Nelson designed the BOM to make it easy to order parts, but there is no reason one can't deviate from it if their system allows for it. The 10uF output caps are more than adequate (and sound fine) unless the input impedance of your amps is ridiculously low. Besides, there isn't room to fit a much larger cap.
In my setup, I have replaced the Nichicon NP output caps for the high pass with 1uF Kemet PPS caps (my mid/high amps have 100k input impedance), and they do provide a bit more clarity. They fit with very minor lead bending. I wouldn't bother changing or bypassing the low pass caps, as the NP caps sound very good, and bypassing caps always brings its own risks of negatively affecting the sound.
On the input side, I have installed 0.1uF Mundorf Evo Aluminum Oil caps. Remember, there is a 1 megaohm resistor to ground, so that small cap still provides a suitable cutoff. There is no need for a 10uF cap on the input. The Mundorf also requires some simple lead bending. One could also try the PPS cap there if cost is a concern.
In my setup, I have replaced the Nichicon NP output caps for the high pass with 1uF Kemet PPS caps (my mid/high amps have 100k input impedance), and they do provide a bit more clarity. They fit with very minor lead bending. I wouldn't bother changing or bypassing the low pass caps, as the NP caps sound very good, and bypassing caps always brings its own risks of negatively affecting the sound.
On the input side, I have installed 0.1uF Mundorf Evo Aluminum Oil caps. Remember, there is a 1 megaohm resistor to ground, so that small cap still provides a suitable cutoff. There is no need for a 10uF cap on the input. The Mundorf also requires some simple lead bending. One could also try the PPS cap there if cost is a concern.
Last edited:
Here is some practical advice I've learned after building and fiddling with this kit.
My xovers are not in an enclosure. I used RCA's soldered to the board (Connex from Parts Connexion, but they require some modification to fit.). The xovers rest on a foam pad directly on the floor between the amps, and they are completely silent. If you have wired RCA's, don't attach everything inside an enclosure until you've achieved your best sound. If you're using this kit appropriately, that's somewhat unlikely on the first try.
You can set the values of the pots before soldering them to the board, but you'll probably end up adjusting them. Eventually, because they have no markings, you'll lose your place and not know where the heck you're at. Flip the board over and measure total resistance through each 10k resistor and its associated pot, combined. At least for me, finding and remembering which little solder pads are for which parts was difficult, so I used a fine point marker to draw the signal path for each pot/resistor combo on the back of the board so I could easily find where to attach my probes. If you do it once, you won't need to sort out their locations the next time(s) you want to make an adjustment.
How do you know when you've got it right? FR is not the only indicator, and our ears are easily fooled. Even measuring with REW doesn't reveal every fine point. Listen to a recording with a prominent vocal. When you nail proper phase alignment, voices should be centered between the speakers (assuming it was mixed that way, which most modern recordings are), floating there in three dimensions. I have found that improper phase will spread the voice (and any other centered information) out so it comes from each speaker, not the center. Easy test, no equipment required.
After living with this for a while and realizing you can't make it any better, then go ahead and mount the boards in your enclosure. My speakers sound better than they ever have, after experimenting with various xovers, passive and active, for several YEARS.
My xovers are not in an enclosure. I used RCA's soldered to the board (Connex from Parts Connexion, but they require some modification to fit.). The xovers rest on a foam pad directly on the floor between the amps, and they are completely silent. If you have wired RCA's, don't attach everything inside an enclosure until you've achieved your best sound. If you're using this kit appropriately, that's somewhat unlikely on the first try.
You can set the values of the pots before soldering them to the board, but you'll probably end up adjusting them. Eventually, because they have no markings, you'll lose your place and not know where the heck you're at. Flip the board over and measure total resistance through each 10k resistor and its associated pot, combined. At least for me, finding and remembering which little solder pads are for which parts was difficult, so I used a fine point marker to draw the signal path for each pot/resistor combo on the back of the board so I could easily find where to attach my probes. If you do it once, you won't need to sort out their locations the next time(s) you want to make an adjustment.
How do you know when you've got it right? FR is not the only indicator, and our ears are easily fooled. Even measuring with REW doesn't reveal every fine point. Listen to a recording with a prominent vocal. When you nail proper phase alignment, voices should be centered between the speakers (assuming it was mixed that way, which most modern recordings are), floating there in three dimensions. I have found that improper phase will spread the voice (and any other centered information) out so it comes from each speaker, not the center. Easy test, no equipment required.
After living with this for a while and realizing you can't make it any better, then go ahead and mount the boards in your enclosure. My speakers sound better than they ever have, after experimenting with various xovers, passive and active, for several YEARS.
I fully agree with madisonears. An active crossover is not a 'link in your chain' that is simply plugged in and everything is better.
It deserves -most often - a lot of finetuning and listening.
And it also depends on what speakerdrivers you are using, what amps, filter slopes in the 6-24AXO, ...
I hope, that everybody who is building a 6-24AXO, will have the same great results like I have! I t is a very versatile audiotool.
Cheers
Dirk 🙂
It deserves -most often - a lot of finetuning and listening.
And it also depends on what speakerdrivers you are using, what amps, filter slopes in the 6-24AXO, ...
I hope, that everybody who is building a 6-24AXO, will have the same great results like I have! I t is a very versatile audiotool.
Cheers
Dirk 🙂
sadly I am not yet free to try the suggestion. But I did found out that turning the trimpot anti-clockwise is wrong. I should turn it clovkwise to the end to get 0 ohm
Hello 14071466d,
this is agreat first step. Don't be dissappointed and go on. I am sure, that you will have a smile on your face when the problems
will be solved.
And give us some feedback. We are all willed to learn. You can be sure, that I made a lot of mistakes... Till now without exploding
caps.
Cheers
Dirk 😉
this is agreat first step. Don't be dissappointed and go on. I am sure, that you will have a smile on your face when the problems
will be solved.
And give us some feedback. We are all willed to learn. You can be sure, that I made a lot of mistakes... Till now without exploding
caps.
Cheers
Dirk 😉
Yes 🤣 , thats what i am planning to. But your suggestion of turning it into 3 way speaker sound is actually clever. You mentioned about mix use of active and passive crossover. I wonder how you decide which part use active crossover and which part use passive crossover. Like is it better to use passive design on midrange and bass because of smaller difference on impedence, Therefore less voltage swing and less demand on amplifer?
I actually did rew measurement on speaker frequencty response as a whole in passive crossover design. But that was intended to do the room correction. You are right that I should do measurement too when doing active crossover test.
I think one step at the time. I better learn how to set the active crossover board to work correctly first. 🤣 Essential first big step
Finally got one of these built. My intent was to use high pass to for speakers cut at 45 hz. Low pass for 80hz and below. Main amplifer is Kenwood 11 mk3. Pre out to crossover with high pass back to power amplifer in. Low pass to Kenwood 801 tape in (no pre out main in). The signal is very quiet, as in both ampliers need to be maxed out to barely hear it. I think I already made a mistake on the high pass with caps. I did c and c/2 and it appears that's for the low pass only.
I wanted to double check my resistor and cap selection but it appears the double secret lab link is now dead.
High Pass
R1, R3 18k
R2, R4 35k
C1, C3 150nf
C2, C4 68nf (need to change to 150)
Low Pass
R1, R3 35k
R2, R4 18k
C1, C,3 120nf
C2, C4 60nf
Any advice is greatly appreciated.
I wanted to double check my resistor and cap selection but it appears the double secret lab link is now dead.
High Pass
R1, R3 18k
R2, R4 35k
C1, C3 150nf
C2, C4 68nf (need to change to 150)
Low Pass
R1, R3 35k
R2, R4 18k
C1, C,3 120nf
C2, C4 60nf
Any advice is greatly appreciated.
Hello Ramblr,
i had to check for the Kenwood 11 MK3... This is a stereo amp. I am a little bit confused, how you want to feed 4 necessary channels after the 6-24AXO? You need 2 Stereo poweramps (or 4 monoblocks) after the 6-24 AXO to feed a 2-way stereo setup. Perhaps you have an active subwoofer? This would make sense to me.
Another question: Why do you 'cut' the main speakers at 45 Hz (highpass) and the bass / subsection at 80 Hz (lowpass)?
I could imagine, that you have a two-way-speaker with a smaller bass/mid-driver, which has a rolloff below 100 - 120 Hz?
I will try to simulate your values (R and C) later. I am a little bit busy at the moment with repairs of vehicles...
Perhaps some other members can help also?
Cheers
Dirk
i had to check for the Kenwood 11 MK3... This is a stereo amp. I am a little bit confused, how you want to feed 4 necessary channels after the 6-24AXO? You need 2 Stereo poweramps (or 4 monoblocks) after the 6-24 AXO to feed a 2-way stereo setup. Perhaps you have an active subwoofer? This would make sense to me.
Another question: Why do you 'cut' the main speakers at 45 Hz (highpass) and the bass / subsection at 80 Hz (lowpass)?
I could imagine, that you have a two-way-speaker with a smaller bass/mid-driver, which has a rolloff below 100 - 120 Hz?
I will try to simulate your values (R and C) later. I am a little bit busy at the moment with repairs of vehicles...
Perhaps some other members can help also?
Cheers
Dirk
Hi Dirk,
I have two stereo amps. Kenwood Eleven for Polk M10 speakers, and Kenwood KA-801 for two passive Snell sub500 subwoofers. The subwoofers do not have a crossover of any kind. My reason for this device was to send full range to speakers. 45 Hz was chosen for high pass kind of arbitrarily. It was intended to leave the lower frequencies to the subwoofers and hopefully cut the subsonic noise I get from some records. Without a crossover I'm getting highs through my subwoofers. The frequency response for the M10's is down to 25 hz according to specs but that's doubtful.
Am I better off with a passive 80 hz crossover? Should I pick different frequencies?
I have two stereo amps. Kenwood Eleven for Polk M10 speakers, and Kenwood KA-801 for two passive Snell sub500 subwoofers. The subwoofers do not have a crossover of any kind. My reason for this device was to send full range to speakers. 45 Hz was chosen for high pass kind of arbitrarily. It was intended to leave the lower frequencies to the subwoofers and hopefully cut the subsonic noise I get from some records. Without a crossover I'm getting highs through my subwoofers. The frequency response for the M10's is down to 25 hz according to specs but that's doubtful.
Am I better off with a passive 80 hz crossover? Should I pick different frequencies?
I will be interested in your impressions of the active set up.
My earlier posts were directed to the the active equivalent version of the Criton loudspeaker system.
To the comments about tweaking that applies to any multiway loudspeaker be it active or passive being designed from scratch.
For correct blending the transfer function of the acoustic crossover point must be as accurate as possible.
The trim posts allow movement of the crossover point. The switching allows different slopes. The levels allow trim of the output.
A level shift in the 1-4 khertz region of +-2 db is significant.
As a rule your woofer and tweeter is not a text book perfectly flat response. Therefore by by specific placement of the crossover point you can equalise and optimise the response. By using the trim pots on the tweeter or the woofer as this will modify the acoustic response by moving the voltage drive. There is nothing magical about it.
The psycho acoustics of how we listen in an enclosed listening space - small room acoustics mean that an overall nn flat response may be preferred to a ruler flat response.
The rules of how this is managed depends on the application. Home listening in a non acoustically treated living room, critical listening in a treated room. nearfield recording monitoring in a treated home studio. In the latter listening reference levels are used and the loudspeakers are equalised precisely with Sonar Works or a similar system.
In contrast virtually all home listening is in an uncontrolled listening space with random acoustic interactions with the room. A passive hifi loudspeaker typically has few adjustments. But modern hifi loudspeakers are voiced to produce an enjoyable listening experience in a variety of home listening room situations.
A manufacturer like Kef, Focal or B&W know from experience what voicing characteristics to apply for a given size or design of loudspeaker. The voicing is the overall tonal balance based on listening in a range of home listening situations. Tweeter dispersion is factored into the voicing. Diaphragm materials also become a characteristic.
The voicing is the characterised in the response measurements of the loudspeaker using highly advanced loudspeaker design software by Loudsoft. The program can engineer the driver requirements very accurately. The driver is built by an OEM.
Slight revisions are made with model updates depending on the competition.
So it’s a fairly well understood situation in the manufacturing of modern loudspeakers.
Last edited:
Hello Ramblr,
thanks for the informations. This makes it much better to understand what your goal is.
The POLK M10 speakers use a 5 1/4 inch midwoofer. It is a bassreflex - construction. So I assume, that the spl-curve of this speaker
is down a few (or many dB) around 25 Hz - that is normal for this kind of speakers. I found some specs which say 60 - 24khz (those will be the -3dB points?).
I would drive your system semi-active. Let your POLK M10 be driven directly by one of your amps.
For the subwoofer (Snell SUB500) I would only use the lowpass - section of the 6-24AXO. I can only assume a good crossoverfrequency. From my experience I would say, that around 60 -100 Hz would be a good starting point. Perhaps a 12dB-slope?
You could also make your setup fully active. Then I would setup the 6-24 AXO at 80 - 120 Hz crossoverfrequency. I would use a 24dB-slope. For the Highpass (which would drive the POLK M10 perhaps a 18dB-slope could be better.But I can't tell you what happens in
the bass /lower midrange of your POLK M10?
These are only assumptions..
Perhaps macka can chime in with his opinion?
Greets
Dirk
thanks for the informations. This makes it much better to understand what your goal is.
The POLK M10 speakers use a 5 1/4 inch midwoofer. It is a bassreflex - construction. So I assume, that the spl-curve of this speaker
is down a few (or many dB) around 25 Hz - that is normal for this kind of speakers. I found some specs which say 60 - 24khz (those will be the -3dB points?).
I would drive your system semi-active. Let your POLK M10 be driven directly by one of your amps.
For the subwoofer (Snell SUB500) I would only use the lowpass - section of the 6-24AXO. I can only assume a good crossoverfrequency. From my experience I would say, that around 60 -100 Hz would be a good starting point. Perhaps a 12dB-slope?
You could also make your setup fully active. Then I would setup the 6-24 AXO at 80 - 120 Hz crossoverfrequency. I would use a 24dB-slope. For the Highpass (which would drive the POLK M10 perhaps a 18dB-slope could be better.But I can't tell you what happens in
the bass /lower midrange of your POLK M10?
These are only assumptions..
Perhaps macka can chime in with his opinion?
Greets
Dirk
Dirk,
I double checked my values and found my 68nf was actually a 68pf. I replaced all 4 of high pass customization caps with a .1uf wima and now the high pass has the volume. I still have to turn it up one notch higher than before the crossover was added but the volume now works. The subwoofer amp needs turned pretty high but it is only 110 watts.
I might try the semi active tonight. Maybe try using tape out.
I double checked my values and found my 68nf was actually a 68pf. I replaced all 4 of high pass customization caps with a .1uf wima and now the high pass has the volume. I still have to turn it up one notch higher than before the crossover was added but the volume now works. The subwoofer amp needs turned pretty high but it is only 110 watts.
I might try the semi active tonight. Maybe try using tape out.