Hi all,
In fact I am not sure does that topic belong to here or "loudspeakers" forum???
I am sure that the moderators will move where it must be!
Anyway,
I am planning to make a "cheap" and "hi-fi" (as possible) active speaker box with my stock components.
I have a pair of Westra KW160-1374, 4 ohm, 93dB SPL;
And will buy a pair of Vifa DPC28, 4ohm, 94dB SPL;
My plan is,
- One 12dB/oct sallen key LPF for bass tuned @ 1Khz,
- One 12dB/oct sallen key HPF for tweeter also tuned @ 1Khz, pot on output.
- One T Amp kit from 41Hz (Amp3) that can give 25W per channel @ 4R to use with both bass and tweeter channels.
- One 12v/5A SMPS for each box.. The filters will also work with that PSU.
- Some woodworks...
So...
First; may you say something about that idea? Has anybody tried a project like that?
Second; What corner frequency should I determine for tweeter channel? Because if I select 1Khz for both bass and tweeter channels, then near 1 Khz point I will have "more" loud than I expect because of the slope of the filters..
Any suggestions?
In fact I am not sure does that topic belong to here or "loudspeakers" forum???
I am sure that the moderators will move where it must be!
Anyway,
I am planning to make a "cheap" and "hi-fi" (as possible) active speaker box with my stock components.
I have a pair of Westra KW160-1374, 4 ohm, 93dB SPL;
An externally hosted image should be here but it was not working when we last tested it.
And will buy a pair of Vifa DPC28, 4ohm, 94dB SPL;
An externally hosted image should be here but it was not working when we last tested it.
My plan is,
- One 12dB/oct sallen key LPF for bass tuned @ 1Khz,
- One 12dB/oct sallen key HPF for tweeter also tuned @ 1Khz, pot on output.
- One T Amp kit from 41Hz (Amp3) that can give 25W per channel @ 4R to use with both bass and tweeter channels.
- One 12v/5A SMPS for each box.. The filters will also work with that PSU.
- Some woodworks...
So...
First; may you say something about that idea? Has anybody tried a project like that?
Second; What corner frequency should I determine for tweeter channel? Because if I select 1Khz for both bass and tweeter channels, then near 1 Khz point I will have "more" loud than I expect because of the slope of the filters..
Any suggestions?
As a general rule the crossover frequency should be at least one octave away from the resonance frequency. The tweeter shows a resonance frequency slightly above 1000Hz, and the frequency response also rolls off correspondingly. That roll-off adds up to the crossovers roll-off and might give you a too steep acoustic crossover. So from the tweeters point of view you should choose something like 2000..2500Hz.
On the other hand the woofer has a strong peak at 4000 Hz, probably a material resonance. So, if you choose 2000Hz for crossover that resonance might not be sufficiently filtered out and influence the sound badly. That means from the woofers point of view you should go for either 1000Hz or for a steeper filter, e. g. 18 dB/oct to suppress that 4000Hz peak.
The tricky task is to find the right compromise, maybe somewhere between 1500..2000Hz and the right filter characteristics. On the other hand you may find the 4000Hz peak less disturbing that it looks on the frequency response printout, and land at an even higher crossover frequency.
One more point. You want a symmetric acoustic steepness at the crossing frequency. The acoustic steepness is the addition of the filter steepness and the speakers frequency response (in the cabinet). Maybe you can use some freeware program to simulate the result before you start to build. I haven't tried that program, but maybe Speaker Workshop can be used for that. Or you make the crossover adjustable for later tweaking.
On the other hand the woofer has a strong peak at 4000 Hz, probably a material resonance. So, if you choose 2000Hz for crossover that resonance might not be sufficiently filtered out and influence the sound badly. That means from the woofers point of view you should go for either 1000Hz or for a steeper filter, e. g. 18 dB/oct to suppress that 4000Hz peak.
The tricky task is to find the right compromise, maybe somewhere between 1500..2000Hz and the right filter characteristics. On the other hand you may find the 4000Hz peak less disturbing that it looks on the frequency response printout, and land at an even higher crossover frequency.
One more point. You want a symmetric acoustic steepness at the crossing frequency. The acoustic steepness is the addition of the filter steepness and the speakers frequency response (in the cabinet). Maybe you can use some freeware program to simulate the result before you start to build. I haven't tried that program, but maybe Speaker Workshop can be used for that. Or you make the crossover adjustable for later tweaking.
Hi!
I sugges you to uuse higher cut off frequency depending of before measured thiele small parameters of the drivers you like to use in active speaker, and the best way for that is to make active filter 12 or 18 db/oct two way and use higher cut of freq. 1KHz is too low for that kind of HF driver, My suggestion is 2.5KHz or high .
Don't worry about peak at the filter cut off point freq, that's the jobfor the filter, , but when you finished your project it is very reccomended to measure frequency responce of the complete loudspeaker and if need it to re tune again/ balance of spl level of bass and tweeter also Phase of the tweeter!!!!
Good luck, bujrum.
I sugges you to uuse higher cut off frequency depending of before measured thiele small parameters of the drivers you like to use in active speaker, and the best way for that is to make active filter 12 or 18 db/oct two way and use higher cut of freq. 1KHz is too low for that kind of HF driver, My suggestion is 2.5KHz or high .
Don't worry about peak at the filter cut off point freq, that's the jobfor the filter, , but when you finished your project it is very reccomended to measure frequency responce of the complete loudspeaker and if need it to re tune again/ balance of spl level of bass and tweeter also Phase of the tweeter!!!!
Good luck, bujrum.
Hmmm..
Youre both right!
But what about the impedance curve of the bass driver?
It looks I will have difficulties at especially around 50Hz. I dont care the tweeters, if I use 2Khz or higher HPF..
And another problem with the bass driver;
It looks there is another unlinearity problem between 750-850Hz band (~87dB). And if you look closer to 1Khz point there is a 93-94dB peak! That makes 7dB diffrerence in just 200Hz band....
I think that bass driver is for three way systems. Do you agree with me?
Should I abort that project totally or ignore that unlinearity problem and go with 2Khz filters (may be 24dB ones)...?
Youre both right!
But what about the impedance curve of the bass driver?
It looks I will have difficulties at especially around 50Hz. I dont care the tweeters, if I use 2Khz or higher HPF..
And another problem with the bass driver;
It looks there is another unlinearity problem between 750-850Hz band (~87dB). And if you look closer to 1Khz point there is a 93-94dB peak! That makes 7dB diffrerence in just 200Hz band....
I think that bass driver is for three way systems. Do you agree with me?
Should I abort that project totally or ignore that unlinearity problem and go with 2Khz filters (may be 24dB ones)...?
What problems do you expect at 50Hz? A speaker of that size will not deliver much deeper bass at reasonable volume. Something around 60 to 80Hz in a closed box, and maybe a little below 50Hz in bass-reflex or transmission-line speakers.
The Westra is a low-cost driver that is performing quite well for its price. You will find it in many commercial low-cost 2-way speakers. It is not a high-end speaker, so you will have to accept some unlinearities. And we don't know how that measurement was taken, so there is a lot of information missing to assess its quality.
2000Hz and 12dB is worth a try. Steeper filters bring their own problems about, when it comes to phase and resonator quality factor. Anyhow it takes some trial and error to tune a crossover. One big advantage of having active crossovers is that changing components is much cheaper than in a passive crossover with much bigger values and voltage rating. And you can even make them adjustable by replacing resistors with potentiometers in the right places.
The Westra is a low-cost driver that is performing quite well for its price. You will find it in many commercial low-cost 2-way speakers. It is not a high-end speaker, so you will have to accept some unlinearities. And we don't know how that measurement was taken, so there is a lot of information missing to assess its quality.
2000Hz and 12dB is worth a try. Steeper filters bring their own problems about, when it comes to phase and resonator quality factor. Anyhow it takes some trial and error to tune a crossover. One big advantage of having active crossovers is that changing components is much cheaper than in a passive crossover with much bigger values and voltage rating. And you can even make them adjustable by replacing resistors with potentiometers in the right places.
It looks, it has ~16ohm impedance at 50Hz. That means basses will reduce 4 times at this peak!
As my calculations; I can have 35Hz F3 with 40lt bass reflex box tuned @40Hz.. But if I take care that 16ohm peak then my 35Hz expectation is going to the hell!
And OK... My Westras are not hi-end, I know that. And I dont expect too much things from them. But if you consider my only speaker pair is just Wharfedale 9.1s! Do you think this project is worth to make?
As my calculations; I can have 35Hz F3 with 40lt bass reflex box tuned @40Hz.. But if I take care that 16ohm peak then my 35Hz expectation is going to the hell!
And OK... My Westras are not hi-end, I know that. And I dont expect too much things from them. But if you consider my only speaker pair is just Wharfedale 9.1s! Do you think this project is worth to make?
I hope I will finish (or start).
In fact, the active monitor idea is a very old dream of me... I have tried 2+1 systems but not satisfied from the results.
But these were in old times and I had not enoughly experiences about designing...
The drivers are from these old days and waiting at least for 10 years! Theyre not expensive ones (may be 50€ / pair) but I want to use them in this project..
Anyway,
I will make a decision in these days. And will inform you about the progress.
Have a nice day!
In fact, the active monitor idea is a very old dream of me... I have tried 2+1 systems but not satisfied from the results.
But these were in old times and I had not enoughly experiences about designing...
The drivers are from these old days and waiting at least for 10 years! Theyre not expensive ones (may be 50€ / pair) but I want to use them in this project..
Anyway,
I will make a decision in these days. And will inform you about the progress.
Have a nice day!
Are you prejudiced against Germans? 
You can trust the curve, if you know the measurement conditions. Since we don't know them, the curve only show us a tendency. In a speaker a flat on-axis response is not always the best sounding solution. Much of the SP goes to the sides, and a good sounding speaker usually has a balanced overall energy response. I. e. the average SPL is flat although the on-axis response may not be.
It also depends on the place, where the speaker will be used. The Westra might e. g. make a good bookshelf speaker. If you put a speaker in a bookshelf, the superfices around it will boost the middle bass. A speaker for that purpose may sound better, if it has a dip in just that frequency range.
A free standing speaker in a box suffers from the so called baffle step. A speaker design may take advantage of that and choose the best speaker dimensions to counteract certain irregulatrities in the frequency response. On the other hand, if not taken into account, the baffle step may produce terrible dips and peaks in the frequency response.
Design a crossover with adjustable frequencies. Don't expect to arrive at a decent crossover without listening tests and further adjustments. It is posible that crossing at 2,5..3k gives good results. But it may also result in a very focused directivity in the mids that forces you to a sweet spot for good sound.
And ask in the speaker forum for help. Maybe this is an interesting starting point.
You can trust the curve, if you know the measurement conditions. Since we don't know them, the curve only show us a tendency. In a speaker a flat on-axis response is not always the best sounding solution. Much of the SP goes to the sides, and a good sounding speaker usually has a balanced overall energy response. I. e. the average SPL is flat although the on-axis response may not be.
It also depends on the place, where the speaker will be used. The Westra might e. g. make a good bookshelf speaker. If you put a speaker in a bookshelf, the superfices around it will boost the middle bass. A speaker for that purpose may sound better, if it has a dip in just that frequency range.
A free standing speaker in a box suffers from the so called baffle step. A speaker design may take advantage of that and choose the best speaker dimensions to counteract certain irregulatrities in the frequency response. On the other hand, if not taken into account, the baffle step may produce terrible dips and peaks in the frequency response.
Design a crossover with adjustable frequencies. Don't expect to arrive at a decent crossover without listening tests and further adjustments. It is posible that crossing at 2,5..3k gives good results. But it may also result in a very focused directivity in the mids that forces you to a sweet spot for good sound.
And ask in the speaker forum for help. Maybe this is an interesting starting point.
AndrewT said:
What invest? I will use my "stock" components as mentioned. Anyway, I've asked "is this idea worth to make" and you say "make this this and this etc..",
But its a lot of job and I havent got enough time to do tall these!
"Time-poor man" means that! Not about the money!
Now; May you help me to move this idea to the real world?
pacificblue said:Are you prejudiced against Germans?
You can trust the curve, if you know the measurement conditions. Since we don't know them, the curve only show us a tendency. In a speaker a flat on-axis response is not always the best sounding solution. Much of the SP goes to the sides, and a good sounding speaker usually has a balanced overall energy response. I. e. the average SPL is flat although the on-axis response may not be.
It also depends on the place, where the speaker will be used. The Westra might e. g. make a good bookshelf speaker. If you put a speaker in a bookshelf, the superfices around it will boost the middle bass. A speaker for that purpose may sound better, if it has a dip in just that frequency range.
A free standing speaker in a box suffers from the so called baffle step. A speaker design may take advantage of that and choose the best speaker dimensions to counteract certain irregulatrities in the frequency response. On the other hand, if not taken into account, the baffle step may produce terrible dips and peaks in the frequency response.
Design a crossover with adjustable frequencies. Don't expect to arrive at a decent crossover without listening tests and further adjustments. It is posible that crossing at 2,5..3k gives good results. But it may also result in a very focused directivity in the mids that forces you to a sweet spot for good sound.
And ask in the speaker forum for help. Maybe this is an interesting starting point.
Sorry for misunderstanding.. I love Germans always...
I mentioned that because we dont know the testing conditions..
Anyway,
The adjustable filter idea was always in my mind.. But I am confused on 12/18 or 24dB curve alternatives.. If I will make for example a 24dB/oct filter then the adjustable options becomes too hard to make...
And ok... I will ask that in the loudspeakers forum..
THX..
Ok..
At the beginning (yesterday at 10:00) , I supposed to start an easy and cheap project.. And take some advices from you (or more experienced diyers like you)..
But, by the time (later hours in the day) I saw that it wont be that easy because of the drivers parameters. I understand that fact just after Pacificblue's first posting..
So,
Please forgive me for taking your precious times!
Is it ok now?
At the beginning (yesterday at 10:00) , I supposed to start an easy and cheap project.. And take some advices from you (or more experienced diyers like you)..
But, by the time (later hours in the day) I saw that it wont be that easy because of the drivers parameters. I understand that fact just after Pacificblue's first posting..
So,
Please forgive me for taking your precious times!
Is it ok now?
Hi Ozgur,
Active loudpeakers are much better to customise than passive, I swear by it now. Also they are very cheap to change... fer R and C here and there and changing xo frequency only require $1 or $2 !! compare that with the price of inductors and I'm not talking Solen !!
Build using breadboard first. No, the electrons don't mind. In fact I have 2 systems built on breadboard permanently. One of them is the revered Linkwitz Pluto
It's dry here in australia and there's no humidity/corrosion issues.
Building enclosures is definitely the biggest hurdle and requires simulation etc. Difficult to setup if you're using speaker workshop measurement. But closed enclosure is quite forgiving and gives more natural bass (albeit not as deep as ported).
Let us know how you go with your project !
Active loudpeakers are much better to customise than passive, I swear by it now. Also they are very cheap to change... fer R and C here and there and changing xo frequency only require $1 or $2 !! compare that with the price of inductors and I'm not talking Solen !!
Build using breadboard first. No, the electrons don't mind. In fact I have 2 systems built on breadboard permanently. One of them is the revered Linkwitz Pluto
It's dry here in australia and there's no humidity/corrosion issues.Building enclosures is definitely the biggest hurdle and requires simulation etc. Difficult to setup if you're using speaker workshop measurement. But closed enclosure is quite forgiving and gives more natural bass (albeit not as deep as ported).
Let us know how you go with your project !
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.