Analog crossover can be better sounding than digital, if done right. The big, big advantage to the digital crossover is its flexibility. You can easily have different slopes and frequencies for the different sections. Fine control of phase (delay) is also possible.
To get the best out of your speakers, stock crossover points will not be the choice. They can work, but won't be ideal.
To get the best out of your speakers, stock crossover points will not be the choice. They can work, but won't be ideal.
If you are going fully active, the cap as tvrgeek suggested is for protection not for influencing the slope of the crossover 
you actually want it to not be having any effect for probably at least an octave below the crossover frequency otherwise it may mess up the slope of your active crossover. I'd be guessing a cap of at least 50uF but I could be way off (just simmed that with 6.5 ohms and it is down 3db at about 500Hz.
I'm not sure about the question about putting the cap in the +ve or -ve leg. The norm in passive crossovers is to put it in the positive. I would think that it is better to block the DC before it gets to the driver than after the voice coil though when used as a protection device.
Tony.
you actually want it to not be having any effect for probably at least an octave below the crossover frequency otherwise it may mess up the slope of your active crossover. I'd be guessing a cap of at least 50uF but I could be way off (just simmed that with 6.5 ohms and it is down 3db at about 500Hz. I'm not sure about the question about putting the cap in the +ve or -ve leg. The norm in passive crossovers is to put it in the positive. I would think that it is better to block the DC before it gets to the driver than after the voice coil though when used as a protection device.
Tony.
Thank you! I'm learning a lot here!
All right. I understand that the frequency where the cap cut's off has nothing to do with the minimal cross-over frequency of the driver. The cap has to cross at least an octave below the active cross-over frequency. (so for a 4500hz active cross-over for the tweeter, I could take any cap that starts cutting off at 2250hz? But why a cap of 50µF that is 3db down at 500hz? Because the slope is more steep then the slope of a crossover at e.g. 2250hz? Or because 500hz is safe area that doens't influences the active cross-over slope? Just asking)
According to the spec sheet, the DC impedance of the midrange electrovoice drivers is 4 ohms (it' the DC impedance that matters, right?). I'd like to cross it at about 500hz. So, If I understand right, I'll have to look for a cap that starts cutting of at least an octave below that frequency. Let's say: 250Hz.
Calculated: 4ohms impedance, Cross-over at 250hz --> value cap: about 160µF
Or should I get a lower frequency cross-over point for the cap to not influence the active cross-over point?
One last question: do I need a cap for the woofer too, or can this 'coil' handle some DC of the amp? Or just to avoid phase problems?
PS: As I spent already a lot of money for the speakers, I think I'll go analog for the active cross-over.
Found two of them for a reasonable price (about 80€) in my region. Costed about 300€ new.
- Inkel DIV 23 Cross-over (3-way stereo cross-over, costed about 300€ new. In good condition. XLR inputs. But you have to choose between about 10 predefined cross-over frequencies for each active cross-over)
- Unika PX23S cross-over (made in china, more advanced then the Inkel DIV 23 cross-over, jack 6.3mm inputs, 4-way cross-over (there is also a subwoofer cross-over))
PX-23S - Products, ????(?)??
http://www.hdaudio.com.tw/file/unika/px/px-23s.pdf
any thoughts? Are they worth its money?
All right. I understand that the frequency where the cap cut's off has nothing to do with the minimal cross-over frequency of the driver. The cap has to cross at least an octave below the active cross-over frequency. (so for a 4500hz active cross-over for the tweeter, I could take any cap that starts cutting off at 2250hz? But why a cap of 50µF that is 3db down at 500hz? Because the slope is more steep then the slope of a crossover at e.g. 2250hz? Or because 500hz is safe area that doens't influences the active cross-over slope? Just asking
)According to the spec sheet, the DC impedance of the midrange electrovoice drivers is 4 ohms (it' the DC impedance that matters, right?). I'd like to cross it at about 500hz. So, If I understand right, I'll have to look for a cap that starts cutting of at least an octave below that frequency. Let's say: 250Hz.
Calculated: 4ohms impedance, Cross-over at 250hz --> value cap: about 160µF
Or should I get a lower frequency cross-over point for the cap to not influence the active cross-over point?
One last question: do I need a cap for the woofer too, or can this 'coil' handle some DC of the amp? Or just to avoid phase problems?
PS: As I spent already a lot of money for the speakers, I think I'll go analog for the active cross-over.
Found two of them for a reasonable price (about 80€) in my region. Costed about 300€ new.
- Inkel DIV 23 Cross-over (3-way stereo cross-over, costed about 300€ new. In good condition. XLR inputs. But you have to choose between about 10 predefined cross-over frequencies for each active cross-over)
- Unika PX23S cross-over (made in china, more advanced then the Inkel DIV 23 cross-over, jack 6.3mm inputs, 4-way cross-over (there is also a subwoofer cross-over))
PX-23S - Products, ????(?)??
http://www.hdaudio.com.tw/file/unika/px/px-23s.pdf
any thoughts? Are they worth its money?
Last edited:
There are two approaches to the protection cap in an active crossover. The first is to as wintermute says, make its influence negligible by moving it's cut-off out of the passband. This calls for a big cap though. The second is to use its cut-off as part of the crossover. This doesn't need the cap to be as big but you do suffer from a loss of damping around the crossover frequency, which is one of the benefits of going active in the first place.
It is your choice which route you take. I bought some 40uF motor run caps to use as protection caps, but they are almost the size of coke cans and I never bothered fitting them as my amps are all relay controlled/protected.
It is your choice which route you take. I bought some 40uF motor run caps to use as protection caps, but they are almost the size of coke cans and I never bothered fitting them as my amps are all relay controlled/protected.
Coke cans??? Ok... :s
As I would not like to influence the active x-over point, I'd take the first approach. I'll think about it and ask prices at the local electro-shop.
Another (completely different) question: I'm not very happy with the horns on the ev-1823M drivers. The horns are bolt-on horns that are fitted to the drivers with a bolt-on/screw-on adapter https://reconingspeakers.com/products-page/horns-drivers-xovers/adapter-bolt-hornscrew-driver/). Not very pro. I don't find any specs of the horns (bought used for 15$), and the dispersion isn't great.
Someone had an idea for a good but acceptably priced horn that fits the midrange screw-on driver?
As I would not like to influence the active x-over point, I'd take the first approach. I'll think about it and ask prices at the local electro-shop.
Another (completely different) question: I'm not very happy with the horns on the ev-1823M drivers. The horns are bolt-on horns that are fitted to the drivers with a bolt-on/screw-on adapter https://reconingspeakers.com/products-page/horns-drivers-xovers/adapter-bolt-hornscrew-driver/). Not very pro. I don't find any specs of the horns (bought used for 15$), and the dispersion isn't great.
Someone had an idea for a good but acceptably priced horn that fits the midrange screw-on driver?
good quality polyprop caps in big values (that aren't quite as big as coke cans) and won't break the bank, can be had at partsconnexion AXON True Cap Metallized Polypropylene Capacitors
I used Axons in my passive crossover and have no complaints at all. The biggest I have are 51uF which is still completely manageable Pic of one of my crossovers attached The coils are much bigger than the 51uF cap (it measures 53mm long with diameter ~38mm)
Tony.
I used Axons in my passive crossover and have no complaints at all. The biggest I have are 51uF which is still completely manageable
Pic of one of my crossovers attached The coils are much bigger than the 51uF cap (it measures 53mm long with diameter ~38mm) Tony.
Attachments
Last edited:
Ok, thank you!
So I'd need 2 x 50µF caps for the tweeters, what do you suggest for the mids?
ev1823M have a DC impedance of 4 ohm and I'd like to do an active cross at about 500hz.
(I guess it's better just to ask, because I can't simulate those X-over slopes and define good values)
Any thoughts about a good midrange horn?
So I'd need 2 x 50µF caps for the tweeters, what do you suggest for the mids?
ev1823M have a DC impedance of 4 ohm and I'd like to do an active cross at about 500hz.
(I guess it's better just to ask, because I can't simulate those X-over slopes and define good values
)Any thoughts about a good midrange horn?
Pano is right!! That's unfortunate it is 4 ohms. An actual impedance sweep would be good as it might show that it is higher in the range that matters.
Simulating at 4 ohms impedance, even with 400uF shows 0.65db down at 250Hz. 800uF takes that to 0.17db... so I think for the mid you are getting into teritory where a protection cap is going to have to be part of the slope. If you go digital you should be have the flexibility to deal with it if you are doing measurements.
The cost of caps big enough to not have an effect in the pass band on the mid is probably going to be too high. Unless you go with electrolytics. You could use two of these 400uF 100V Non-Polarized Capacitor 027-376 in parallel on each mid for 800uF. PE have 200uF polyprops, but they are $64 each
I won't get into whether or not electros will be audible or not...
You could also look into DC protection circuitry. Basically a circuit that detects DC on the output and disconnectes (via a relay) the speaker output of the amp. Often these are used for turn on and turn off muting as well to stop pops and thuds. One of your existing amps may already have this as well.
BTW once you know a few basics, simulating a circuit like this with LTspice is really easy (and it is free). Attached is the ASC file I used if you want to have a play with it to change the value of the components just right click on them and change in the dialog box that comes up. Click the little running man to run the sim, and click on the out1 and out2 to get the waveforms to come up.
Tony.
Simulating at 4 ohms impedance, even with 400uF shows 0.65db down at 250Hz. 800uF takes that to 0.17db... so I think for the mid you are getting into teritory where a protection cap is going to have to be part of the slope. If you go digital you should be have the flexibility to deal with it if you are doing measurements.
The cost of caps big enough to not have an effect in the pass band on the mid is probably going to be too high. Unless you go with electrolytics. You could use two of these 400uF 100V Non-Polarized Capacitor 027-376 in parallel on each mid for 800uF. PE have 200uF polyprops, but they are $64 each
I won't get into whether or not electros will be audible or not...
You could also look into DC protection circuitry. Basically a circuit that detects DC on the output and disconnectes (via a relay) the speaker output of the amp. Often these are used for turn on and turn off muting as well to stop pops and thuds. One of your existing amps may already have this as well.
BTW once you know a few basics, simulating a circuit like this with LTspice is really easy (and it is free). Attached is the ASC file I used if you want to have a play with it
to change the value of the components just right click on them and change in the dialog box that comes up. Click the little running man to run the sim, and click on the out1 and out2 to get the waveforms to come up. Tony.
Attachments
I bought an active electronic crossover. It has XLR outputs and inputs, but I'll look for some adapters so I can connect it with the amps (rca)
I guess the Marantz has a built-in protection circuit (you can hear the relay clicking a second after turning on the amp), so there is no extra DC protection needed I guess.
Now, for the mids and highs I could use the DIY 6 channel amp (without DC protection). What about building a DC-protection in the DIY amp? I found this kit from velleman, a company that provides DIY electronic kits: Velleman nv - Item - Details
It costs about 20$ for a 2 channel protection kit. I'll need two of them.
There is a second solution: I have a nice sounding Kenwood KA5700 stereo amp (with DC protection) that needs some clean-up, but could do the job for the mids and the highs, together. But with this stereo amp the mids and highs have to be crossed with a passive filter.
What should I do? Going 3-way active and rebuilding my DIY amp with speaker protection while I'm looking for 2 good stereo amps that will do the job later?
Or going 2-way active = staying with the marantz (lows) and the Kenwood (mids+highs) and using the passive filter that I use now for crossing the mids and the highs?
I guess the Marantz has a built-in protection circuit (you can hear the relay clicking a second after turning on the amp), so there is no extra DC protection needed I guess.
Now, for the mids and highs I could use the DIY 6 channel amp (without DC protection). What about building a DC-protection in the DIY amp? I found this kit from velleman, a company that provides DIY electronic kits: Velleman nv - Item - Details
It costs about 20$ for a 2 channel protection kit. I'll need two of them.
There is a second solution: I have a nice sounding Kenwood KA5700 stereo amp (with DC protection) that needs some clean-up, but could do the job for the mids and the highs, together. But with this stereo amp the mids and highs have to be crossed with a passive filter.
What should I do? Going 3-way active and rebuilding my DIY amp with speaker protection while I'm looking for 2 good stereo amps that will do the job later?
Or going 2-way active = staying with the marantz (lows) and the Kenwood (mids+highs) and using the passive filter that I use now for crossing the mids and the highs?
Again: a well done passive crossover is often better than an 'adapted' (as in not specifically designed/tailored for the drivers) Active.
And there's' none of the exact same gain requirement minefield.
Some have even Bi amped through their passives with good result too.. same gain issues though. Best (necessary?) to have identical amps though
No need to mention?? that the tonal differences in a multi amp setup can cause genuine havoc /chaos in the final sound balances
Lots of experimentation required, along all directions.
And there's' none of the exact same gain requirement minefield.
Some have even Bi amped through their passives with good result too.. same gain issues though. Best (necessary?) to have identical amps though
No need to mention?? that the tonal differences in a multi amp setup can cause genuine havoc /chaos in the final sound balances
Lots of experimentation required, along all directions.
Last edited:
Thank you for your input, Bare. But passive filters are not the way that I will go now. Simply because I don't know at all how to make them.
I rather like to make an acceptable stereo setup with good drivers that aren't disturbed by a cheap passive off-the-shelf filter.
I could buy a lot of passive filters and try, try, try until I get a sound that I like, but this would be expensive and not creative at all.
To be honest, I like the idea of controlling completely the speakers by searching for good cross-over points and changing the volumes of the different frequency-bands and to listen, listen, listen until I get a good sound and to measure it eventually (and searching, listening,.... again).
Besides, the active cross-over setup gives me more possibilities if I wanted to change a driver/a horn/the complete speaker in the future
The low frequencies will be done by the Marantz amp
The mid and high frequencies will be done by two equal amps with a similar gain as the Marantz.
Thank you all for your input and ideas, I can't wait untill I can start experimenting and testing
PS: is REW a good program to measure SPL? My father has a good mic so this wouldn't be a problem.
Kind regards
I rather like to make an acceptable stereo setup with good drivers that aren't disturbed by a cheap passive off-the-shelf filter.
I could buy a lot of passive filters and try, try, try until I get a sound that I like, but this would be expensive and not creative at all.
To be honest, I like the idea of controlling completely the speakers by searching for good cross-over points and changing the volumes of the different frequency-bands and to listen, listen, listen until I get a good sound and to measure it eventually (and searching, listening,.... again
).Besides, the active cross-over setup gives me more possibilities if I wanted to change a driver/a horn/the complete speaker in the future
The low frequencies will be done by the Marantz amp
The mid and high frequencies will be done by two equal amps with a similar gain as the Marantz.
Thank you all for your input and ideas, I can't wait untill I can start experimenting and testing
PS: is REW a good program to measure SPL? My father has a good mic so this wouldn't be a problem.
Kind regards
REW is a nice program, I've only used the pink noise function for SPL measurements in room, I've not tried the gated measurements with it. I used holmimpulse for that.
REW is also good for measuring impedance curves of the drivers (with a suitable simple jig).
Whether active or passive you ultimately want to aim for acoustic slopes of the drivers matching your crossover target, not the filter slopes. for example a 12db flter slope when combined with the driver will often give a 24db accoustic slope (due to the drivers natural rolloff). However if one of the drivers is relatively flat in the range of the crossover frequency then it may only roll off at the filter slope of 12db. It's of course a bit more complex than that, but hope that gives an idea.
Tony.
REW is also good for measuring impedance curves of the drivers (with a suitable simple jig).
Whether active or passive you ultimately want to aim for acoustic slopes of the drivers matching your crossover target, not the filter slopes. for example a 12db flter slope when combined with the driver will often give a 24db accoustic slope (due to the drivers natural rolloff). However if one of the drivers is relatively flat in the range of the crossover frequency then it may only roll off at the filter slope of 12db. It's of course a bit more complex than that, but hope that gives an idea.
Tony.
Thinking about 2 stereo amps for the mids and the highs, I read something about 'gain clone amplifiers'. Almost everybody love these kits because they seem affordable and are sounding very nice!
I can choose between 4 x mono LM3886 amps (68W /channel) or 2 x stereo LM4780 amps, which is a dual LM3886.
I prefer 2 x LM4780 because it would be cheaper and smaller. Is there a big difference between 4 x mono and 2 x stereo?
I can choose between 4 x mono LM3886 amps (68W /channel) or 2 x stereo LM4780 amps, which is a dual LM3886.
I prefer 2 x LM4780 because it would be cheaper and smaller. Is there a big difference between 4 x mono and 2 x stereo?
I'm wondering just how critical it is to have a high damping factor for a driver at frequencies not close at all to the resonant frequency of the driver. At resonance, there is a high level of back EMF generated, but other than at resonance, not much at all. Have you seen any hard data (measurements) of distortion vs damping factor at frequencies other than the resonant frequency of the driver?
Regards,
Pete
Hey, I didn't mean to bring all of the discussion to an END! If you all don't like my question, you can certainly continue with what you were discussing and totally ignore my question. For sure, I've had that happen before.
I realize that what I'm asking is outside of the mainstream of what the thread is about. But I do think that a quick reply by somebody wouldn't hurt.
Where are the great defenders of crossing over actively? What I'm suggesting is a frontal assault on the supposed great advantages of active crossovers.
Please do excuse my impertinence for posting in this thread. I didn't realize that I wasn't invited to the party.
-Pete
I realize that what I'm asking is outside of the mainstream of what the thread is about. But I do think that a quick reply by somebody wouldn't hurt.
Where are the great defenders of crossing over actively? What I'm suggesting is a frontal assault on the supposed great advantages of active crossovers.
Please do excuse my impertinence for posting in this thread. I didn't realize that I wasn't invited to the party.
-Pete
How about a DSP?
Hello everybody,
after reading trough this thread I think a DSP could be a very nice option for this system. miniDSP comes to mind. Home | miniDSP
complete freedom in choosing crossover types, slopes, time alignment, ...
If you want to try simulating filters (frequency response, phase response, group delay, ..) also take a look at AADE filter design: http://aade.com/filter.htm easy to use and free.
Hello everybody,
after reading trough this thread I think a DSP could be a very nice option for this system. miniDSP comes to mind. Home | miniDSP
complete freedom in choosing crossover types, slopes, time alignment, ...
If you want to try simulating filters (frequency response, phase response, group delay, ..) also take a look at AADE filter design: http://aade.com/filter.htm easy to use and free.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.