Normally I’m not into DIY stuff because what I do is geared toward studio stuff and making music, but lately I’m starting to realize that I have an itch to do more tinkering to get results that may cost upwards of $30k otherwise.
What I want to accomplish is an active system that has digital crossovers taken care of by Trinnov Nova.
What I don’t want to have happen is “bi or tri-amping”, into a speaker that already has passive crossovers which defeats the purpose as I understand.
Here are some options I’ve come across to make this happen, and they are not reasonable. Maybe I’m not looking in the right spots.
XTZ alpha divine speakers allow you completely take out the passive crossover network. I don’t see many speakers like this.
Linn Akurate Exaktbox: these allow for active crossovers upstream of power amps, but is that really necessary? The cost is ridiculous.
Can I take a speaker like the KEF 11r and just direct wire to the drives and essentially “convert” these speakers true actives by using something like the Trinnov Nova to safely take care of crossovers?
Just curious what options are there for this from a DIY approach.
What I want to accomplish is an active system that has digital crossovers taken care of by Trinnov Nova.
What I don’t want to have happen is “bi or tri-amping”, into a speaker that already has passive crossovers which defeats the purpose as I understand.
Here are some options I’ve come across to make this happen, and they are not reasonable. Maybe I’m not looking in the right spots.
XTZ alpha divine speakers allow you completely take out the passive crossover network. I don’t see many speakers like this.
Linn Akurate Exaktbox: these allow for active crossovers upstream of power amps, but is that really necessary? The cost is ridiculous.
Can I take a speaker like the KEF 11r and just direct wire to the drives and essentially “convert” these speakers true actives by using something like the Trinnov Nova to safely take care of crossovers?
Just curious what options are there for this from a DIY approach.
Had a look at the Trinnov Nova and could not initially see if it had crossover‘s built in but to me it looks like a Room EQ system.
The most common system for doing what you want is to use one of the Minidsp boxes to handle crossovers and EQ, add a stereo amp per per driver pair and off you go. There are other digital active crossover boxes about like the Dayton DSP408 a search on here will identify more.
However, it is not as simple as just plugging in and the system works everything out for you. The usual process involves takin impredance and frequency measurements of each driver in the box (need a calibrated microphone and free REW software), then using a simulator like Vituixcad to model the crossover. The crossover is then loaded into the minidsp box and you measure the complete system and iterate until you have both a response you are happy with (assumed flat on axis for studio work) and it produces a sound that you like to listen to.
Another issue I have is that the Trinkov Nova is £5k plus just for the setereo system, you will need the six channel system! Minidsp boxes are around £500 and then you just need three power amps (for a three way speaker pair), the Behringer A800 would be good enough for most applications - so around £1200 all in Plus another £100 for a calibrated microphone.
The most common system for doing what you want is to use one of the Minidsp boxes to handle crossovers and EQ, add a stereo amp per per driver pair and off you go. There are other digital active crossover boxes about like the Dayton DSP408 a search on here will identify more.
However, it is not as simple as just plugging in and the system works everything out for you. The usual process involves takin impredance and frequency measurements of each driver in the box (need a calibrated microphone and free REW software), then using a simulator like Vituixcad to model the crossover. The crossover is then loaded into the minidsp box and you measure the complete system and iterate until you have both a response you are happy with (assumed flat on axis for studio work) and it produces a sound that you like to listen to.
Another issue I have is that the Trinkov Nova is £5k plus just for the setereo system, you will need the six channel system! Minidsp boxes are around £500 and then you just need three power amps (for a three way speaker pair), the Behringer A800 would be good enough for most applications - so around £1200 all in Plus another £100 for a calibrated microphone.
Yes. You may want to begin with a speaker that has a standard acoustic configuration, such as having single units in a straight line on one face of a box.
As Ugg10 suggests, if you expect to improve on the original crossover you will have to lead the way, using the same kind of knowledge that was used to create it.
For my situation, I’ve been waiting to get Trinnov for a number of years now. There’s a few things it can do that miniDSP and other room software correction can’t do (although miniDSP would be my second choice. It has better bass management)
Trinnov hasn’t advertised the active crossovers yet, but I have asked them directly, and they will soon have it available for the NOVA so you will have 6 channels total. I’m confident they will follow through as some of their earlier versions have it. Trinnov can also remote into your system and make adjustments for you.
What you mention about measuring each driver individually and doing calculations, measurements etc, what if you take a speaker that has known crossover points, then use those as a guideline?
Example: the KEF r11 has crossover points at 400 hz and 2900 hz. Wouldn’t I be safe using those values as a starting point?
I suppose taking all the passive components out of the speaker will change things quite a bit in terms of total circuit impedance.
Trinnov hasn’t advertised the active crossovers yet, but I have asked them directly, and they will soon have it available for the NOVA so you will have 6 channels total. I’m confident they will follow through as some of their earlier versions have it. Trinnov can also remote into your system and make adjustments for you.
What you mention about measuring each driver individually and doing calculations, measurements etc, what if you take a speaker that has known crossover points, then use those as a guideline?
Example: the KEF r11 has crossover points at 400 hz and 2900 hz. Wouldn’t I be safe using those values as a starting point?
I suppose taking all the passive components out of the speaker will change things quite a bit in terms of total circuit impedance.
Using the original crossover points is not enough to ensure an improvement. For example, you also want to make up for the natural differences.
What should be an improvement is use the original crossover and apply equalisation on top of it.
What should be an improvement is use the original crossover and apply equalisation on top of it.
If you go to 12:30 on this video, they make it seem doable, but I kinda wondered why I haven’t heard of more people doing this. In other words, they make it sound “too easy” or too good to be true
What the advertised crossover frequency specs don't tell you is what slope they are (6db, 12db, 18db etc) also known as order (1st, 2nd, 3rd etc.) and also what type of filer (Linkwitz Reilly, Bessel, Butterworth etc.) and if any compensation has been applied to account for peaks (notch filter), breakups (tanks), modifiers to the crossover slopes or phase to overcome physical driver offset etc. If you have the original crossover schematic some of this can be inferred.
As for impedance, in a truly active stystem impedance is not an issue as the power amp is directly connected to the driver, most drivers naturally are above 4ohms so most amps will deal with this easily.
Often commercial speakers are also designed to have a "house sound", B&W being a good example that have bass and treble boosts often called a smiley face frequency response, do you want to replicate this or replace it with your own sound signature even if that is flat on axis.
As for impedance, in a truly active stystem impedance is not an issue as the power amp is directly connected to the driver, most drivers naturally are above 4ohms so most amps will deal with this easily.
Often commercial speakers are also designed to have a "house sound", B&W being a good example that have bass and treble boosts often called a smiley face frequency response, do you want to replicate this or replace it with your own sound signature even if that is flat on axis.
You do need acoustic measurements but you can skip the modeling entirely - totally unnecessary- and just tune the DSP crossover to your liking. No need for impedance measurements either.
Also, it’s extremely rare for passive crossovers in commercial systems to actually conform to any formal alignment such as “linkwitz Riley” etc. The manufacturer might say that it is Linkwitz Riley and that might be true electrically, but that doesn’t mean it’s true acoustically.
For that matter, the same comment applies to something as basic as whether it is a 6dB or 12dB crossover or whatever.
If you have the physical crossover, you can figure out what the electrical slope is, but you still don’t know the impedance interactions. Nor do you know what corners were cut. A Crossover is one of the easiest places for a manufacturer to cut corners.
Let’s be clear that you are completely eliminating the original passive crossover. You’ll want to put capacitors in series with the midrange and with the tweeter to reduce the possibility of damage if there is DC in the output or if a cable gets unplugged and causes a buzz.
So if you halfway know what you’re doing, Converting a passive system to active DSP will probably be a considerable improvement over the original.
What you are describing is not often done but it’s one of the easier things to pull off.
For that matter, the same comment applies to something as basic as whether it is a 6dB or 12dB crossover or whatever.
If you have the physical crossover, you can figure out what the electrical slope is, but you still don’t know the impedance interactions. Nor do you know what corners were cut. A Crossover is one of the easiest places for a manufacturer to cut corners.
Let’s be clear that you are completely eliminating the original passive crossover. You’ll want to put capacitors in series with the midrange and with the tweeter to reduce the possibility of damage if there is DC in the output or if a cable gets unplugged and causes a buzz.
So if you halfway know what you’re doing, Converting a passive system to active DSP will probably be a considerable improvement over the original.
What you are describing is not often done but it’s one of the easier things to pull off.
Last edited:
I thought I might slightly revive this thread as what the OP had probably come across was the option for Trinnov products to automatically determine and then implement active crossovers..
This was first done with their cinema processors where it's common to use separate bass units and then horn mids/tweeters for small, independent cinemas with stereo. That gives the cinema excellent dgital and remote control over the audio.
It has been a feature in their studio correction systems too and that new Nova certainly has the feature - in that YouTube video above, the guy originally used their Trinnov to determine rough crossover points when testing the Bliesma drivers in their old ATC cabinets (ATC mid based Questeds actually , I think ) .
Id be interested to know how well their automatic crossover system actually works with a well designed speaker - has anyone tried it? Deqx processors used to or still do, have a similar feature ..
This was first done with their cinema processors where it's common to use separate bass units and then horn mids/tweeters for small, independent cinemas with stereo. That gives the cinema excellent dgital and remote control over the audio.
It has been a feature in their studio correction systems too and that new Nova certainly has the feature - in that YouTube video above, the guy originally used their Trinnov to determine rough crossover points when testing the Bliesma drivers in their old ATC cabinets (ATC mid based Questeds actually , I think ) .
Id be interested to know how well their automatic crossover system actually works with a well designed speaker - has anyone tried it? Deqx processors used to or still do, have a similar feature ..
Hey OP,
I think when people first learn about crossovers they underestimate just how much goes into one. A crossover does frequency splitting but it also does level matching and significant amounts of equalization as well, and those choices are carefully balanced with the challenges of matching the phase through the cvrossover region.. I tried to find you a better example fro my own projects, but I couldn't. Take a look at the blue line, the tweeter. That line has a high pass filter but it also has significant EQ happening from 2 kHz to 20kHz.
While stage speakers may translate well to very simple crossovers, with Hz and db dials, if you want to reproduce an existing passive crossover using exclusively active EQ you need to know what the whole thing does. Of course, many who rip out the original crossovers do not attempt a direct replacement. They use different slopes, add delays, etc. but to do that you should know how to analyze the speaker from end to end.
I like miniDSP for these projects because they are pretty full featured, and affordable. Definitely worth it for a learning project.
I think when people first learn about crossovers they underestimate just how much goes into one. A crossover does frequency splitting but it also does level matching and significant amounts of equalization as well, and those choices are carefully balanced with the challenges of matching the phase through the cvrossover region.. I tried to find you a better example fro my own projects, but I couldn't. Take a look at the blue line, the tweeter. That line has a high pass filter but it also has significant EQ happening from 2 kHz to 20kHz.
While stage speakers may translate well to very simple crossovers, with Hz and db dials, if you want to reproduce an existing passive crossover using exclusively active EQ you need to know what the whole thing does. Of course, many who rip out the original crossovers do not attempt a direct replacement. They use different slopes, add delays, etc. but to do that you should know how to analyze the speaker from end to end.
I like miniDSP for these projects because they are pretty full featured, and affordable. Definitely worth it for a learning project.