I'm going to be building a pair of diy powered AR-LST speakers and need an active 3 way crossover to go in betweeen the input and the amplifiers. I will be using 2 class D amps with 4 channels (1 channel per tweeter) and then another class D amp for the 12 i will be putting in the center.
looking for something that will fit nicely in the back of the speaker with the amps. I was even thinking if I had to I could take a crossover already built in a case, take it out and mount it in the enclosure in my speaker somehow, then take the knobs and build them into a back panel of my own. then figure out a way to power everything including amps and crossover off 1 power plug, with mabe even 1 power supply?
looking for something that will fit nicely in the back of the speaker with the amps. I was even thinking if I had to I could take a crossover already built in a case, take it out and mount it in the enclosure in my speaker somehow, then take the knobs and build them into a back panel of my own. then figure out a way to power everything including amps and crossover off 1 power plug, with mabe even 1 power supply?
This is a very good one if you sub the TLO74 quad opamps for OPA4134's
Silicon Chip Online - Active 3-Way Crossover for Loud Speaker Systems
Cheers George
Silicon Chip Online - Active 3-Way Crossover for Loud Speaker Systems
Cheers George
ESP (Rod Elliott) has a choice of two PCBs. I use the earlier one and am very happy with it.
Linkwitz-Riley Electronic Crossover
Project 125
The boards are reasonably priced and good quality.
Frank
Linkwitz-Riley Electronic Crossover
Project 125
The boards are reasonably priced and good quality.
Frank
Last edited:
I still use the Ben Duncan crossover published in Hi Fi News in the early '80s. You can make it any number of ways, I've used it 2 and 3 way.
"Here are the directions to get to the Ben Duncan HFN/RR 1981 active cross over.
Google "Saturn Sound Recording Services" go to the LHS options, click HISTORY then "The HiFi Section" and then "diyprojects". Fair bit of fine print but it is all there.........The cross over is of course nearly at the bottom of the page! That should do it."
Ben will sell you a reprint of the article with updates, but make sure you make it VERY clear exactly what you want, as he can be hard work! Go to Ben Duncan Research: Archives & Libraries. I now use AD825s from Chris Daley in Oz. Go to my.opera.com and find a member called "opamp".
"Here are the directions to get to the Ben Duncan HFN/RR 1981 active cross over.
Google "Saturn Sound Recording Services" go to the LHS options, click HISTORY then "The HiFi Section" and then "diyprojects". Fair bit of fine print but it is all there.........The cross over is of course nearly at the bottom of the page! That should do it."
Ben will sell you a reprint of the article with updates, but make sure you make it VERY clear exactly what you want, as he can be hard work! Go to Ben Duncan Research: Archives & Libraries. I now use AD825s from Chris Daley in Oz. Go to my.opera.com and find a member called "opamp".
The topologie of the crossover in "Linkwitz-Riley Electronic Crossover", figure 1c, is not the same as in Woofer crossover & offset.
KLaus.
KLaus.
the crossover is the easy bit.
The EQ required for each driver is far more difficult in that there are no standard formulae to determine the component values for the multiplicity of EQ F & Q & Notch & slope etc for each driver. Every one is different and requires a different EQ.
Read Linkwitz site for good info.
The EQ required for each driver is far more difficult in that there are no standard formulae to determine the component values for the multiplicity of EQ F & Q & Notch & slope etc for each driver. Every one is different and requires a different EQ.
Read Linkwitz site for good info.
To significantly ameliorate these problems, I choose drivers that have considerably overlapping frequency response and also fairly flat FR. Therefore, less (nil) EQ to be done.
I've never understood why people would use drivers that are APIT to implement when there are plenty around that aren't.
Frank
I agree with AndrewT. Opamp based Active XO is the easy bit. The harder bit is the ability to change the Frequency, Slope and Propagation delay on the fly and match (attenuate or equalize) the outputs for the drivers to have a nice FR curve.
Some cheap active DIY XO alternative is MiniDSP, DCX2496, etc. And Then, there is more costly DEQX for all the things you want to do with the active setup.
Some cheap active DIY XO alternative is MiniDSP, DCX2496, etc. And Then, there is more costly DEQX for all the things you want to do with the active setup.
Hello,
For me this is the right approach by Riley topology. http://www.linkwitzlab.com/images/graphics/xo_topo-improved-s.png
For me this is the right approach by Riley topology. http://www.linkwitzlab.com/images/graphics/xo_topo-improved-s.png
If you are going to use active XO , read S. Linkwitz and understand how he designed his active crossovers for the Orion and Phoenix. It's all there for your education. Those are 3 way open baffle but the principles are the same for regular 3 ways. The filters are easy, very cookbook. The hard part is the allpass networks in the tweeter path and in the midrange path. Those allpass networks are there to add delay and adjust the phase between the T, MID and MID,W. and assist with the time alignment of the drivers. SL comment in those allpass networks is that they are needed and without them the results will be less than desirable. SL tunes them by ear. Plus, you need BSC (baffle step correction) for your woofer. That network goes in the woofer filter path. Pretty easy and cookbooky to design. Just be aware that you need them. I've simulated all of his circuits using LTSpice and my favorite opamp, LME49860 and find his designs work very well. Use 1% metal film resistors and 2% polyprop capacitors if you can.
OR, do it in the digital domain. The DBX PA+ and Behringer DCX2496 are almost perfect. analog/digital inputs and 6 analog outputs. They can do 4th order LR XO filters, BSC function, automatic equalization using a microphone with pink noise and delay for each output to do the allpass function. Very flexible and powerful. The miniDSP (2X8) is not up to the task, IMHO, yet. Plus it ain't cheaper than the DCX2496 - $288.00USD at Parts Express.
You might need a 6 input system volume control to go between the active XO, whether analog or digital, and the power amps - all 6 on them. This will be hard to find. I had a Rane MA6 but sold it. Wish I had it.
OR, do it in the digital domain. The DBX PA+ and Behringer DCX2496 are almost perfect. analog/digital inputs and 6 analog outputs. They can do 4th order LR XO filters, BSC function, automatic equalization using a microphone with pink noise and delay for each output to do the allpass function. Very flexible and powerful. The miniDSP (2X8) is not up to the task, IMHO, yet. Plus it ain't cheaper than the DCX2496 - $288.00USD at Parts Express.
You might need a 6 input system volume control to go between the active XO, whether analog or digital, and the power amps - all 6 on them. This will be hard to find. I had a Rane MA6 but sold it. Wish I had it.
drivers + crossover circuit - measurements - crossover design = bad
Taking some drivers and matching them with a standard crossover topology will probably result in a very lame sounding speaker. You need to do a full design, taking in to account the driver SPL, frequency response, phase and group delay, and then customizing the crossover functions to work together with the driver responses. You will need to make accurate measurements of the driver responses in order to include all these factors. If you don't want to go to these lengths, just buy a pro audio crossover and tweak it as best you can. It might work, but probably not well.
I am soon to release a set of free active crossover design tools, for those who want to get serious about speaker crossover design. There is a thread in the Multi-way forum about it:
http://www.diyaudio.com/forums/mult...ource-excel-based-crossover-design-tools.html
-Charlie
Taking some drivers and matching them with a standard crossover topology will probably result in a very lame sounding speaker. You need to do a full design, taking in to account the driver SPL, frequency response, phase and group delay, and then customizing the crossover functions to work together with the driver responses. You will need to make accurate measurements of the driver responses in order to include all these factors. If you don't want to go to these lengths, just buy a pro audio crossover and tweak it as best you can. It might work, but probably not well.
I am soon to release a set of free active crossover design tools, for those who want to get serious about speaker crossover design. There is a thread in the Multi-way forum about it:
http://www.diyaudio.com/forums/mult...ource-excel-based-crossover-design-tools.html
-Charlie
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.