I really don't understand why I'm having such negative feedback here. I do not have an "Over-simplified" view of what it takes to design a crossover. I specifically said that I was "Not looking for perfection".
My original post read - "I have been looking to do a speaker build for many years and need A LITTLE GUIDANCE. I am here looking for help from someone with experience and the necessary software to POINT ME IN THE RIGHT DIRECTION.".
I literally just wanted to bullsh** with someone back and forth through email or over the phone about speaker design and have them give me some pointers and maybe use some software to help me design a simple crossover for a personal project. I have owned many hifi speakers over the years including Monitor Audio, Tannoy, Wharfedale, Martin Logan, Dali and Vandersteen. I understand that it takes years of dedication and perseverance to design a pair of Wilson Alexandria's, but I as I said, I am not looking for perfection.
I have a very good idea of the functions of a crossover from all my time spent recording and engineering. I have worked for 15 years with equalizers, high, low and bandpass filters, notch filters, attenuation etc, but I have never designed and/or built a crossover myself. I figured with a little guidance from someone with experience building speakers I could build a simple 1st or 2nd order 3-way crossover for this project, have a hell of lot of fun in the process and hopefully end up with a decent sounding pair of speakers.
I absolutely love speakers and have always wanted to build a pair. I have been looking through Parts-Express's list of DIY projects for years and years and have always wanted to build a middle of the road project. I was just looking for a little help with driver selection and crossover design. I thought it would be beneficial to have a chat with someone who has a few builds under their belt and knows a bit more about the ins and outs than I do.
My original post read - "I have been looking to do a speaker build for many years and need A LITTLE GUIDANCE. I am here looking for help from someone with experience and the necessary software to POINT ME IN THE RIGHT DIRECTION.".
I literally just wanted to bullsh** with someone back and forth through email or over the phone about speaker design and have them give me some pointers and maybe use some software to help me design a simple crossover for a personal project. I have owned many hifi speakers over the years including Monitor Audio, Tannoy, Wharfedale, Martin Logan, Dali and Vandersteen. I understand that it takes years of dedication and perseverance to design a pair of Wilson Alexandria's, but I as I said, I am not looking for perfection.
I have a very good idea of the functions of a crossover from all my time spent recording and engineering. I have worked for 15 years with equalizers, high, low and bandpass filters, notch filters, attenuation etc, but I have never designed and/or built a crossover myself. I figured with a little guidance from someone with experience building speakers I could build a simple 1st or 2nd order 3-way crossover for this project, have a hell of lot of fun in the process and hopefully end up with a decent sounding pair of speakers.
I absolutely love speakers and have always wanted to build a pair. I have been looking through Parts-Express's list of DIY projects for years and years and have always wanted to build a middle of the road project. I was just looking for a little help with driver selection and crossover design. I thought it would be beneficial to have a chat with someone who has a few builds under their belt and knows a bit more about the ins and outs than I do.
If you have experience with electronic crossover/notch and understand recording etc. then running a dsp multi amp system may be a better option. Get yourself a minidsp 4x10hd and a umik-1 plus a 6 channel poweramp (rotel rmb1066, cheap second hand?), build the speaker you think will work using on line calculators and then use the dsp to set crossover and notches until you get a sound you like.
Otherwise, I stand by others comments, find a design that is already done and copy.
If the 4x10hd is a bit too much £/$ then use two standard 2x4 and work up from there.
Or go back to my original suggestion and get the double gem plus ls50 and add a minidsp 2x4 or 2x4hd plus umik-1 to hand the crossover and also sort out room eq.
Otherwise, I stand by others comments, find a design that is already done and copy.
If the 4x10hd is a bit too much £/$ then use two standard 2x4 and work up from there.
Or go back to my original suggestion and get the double gem plus ls50 and add a minidsp 2x4 or 2x4hd plus umik-1 to hand the crossover and also sort out room eq.
Just getting "working well together" on a coax is a heavy lift. I don't think you're keeping the right perspective on what effort "good enough" will actually take with you having already penciled in a design decision without having considered all the moving pieces.
I'd recommend finding a good commercial coax you like the look of that sells a go-along crossover, then basically just adding subs, even if they're in the same cab. Then you can high pass the coaxes actively (in many setups) and just tweak the sub integration with the sub amp until it sounds right.
There are plenty here happy to chat speakers all day long, but you have requirements without having done as much of the homework- generally that leads to less than productive collaboration. If someone asks me to help design things but has already made design decisions without my input, it's pretty hard for me to care about optimizing things- the process is already compromised.
That's an honest assessment of what the issue you're having here is. DIYaudio, the first time I saw it, was a little overwhelming. We have some of the finest minds on the planet here, quite literally. It's almost never IRL that I encounter something I'm completely ignorant on, but around here, you'll encounter some people who will write an essay that makes you feel pretty far out of your depth even on topics you thought you knew the basics of. Even the finest designers have SOME multidisciplinary limitations. I'm pretty poor at solid state electronics (not saying I'm in that "finest" range for the other stuff but I'm pretty solid with speakers).
Last edited:
jReave - Thank You! Much appreciated.
I was unaware of what baffle compensation was until now. I was concerned about the overall SPL of each driver being different from one another once in their enclosures which is why I planned on bi-amping, but I would prefer to have the opportunity to power these speakers with one amplifier in the future without the woofers over powering the coaxial. If changing the drivers makes that more viable I am all for it.
I do like the look of the Peerless and SB acoustics drivers you suggested, but I seem to be getting a lot of feedback reporting that coaxials are more difficult to work with. I was aiming at Coaxial for their off-axis response and notorious imaging, but if it really makes things that much easier on the design end I am open to switching this to a separate tweeter and midrange.
If I wire the SB acoustics tweeter and midrange in series (Both 4ohm) and the Peerless woofers in parallel (Both 8ohm) then I should end up with 8 ohms for the Tweeter/Midrange and 4 ohms for the woofers which would net me 6 ohms if I powered these with one amplifier. This is a good place to start, no?
The SB Acoustics says it will get me an F3 of 90hz in a .25 cubic foot sealed cabinet. Is this large enough to absorb some of the backwave you were talking about?
I was unaware of what baffle compensation was until now. I was concerned about the overall SPL of each driver being different from one another once in their enclosures which is why I planned on bi-amping, but I would prefer to have the opportunity to power these speakers with one amplifier in the future without the woofers over powering the coaxial. If changing the drivers makes that more viable I am all for it.
I do like the look of the Peerless and SB acoustics drivers you suggested, but I seem to be getting a lot of feedback reporting that coaxials are more difficult to work with. I was aiming at Coaxial for their off-axis response and notorious imaging, but if it really makes things that much easier on the design end I am open to switching this to a separate tweeter and midrange.
If I wire the SB acoustics tweeter and midrange in series (Both 4ohm) and the Peerless woofers in parallel (Both 8ohm) then I should end up with 8 ohms for the Tweeter/Midrange and 4 ohms for the woofers which would net me 6 ohms if I powered these with one amplifier. This is a good place to start, no?
The SB Acoustics says it will get me an F3 of 90hz in a .25 cubic foot sealed cabinet. Is this large enough to absorb some of the backwave you were talking about?
Honestly, when I saw the start of this thread my immediate idea was similar - partially since I've been playing around with a similar idea - a coax + single woofer for a 'desktop' system (I have a big desk, hence being able to look at the RSS210)
IMHO OP should just go buy a pair of Kef Q150s when they're on for $299 (happens about 2x per year). You're not going to find a better coax than this at anywhere close to the price. It's a ~5.5" rather than a 6.5, but hopefully close enough. You can pull the driver and xover and build them into a new cabinet if desired, or just use them as-is. (If you really want to go top-shelf, get the LS50 or an R series, but the Q150 is plenty good on it's own)
For the woofers, based on my experience I'd suggest dropping down to 8" drivers and use a pair of the Dayton RSS210HF-4 per side. They work very well sealed in 0.75-1.0 cubic ft per driver, and a 2 cubic foot cabinet is easy to build into a speaker-stand type of shape. These will easily play up to 2-400Hz to mate with a high-passed Q150. I don't think any of the 10" (RSS265) variants do quite as well in sealed enclosures, but I'd have to double check. If you're willing to go ported though, the RSS265HO-4 is an amazing driver in small cabinets, but might need an xover closer to 200Hz due to higher VC inductance (that would come out in your measurements/sims when you get to that point)
As always, there are lots of ways to approach a project, but the above would result in a very good system.
I have definitely considered using a proven design like the KEF Q350 or LS50 or even the little brother of my Tannoy XT6Fs, but the problem I keep seeing is adding a high pass to one of these. Obviously if I bi-amp I can use some type of bass management or RCA filters, but can I add to the existing crossover in the kEFs? That would honestly be perfect if it’s possible. I assumed it would be difficult to rip apart the very well engineered KEF crossover and add a high pass to the midrange/woofer, no?
Two Peerless give the attached response in 70L with a Qtc of .832. F3 is about 50Hz although it can be lowered a touch with a bigger box, up to about 107L.
I haven't worked with a coaxial before so I'm not sure that they are any harder to design with than a separate tweeter and mid, except for the fact that many previous coaxials have pretty bad frequency responses. Another reason I like the SB - fairly nice responses except for the tweeter on axis at about 7kHz but that corrects itself off-axis so this might be a speaker best listened to slightly off-axis.
No. Speaker impedance doesn't work this way. All the drivers will be wired in parallel but the xo's in front of each driver mean they operate more or less individually so you'll end up with a 4ohm speaker in the end.
F3 is essentially unimportant for the mid in a 3-way because you'll be rolling the response off earlier at about 300Hz with a HP filter. For the SB, a Qtc of .5 is 13L. Somewhere around there would give the driver enough room to breath while leaving enough space to add a goodly amount of absorption material.
Now I could be wrong here, but I think part of your frustration is the result of a false assumption that simulations alone using manufacturers' data will produce a very good speaker. It's possible to get close and sometimes to even hit the nail on the head, but more often than not you need to take real world measurements to get the most out of your drivers. In particular, you need the drivers' in-box frequency and impedance responses plus a measurement of the drivers acoustic centers. Even then, you still need to listen and make adjustments.
To me it sounds like you've been dreaming about building this speaker for a while. Because you're not actually looking for state of the art but as you say, good enough, I would suggest that you try out some not too expensive drivers using just the simulations and see how that goes. If perhaps you are not completely happy with the results, you may want to try diving just a little deeper and learn to do the necessary measurements to fine tune your results. With some help of course. Just so long as you understand that there are no guarantees here.
My 2 cents anyways......
Attachments
No need, as above, use a minidsp 2x4 to manage the crossover between the ls50 (Or other coax) and the sub. The other benefit is that it does room eq which will do more for the sound of the speakers than any passive component. The benefit is you can try a whole range of crossover frequencies and slopes plus add notches/boosts to get the frequency profile you want.
As per the above recommendation, if you use either a good DSP crossover like a Minidsp 2x4hd, or an analog XO like a paradigm x-30 or outlaw ICBM, you can crossover an LS50 to a bass bin like you described. Ideally around 300hz, but anything above 125hz or so would make a great system. As far as I know there are no commercially available DIY coax drivers that can match the KEF drivers, and even if there were, you'd really want measurement and XO modelling capability and experience to get good results.
If you search this forum there are several threads where people are taking this approach, even redoing the LS50 xo while they're at it. IMO this approach will yield as close to a state of the art coax system as is currently possible without huge cost or effort; the downside being I suppose that there's less 'diy' involved. It's a great way to get acquainted with DIY while building your basic skills and experience though, and still getting a top notch result.
If you search this forum there are several threads where people are taking this approach, even redoing the LS50 xo while they're at it. IMO this approach will yield as close to a state of the art coax system as is currently possible without huge cost or effort; the downside being I suppose that there's less 'diy' involved. It's a great way to get acquainted with DIY while building your basic skills and experience though, and still getting a top notch result.
@OP:
you may believe you are not oversimplifying the speaker design process.
However, you need to get a lot more educated on the subject. This link (and the whole website for that matter), would be a good start:
But you want to design your own loudspeaker
People spend hundreds of hours refining their speaker designs. Ask Joe R. on Elsinores thread. Btw I am a professional in the noise & vibration field doing a lot more complicated stuff than a speaker design. Similar to the late Linkwitz who had a PhD in electronics and dealt with RF which is an order of magnitude more complex than anything in the audio frequency range. Like Joachim G. (who btw was also designing/selling speakers priced in tens of thousands of Euros through AudioPhysic) said here: Siegfried could design a spectrum analyzer let alone an amp or a speaker. You should be aware that all these super qualified people post or posted here.
That is why I will usually start with a design by somebody like SL who was a published speaker designer 40 years ago already. Especially if another guy (John K.) also converged independently on the same min baffle shape (in his NaoNote designs), and after I have made their design I will set out small incremental goals to improve on those to suit my application, (for example get the sensitivity higher so I can use smaller wattage class A amps designed by Pass instead of a Chinese 250W/Ch class AB amp to drive power hungry drivers of their choice on a min baffle all the way down to bass), etc. etc. That is how you go about this hobby. You learn from the more knowledgeable first and then then you take pleasure in doing something yourself.
Since you mentioned $50 and 30min: FYI I took 90min earlier this week to just write up an excel VB routine that will decimate files coming out of my professional sound measurement equipment so that I could import them into the Passive Crossover Designer written by Jeff Bagby for older versions of Excel. Btw I could not get a plumber to show up at my door without $150.
I already pointed you to a ready made DIY design using a coax in your choice of size (by a reputable manufacturer like Seas). It measures very well and it has a passive (albeit complex) Xover fully engineered and tested. If those guys gave you documentation for $50 it would be A GIFT. I will spend that on shipping cost when buying iterations of components for a passive Xover.
you may believe you are not oversimplifying the speaker design process.
However, you need to get a lot more educated on the subject. This link (and the whole website for that matter), would be a good start:
But you want to design your own loudspeaker
People spend hundreds of hours refining their speaker designs. Ask Joe R. on Elsinores thread. Btw I am a professional in the noise & vibration field doing a lot more complicated stuff than a speaker design. Similar to the late Linkwitz who had a PhD in electronics and dealt with RF which is an order of magnitude more complex than anything in the audio frequency range. Like Joachim G. (who btw was also designing/selling speakers priced in tens of thousands of Euros through AudioPhysic) said here: Siegfried could design a spectrum analyzer let alone an amp or a speaker. You should be aware that all these super qualified people post or posted here.
That is why I will usually start with a design by somebody like SL who was a published speaker designer 40 years ago already. Especially if another guy (John K.) also converged independently on the same min baffle shape (in his NaoNote designs), and after I have made their design I will set out small incremental goals to improve on those to suit my application, (for example get the sensitivity higher so I can use smaller wattage class A amps designed by Pass instead of a Chinese 250W/Ch class AB amp to drive power hungry drivers of their choice on a min baffle all the way down to bass), etc. etc. That is how you go about this hobby. You learn from the more knowledgeable first and then then you take pleasure in doing something yourself.
Since you mentioned $50 and 30min: FYI I took 90min earlier this week to just write up an excel VB routine that will decimate files coming out of my professional sound measurement equipment so that I could import them into the Passive Crossover Designer written by Jeff Bagby for older versions of Excel. Btw I could not get a plumber to show up at my door without $150.
I already pointed you to a ready made DIY design using a coax in your choice of size (by a reputable manufacturer like Seas). It measures very well and it has a passive (albeit complex) Xover fully engineered and tested. If those guys gave you documentation for $50 it would be A GIFT. I will spend that on shipping cost when buying iterations of components for a passive Xover.
I think going with the KEF LS50 is my best option. Is there a way I can add some crossover components to the LS50 OUTSIDE the enclosure? Not a dsp or Rca crossover. I’d like to be able to power the whole setup with one amplifier. It looks all of these outboard crossover options will require two amplifiers.
While a passive crossover likely could be designed to roll off the KEF's and cross to the additional drivers, likely this will not be optimal, cost effective or keep the singular amplifier happy. ( load )
Your best chance for success is either a single amplifier module with two channels for each speaker, so two such amplifier modules ( 4 channels total ) with DSP integrated, or an outboard DSP unit with plural amplifiers.
Again, thank you guys for all of the help. Much appreciated.. I apologize for getting ahead of myself. I still have a lot to learn.
I think I’ll probably buy something like the Parasound P6 that has bass management and just build those cabinets for the 2 10” woofers. This might yield me some good results without too much stress. A good first project.
One of you suggested going with 8” drivers instead of 10”s. My goal has always been to have a coaxial that has a large sound and can reach down into the 20s with authority with a little EQing, but I think my max volume for the two woofers is about 2.75 cubic feet if I want to keep the appearance I am aiming at. Am I better off giving a pair of 8’s some extra room in that enclosure, or. since I am bi-amping and am going to be using a little bit of bass boost am I better off with the 10’s?
I think I’ll probably buy something like the Parasound P6 that has bass management and just build those cabinets for the 2 10” woofers. This might yield me some good results without too much stress. A good first project.
One of you suggested going with 8” drivers instead of 10”s. My goal has always been to have a coaxial that has a large sound and can reach down into the 20s with authority with a little EQing, but I think my max volume for the two woofers is about 2.75 cubic feet if I want to keep the appearance I am aiming at. Am I better off giving a pair of 8’s some extra room in that enclosure, or. since I am bi-amping and am going to be using a little bit of bass boost am I better off with the 10’s?
"As I said, I have a very good idea of what I want to build. 6-1/2 coaxial in a sealed enclosure with 2 10inch woofers below it in a separate sealed enclosure."
In the Beginning...... the OP planned to use the 6.5" Tang Band W6-2313 coaxial speaker on top of two 10" woofers. A couple pic's for anyone following this thead for that first design... price range.
In the Beginning...... the OP planned to use the 6.5" Tang Band W6-2313 coaxial speaker on top of two 10" woofers. A couple pic's for anyone following this thead for that first design... price range.
Attachments
Recap: Building a sealed 2x10" cabinet to put underneath a pair of KEF LS50s.
So I was looking at the Parasound P6 and Emotiva XSP-1 because they have bass management, but the crossovers only go to 140hz and 250hz, respectively. Will this be high enough to crossover to the KEF LS50's?
If not I only see two other options, which is going with an outboard crossover like the Rolls SX45, or going with the passive Harrison Labs RCA crossovers that they sell on Parts Express. I could buy several different RCA crossovers and find one that works best.
Thoughts?
So I was looking at the Parasound P6 and Emotiva XSP-1 because they have bass management, but the crossovers only go to 140hz and 250hz, respectively. Will this be high enough to crossover to the KEF LS50's?
If not I only see two other options, which is going with an outboard crossover like the Rolls SX45, or going with the passive Harrison Labs RCA crossovers that they sell on Parts Express. I could buy several different RCA crossovers and find one that works best.
Thoughts?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.