By looking at the pics I am sure he used a 45 degree chamfer bit in a router. There is not another good way to get such a clean edge at the same angle all around...
Anyone know the dimensions of the baffle? Like distance between speakers, distance from top to bottom of board, angle against floor?
I am not understanding how the woofers can be 7mh for the 4 ohm load...isn't that like 90hz low pass?
...
Just read this thread. Yes, 2nd order 7mH and 20uF doesn't look like it ought to work. However if you were to shunt the coil with a resistor, maybe 9 to 12 ohms, it models very nicely. Is it possible there's a resistor hidden away somewhere in the Caladan woofer crossover, perhaps below the coil?
Last edited:
the tweeter is 3" from the top, the woofers are 12" from the top and bottom.Anyone know the dimensions of the baffle? Like distance between speakers, distance from top to bottom of board, angle against floor?
Just to illustrate what I said above, here's a sim of the Caladan woofers, based on the info earlier in the thread. Baffle is the correct size, inductor is 7mH, air core, 18 AWG (so 1.6 ohms DCR), cap is 20 uF.
Response is 6dB down at about 635 Hz, which is a long way short of a 1 kHz crossover.
Here's another sim with everything the same, except the coil is shunted by an 18 ohm resistor. Crossover point could be 1.5 kHz now. (Incidentally the dip around 1 kHz is a baffle effect. The little chamfer around the woofers may actually reduce this in practice.)
(PS In my earlier post, I guessed that a 9-12 ohm resistor would work well, but I was assuming that the coil had a DCR of about 0.6 ohms. However it looks as if the coil used in the Caladan actually has about 1.6 ohms DCR, so the shunt resistor probably needs to be bigger. )
Response is 6dB down at about 635 Hz, which is a long way short of a 1 kHz crossover.
Here's another sim with everything the same, except the coil is shunted by an 18 ohm resistor. Crossover point could be 1.5 kHz now. (Incidentally the dip around 1 kHz is a baffle effect. The little chamfer around the woofers may actually reduce this in practice.)
(PS In my earlier post, I guessed that a 9-12 ohm resistor would work well, but I was assuming that the coil had a DCR of about 0.6 ohms. However it looks as if the coil used in the Caladan actually has about 1.6 ohms DCR, so the shunt resistor probably needs to be bigger. )
Last edited:
Of course, my simulation is just a guess, and it doesn't take account of the real-world frequency behaviour of the drivers. Maybe the Beyma woofers have a rising response in the 600-1000 Hz band, which would offset the modelled droop? (The graph on the manufacturer's datasheet could be read that way, but the resolution on the vertical axis is so coarse it's hard to be sure.) The photos I've seen of the Caladan woofer crossover certainly don't show any sign of a shunting resistor.
I am not understanding how the woofers can be 7mh for the 4 ohm load...isn't that like 90hz low pass?
My general understanding/experience with passive filters for OB woofers would make this choice seem reasonable.
My experience has been with 3-way and 4-way OB projects (that turned out well) so the situation is different, but---
I look at it this way: take a 12” woofer like this in a 3 or 4ft3 ported box---it will have a fairly flat response from low frequencies up through it's major breakup, with no filter. But take the same woofer with no filter and mount it on an open baffle like the Caladan, and it will have a dramatically rising response from the lowest frequencies it produces up to something like 1000hz. So the 7mH coil on a 4ohm load just flattens out the rising response.
I think that I got my first education along these lines by reading an old (early 2000's) article by Martin King where he made a 2way OB using a 15” high-QTS woofer and (I think) a Fostex wide-band.
If I had two components to work with for the (combined 4ohm) woofers, and did not use simulation, I would try the large 7mH coil in series, and try a shunt cap that rolled off the highs to meet the tweeter at 1000hz---so nothing like a textbook 2nd order crossover.
For those just getting their passive-crossover-toes wet here: a 4ohm-loaded low pass filter uses a coil of half the value of a similar one with a 8ohm load. For high pass to tweeters, you could think of it as being the opposite; A 4ohm tweeter needs a 2x larger capacitor than does an 8ohm tweeter for the same high pass frequency. Of course there are many more factors to take into account, but this has been helpful for me to remember.
Maybe a smart person here could explain why the above is true.
Yes, and a bit to go that deep at 45 degrees is quite large in overall diameter.By looking at the pics I am sure he used a 45 degree chamfer bit in a router. There is not another good way to get such a clean edge at the same angle all around...
To make a 45 degree chamfer that is 1” deep, a bit like the Whiteside 2310 is needed, which is 2 5/8” in overall diameter ( including the ½” shaft and bearing). Best to have the power and bulk of a 3.25 HP router, or similar.
I heard the Caladans last week-end at the Pacific Audio Fest in Seattle. They were very nice, especially for the money. The bass was quite satisfying. I tried to find out if there was any additional bass boost being used, but was not able to obtain an answer. I am thinking that I asked the question poorly and maybe also that I was being too nosy??? I am now assuming that there was not electronic bass boost being used.
I built a 4way OB using two 12” woofers/side which are most likely less capable bass-wise than the Beyma woofers mentioned here, and the speakers have strong bass to about 40hz, and some output much lower, as one of the 12” drivers has an Fs of about 23hz. These speakers are made with vintage drivers---and are of really complex construction. I have thought about copying the shape and using a more straight-forward construction approach and readily available drivers---so this thread, and the show, got me thinking.
The one thought that I had about the Caladans was that though the mids were great for 12” drivers, adding the right 6” to 8” mid could be even better. Of course that could easily add $1500 retail to the price of the (commercially made) speakers (more if made by some others than Shaw, I am sure) due to more height, another cut out, much more complex crossover and more labor all around. All commercially successful speakers must be made to a price point. I was thinking as a DIY kind of guy, so the mid-driver would largely add labor and some cost for me.
The Caladans did music well, and they were demonstrated (to the best of my remembering) with the kind of music that the audiophiles/music lovers I know would likely listen to at home---which is not true of some of the rooms with very $$$$ $ystems in them.
Way too many speakers at the show had boxy or boomy bass, including some very expensive ones, which the Caladans did not. The Caladans seemed like a great deal compared to most other offerings at the show. The only better deal that I noticed were the Seattle based Vanatoo powered speakers, which sound great and have built-in amps/DSP and lots of connectivity choices.
I built a 4way OB using two 12” woofers/side which are most likely less capable bass-wise than the Beyma woofers mentioned here, and the speakers have strong bass to about 40hz, and some output much lower, as one of the 12” drivers has an Fs of about 23hz. These speakers are made with vintage drivers---and are of really complex construction. I have thought about copying the shape and using a more straight-forward construction approach and readily available drivers---so this thread, and the show, got me thinking.
The one thought that I had about the Caladans was that though the mids were great for 12” drivers, adding the right 6” to 8” mid could be even better. Of course that could easily add $1500 retail to the price of the (commercially made) speakers (more if made by some others than Shaw, I am sure) due to more height, another cut out, much more complex crossover and more labor all around. All commercially successful speakers must be made to a price point. I was thinking as a DIY kind of guy, so the mid-driver would largely add labor and some cost for me.
The Caladans did music well, and they were demonstrated (to the best of my remembering) with the kind of music that the audiophiles/music lovers I know would likely listen to at home---which is not true of some of the rooms with very $$$$ $ystems in them.
Way too many speakers at the show had boxy or boomy bass, including some very expensive ones, which the Caladans did not. The Caladans seemed like a great deal compared to most other offerings at the show. The only better deal that I noticed were the Seattle based Vanatoo powered speakers, which sound great and have built-in amps/DSP and lots of connectivity choices.
Please forgive me if anything following here is "preaching to the choir" . At any rate, I figure that there will be others who have not quite joined the "choir" yet.So,
1 - is this the right crossover?
2 - Have anybody tried it with good results?
3 - what XO have anybody tried with good results?
If everything about your speaker is close to exactly the same as another person's, the you could copy their crossover, theoretically. But likely as not, something significant will be different.
Do you have a measurement set up, like a USB mic + laptop + Room EQ Wizard? With measurement abilities, one can experiment by swapping parts in and out.
Cheaper parts can be used for getting basic frequency response established. Bi-polar electrolytic caps are way cheaper than film types. Ceramic-encased resistors are super cheap. Iron/ferrite core inductors with small wire gauges (with higher resistance) are way less expensive than hefty air-core inductor coils. You can get pretty close using these (ultimately inferior) parts. When you like how it sounds overall, then buy the good stuff to build your crossovers.
Getting good at making useful measurements in a living space is probably as much an art as science. I end up moving the mic around a lot to see if there are characteristics that repeat themselves, like peaks and dips, regardless of mic position. These are likely characteristics of the speaker, rather than room/speaker interactions. But there are exceptions, especially in the bass and lower mids, where floor bounce and room modes tend to swamp measurements. If you want to be sure your woofer is smooth through the desired range, put the mic at about 1/2 inch off the dust-cap to measure. Most hifi woofers are smooth (but not necessarily flat) at least through 500hz.
Howard, good information.
I do have the measurement mic and everything else. I probably should get a crossover design software to experiment and learn the basics.
I did built the speakers but I was not happy with the sound so I put them aside. So should start experimenting like you suggested.
Thank you for your advice.
I do have the measurement mic and everything else. I probably should get a crossover design software to experiment and learn the basics.
I did built the speakers but I was not happy with the sound so I put them aside. So should start experimenting like you suggested.
Thank you for your advice.
Chemarcelo,
If you find a good way to model dipoles, tells us about it. Not saying that such is not available, but it would be helpful to share it here. If you measure the drivers in the open baffle, you should be able to model the 1000hz crossover with some practical results in Xsim and Boxsim , which are free (pretty sure) from VISATON. There is a learning curve to saving measurements and importing them to the software, this takes some time to learn.
I tend to enjoy guessing at part values and then doing iterative swapping/testing. This must fit with my temperament, because I know it does probably take longer. I do learn about the possibilities first hand this way, which works for me, and sticks in my memory best. I have a more highly-educated friend who goes directly to simulations; if it was a job, he would be the one getting the promotions.
If you find a good way to model dipoles, tells us about it. Not saying that such is not available, but it would be helpful to share it here. If you measure the drivers in the open baffle, you should be able to model the 1000hz crossover with some practical results in Xsim and Boxsim , which are free (pretty sure) from VISATON. There is a learning curve to saving measurements and importing them to the software, this takes some time to learn.
I tend to enjoy guessing at part values and then doing iterative swapping/testing. This must fit with my temperament, because I know it does probably take longer. I do learn about the possibilities first hand this way, which works for me, and sticks in my memory best. I have a more highly-educated friend who goes directly to simulations; if it was a job, he would be the one getting the promotions.
I feel like i have to pitch in. I know that I do not have as much knowledge or experience as most of you have. But what I do know is that everything, every small detail, in audio, makes a difference. Even if the variations are small in a DIY build compared to the manufactured counterpart - they are still variations or differences made to the original creation. When to many corners are cut or when a few small changes are made they eventually adds up. Then it is not the same anymore.
Baffle thickness, chamfer depth (AND angle) and mounting the drivers the correct way seems like huge things to tamper with. Especially before a baseline is set. Of course you are allowed to change these things, but then it's just not the Caladans anymore. Maybe that's why some of you think they sound bad? Also, Clayton seems to recomend 200 hours of burn-in before they open up. That is a looong time for the impatient haha
If you think about it - Early on there was an option to order the Caladan with a hardwood OR a MDF baffle. According to Clayton the MDF didn't sound as good and they removed it. There must have been a significant audible difference since MDF probably would have been a lot cheaper to manufacture. Everything makes a difference.
I want to build my speakers as close to the originals as possible and i think we are on our way to figure out what that is supposed to be. I am sure that there are room for improvements since Clayton built these out of a business perspective. If i could i would buy the originals, but also where's the fun in that?
This is the proper crossover design. The only thing I'm not 100% sure about is the 1.25mh coil. Maybe someone can measure an original pair of Caladans?
Baffle thickness, chamfer depth (AND angle) and mounting the drivers the correct way seems like huge things to tamper with. Especially before a baseline is set. Of course you are allowed to change these things, but then it's just not the Caladans anymore. Maybe that's why some of you think they sound bad? Also, Clayton seems to recomend 200 hours of burn-in before they open up. That is a looong time for the impatient haha
If you think about it - Early on there was an option to order the Caladan with a hardwood OR a MDF baffle. According to Clayton the MDF didn't sound as good and they removed it. There must have been a significant audible difference since MDF probably would have been a lot cheaper to manufacture. Everything makes a difference.
I want to build my speakers as close to the originals as possible and i think we are on our way to figure out what that is supposed to be. I am sure that there are room for improvements since Clayton built these out of a business perspective. If i could i would buy the originals, but also where's the fun in that?
This is the proper crossover design. The only thing I'm not 100% sure about is the 1.25mh coil. Maybe someone can measure an original pair of Caladans?
HI ehyfatboy
Thanks for keeping us on track, I did not mean to de-rail things here.
I think that you are right that every aspect of a design matters---and some clearly matter more than others. In this context, if someone has built a speaker, and it does not sound right to them, they may want to do something about it. For my own hypothetical project, I would look at this as a challenge, and be pleased/proud if I successfully innovated/tweaked a design.
For those who are not so much in a place to problem-solve or innovate, but just want a nice pair of speakers to listen to, and be proud of the craftsmanship that they contributed, your focusing on the Caladan design specifics is clearly helpful.
In any case, I am especially glad to know about the Beyma woofers, and I already have a pair of the Dayton Tweeters.
Along with Clayton Shaw, La Dolce Audio was also at the recent (Seattle) Pacific Audio Fest show (ABX Audiophiles Partners Room). They have very good sounding open baffles, they sell the parts, and they provide designs. They tend to be more complicated, and the parts probably cost more, especially if you are not in the USA. But they are oriented towards helping others build their speakers.
Thanks for keeping us on track, I did not mean to de-rail things here.
I think that you are right that every aspect of a design matters---and some clearly matter more than others. In this context, if someone has built a speaker, and it does not sound right to them, they may want to do something about it. For my own hypothetical project, I would look at this as a challenge, and be pleased/proud if I successfully innovated/tweaked a design.
For those who are not so much in a place to problem-solve or innovate, but just want a nice pair of speakers to listen to, and be proud of the craftsmanship that they contributed, your focusing on the Caladan design specifics is clearly helpful.
In any case, I am especially glad to know about the Beyma woofers, and I already have a pair of the Dayton Tweeters.
Along with Clayton Shaw, La Dolce Audio was also at the recent (Seattle) Pacific Audio Fest show (ABX Audiophiles Partners Room). They have very good sounding open baffles, they sell the parts, and they provide designs. They tend to be more complicated, and the parts probably cost more, especially if you are not in the USA. But they are oriented towards helping others build their speakers.
No worries! My post wasn't aimed at anyone in paticular, no need to apologize! I just feel like I have to keep reminding myself to keep focus at times.
I have watched a couple of videos on YT from ABX and their open baffle build, very inspirational. However, for me, it is impossible to even consider a 15-inch (or 18") open baffle build. My room is too small and the wife appeal is almost at 0%...
Anyways... I have been thinking about, maybe, just maybe, make my caladans out of concrete. I know that it probably will introduce new set of problems. Thoughts? Do you guys think it sounds like a good idea or does it sound like I'm about to put two tombstones in my living room?
I have watched a couple of videos on YT from ABX and their open baffle build, very inspirational. However, for me, it is impossible to even consider a 15-inch (or 18") open baffle build. My room is too small and the wife appeal is almost at 0%...
Anyways... I have been thinking about, maybe, just maybe, make my caladans out of concrete. I know that it probably will introduce new set of problems. Thoughts? Do you guys think it sounds like a good idea or does it sound like I'm about to put two tombstones in my living room?
Tombstones? Gothic Audio Research Laboratories gets it's start!
When I was about 18 or 19 I built a 45 liter concrete enclosure for a JBL LE14a. It was the same volume as the plywood cabinets that I normally used. I just placed the 175 DLH horn on top in a stand. The front baffle was plywood. The stock crossover was either 1200hz or 1500hz, probably 2nd order acoustic slopes.
It was an experiment/project for a school class (“The Physics of Music”). It was not that greatly appreciated by the instructor, who thought that it was not academic enough. I remember the same instructor seemed jealous when I told him that I had just acquired a Dynaco integrated amp (good gear at the time) for cheap at the local flea-market. I think he said he wanted one for his kid.
From what I remember the concrete cabinet, which was about 37mm/1.5” thick, was neutral sounding, but lacked some of the liveliness of the homemade wood box; which was about 25mm/1” thick. Both were roughly the shape of a JBL L99 factory-made box. I ended up thinking that the wood box had a more appealing sound.
A concrete open baffle will sound different also, I would guess. At louder levels, open baffles can have very noticeable (to the touch) woofer-induced vibrations across the whole panel. The panel vibrations must contribute something to the timbre of the speaker. Also, the tweeter body (which moves the dome) might move less on a very heavy baffle, or it might just move at a lower vibrational rate? Not sure if this is audible---possibly modulating the pressure waves coming off the dome.
When I was about 18 or 19 I built a 45 liter concrete enclosure for a JBL LE14a. It was the same volume as the plywood cabinets that I normally used. I just placed the 175 DLH horn on top in a stand. The front baffle was plywood. The stock crossover was either 1200hz or 1500hz, probably 2nd order acoustic slopes.
It was an experiment/project for a school class (“The Physics of Music”). It was not that greatly appreciated by the instructor, who thought that it was not academic enough. I remember the same instructor seemed jealous when I told him that I had just acquired a Dynaco integrated amp (good gear at the time) for cheap at the local flea-market. I think he said he wanted one for his kid.
From what I remember the concrete cabinet, which was about 37mm/1.5” thick, was neutral sounding, but lacked some of the liveliness of the homemade wood box; which was about 25mm/1” thick. Both were roughly the shape of a JBL L99 factory-made box. I ended up thinking that the wood box had a more appealing sound.
A concrete open baffle will sound different also, I would guess. At louder levels, open baffles can have very noticeable (to the touch) woofer-induced vibrations across the whole panel. The panel vibrations must contribute something to the timbre of the speaker. Also, the tweeter body (which moves the dome) might move less on a very heavy baffle, or it might just move at a lower vibrational rate? Not sure if this is audible---possibly modulating the pressure waves coming off the dome.
Although the tweeter manufacturer claims it can be used from 1400 hz, having an Fs of 750 ohms requires at least two good octaves with a 12 db cutoff, so I assume the woofer has a 12 db cutoff and the tweeter is cut with a series RLC resonant cell, so you are completely wrong about the crossover frequency...also because a 20 mf in series, as you drew, would bring the cutoff close to the resonance frequency.
Regards
Roberto
Regards
Roberto
- Home
- Loudspeakers
- Multi-Way
- Caladan by Clayton Shaw DIY