Starting a new thread for the build of the SMJ Audio Calliope MT speakers. I worked with Steve to get the design locked down. I went this path because I wanted something with a faster turn around for DIY. My front L/R channel speakers are on loan from a friend who now has space in his new home and would like his speakers back. I also believe in supporting small businesses, and the work Steve does speaks for itself. Therefore it made perfect sense to move forward and I would like to share the build process for these speakers.
SMJ Audio Calliope MT Build
Collapse
X
-
Well the work behind the scenes has begun. Parts and materials are being ordered and fighting with the shipping chaos that is know as Christmas.
Tweeters ..... woofers at the moment have been misplaced by FedEx.
Most of the crossover parts, a few are on back order.
Aluminum tubing for the stands sitting on a stack of dampening material.
Bamboo sheets arrived today. Glad the truck driver was willing to help move it into the shop.
Material for the spacers.
- Bottom
Comment
-
- Bottom
Comment
-
Immediate reaction:
Oh, be still my beating heartā¦.
Ok, a few deep breaths and Iāve regained my composure a bit, perhaps enough to comment.
First off svenarajala you have chosen your fabricator well. Steve has just delivered two pairs of more conventional Ā¼ cubic foot mini monitor boxes to me. I will post these to a new thread once I get them completed. For now suffice to take my word that Steve delivers to the highest quality.
Second comment is that Peerless DA25TZX00-08 tweeterās backside has some of the best industrial design I have seen in Hifi in recent memory. That heatsink and connector area is too pretty (yes, I am such a geek). I want to design it into something with an open back or a window to show it off. And as a bonus apparently some guy who posts here as Jon says it measures and sounds good.
Finally (for now) is the render above indicative of the planned result (of course it is) as I like where the Calliope is going if it is. These are just my opinions on the geeky industrial design details and purely a matter of taste so take or leave as you see fit. What really counts is what svenarajala and Steve like and want.
For years I have studied and built a few speaker designs based on the layered concept. As acknowledged in the A New Adventure thread Gauder Akustik https://gauderakustik.com/ helped inspire the system Steve is developing. Here is Gauderās BERLINA RC 3. As an aside how can Gauder Akustik use such an ugly stand? In contrast Steveās is an integrated part of the design.
Another company that uses a layered concept is Eventus Audio http://www.eventusaudio.com/, this is their iO.
What I prefer in the Eventus and the Calliope render above for svenarajala's is the more subtle lines between the layers compared to the Calliope prototype here.
From a manufacturability standpoint the larger edges radius I expect is better able to take paint finish and hides misalignments down in the tolerances of a few 0.001ā. I do wonder, did you do all that rounding over on CNC or manually at the router table? That would be a lot of time feeding parts across the router table.
I know my preference for tighter transiting between layers is down to Sonus Faber, like these Amati Guarneri. Now on these the layers are blocks of walnut and I expect on some models strips of veneer. My judgement is the Calliope emphasizes performance first looks second while SF does the opposite.
I freely admit one can also argue the more rounded look of the layers on the Calliope prototype celebrates the rare and unique technics used to achieve the desired result of styling + performance + manufacturability. Really looking forward to following the progress of svenarajalaās set.- Bottom
Comment
-
Thanks for the kind words Norman.
With regard to the round overs .... they were all done on the router table. Since it's a two sided operation it seemed to me that it would take less time than having to deal with that on the CNC. When I cut the ribs it's all from one side and their done. Having to set them up again to just do a round over did not make sense to me without making a very accurate holding fixture. Of course I may look at that again in the future.
As to the round overs they can be reduced down to nothing more than just breaking the edge. Generally that's left up to the customer unless they want to me to take care of it and in this case Sven is going for what's on the originals.
The spacers also contribute to the over all look. I went with 3/16" material and with gasket material top and bottom it comes out right at .25". I could see potentially going with an 1/8" spacer, but one has to keep in mind a certain amount of structural integrity. One could look at other materials, though prices go up the crazier you go.- Bottom
Comment
-
Letās look at it this way... a platform for your options
Thanks for sharing the interesting comments Norm- all very valid.
I would like to throw out a few ideas though, that might put them in a somewhat different light.
In one sense, I expect Steve is flattered- you’re looking at this like a finished product, and comparing it to some very capable and established manufacturers.
However, what was intended is a robust development platform... while Steve can take this all the way to a complete turn-key developed speaker if one desires and can afford to hire his labor, that’s not really the basic intent. This is about DIY, and though the “Y” comes last in that version of the acronym, we could justifiably coin a different version, which would go, YDI, for “You Do It!”
So there are many design choices and options you can pursue with this concept. While the least damage is done to the pocket book with MDF or HDF, BB ply, LBL ply, and other materials are also viable options, because this assembly technique is far more robust mechanically than the kinds of Translam construction many besides Steve have used on HT Guide (Cochinada, I’m talking about you....)
MDF does not have very good bending strength, and it’s resistance to tearing is mediocre, like BB ply, but it’s compression strength is quite good, and that’s how it’s used in this design....
BB ply has very good bending strength and fairly good compression strength, but tearing and de-lamination are clear problems to be dealt with - problem solved.
LBL ply has much better overall material properties- ones that make it a very good choice for the end caps, but a bit pricey for the main laminations unless you have deep pockets like Evil Twin, with some of the power of the Imperial treasury behind you...
As Steve notes, the separator layers can be reduced in thickness- if you’re willing to commit to not disassembling, it can be eliminated and you can glue as you assemble, then of course you need to have a good random orbital sander for leveling everything afterwards, somewhat like those previous translam builds, except that the material cut precision probably reduces that work by an order of magnitude.
So, think of what you’ve seen from Steve so far as a jumping off point for your imagination, but remember you might need to schedule a bit of a talk with your banker, too, if your imagination gets the best of you- there’s no changing the physics or finances, it’s just an opportunity for experimentation and results not often open for the YDI or DIY guys like us.the AudioWorx
Natalie P
M8ta
Modula Neo DCC
Modula MT XE
Modula Xtreme
Isiris
Wavecor Ardent
SMJ
Minerva Monitor
Calliope
Ardent D
In Development...
Isiris Mk II updates- in final test stage!
Obi-Wan
Saint-Saƫns Symphonique/AKA SMJ-40
Modula PWB
Calliope CC Supreme
Natalie P Ultra
Natalie P Supreme
Janus BP1 Sub
Resistance is not futile, it is Volts divided by Amperes...
Just ask Mr. Ohm....- Bottom
Comment
-
- Bottom
Comment
-
- Bottom
Comment
-
The spacers are like perfect little cookies that just got cut out. The CNC takes the level of precision to the next level and I enjoy seeing all the cool things that can be created with that device. On another note I got most of the supplies in the list you sent me ordered.Painter in training- Bottom
Comment
-
There only perfect little cookies when the material doesn't move on you and get chewed up by the bit ..... that was yesterday. :cry: A different holding setup including double stick tape was implemented today for better baking results. :W- Bottom
Comment
-
- Bottom
Comment
-
Slow work continues watching the glue dry. This time it's with mdf .... too bad this good stuff doesn't come at 1.5" already. At about the 60% point for the required slabs.
[ATTACH=CONFIG]29916[/ATTACH]
Note in the background the size of the glue bottle ..... yep, that's gallon sized.
[ATTACH=CONFIG]29917[/ATTACH]
Ah, this reminds me of the good old days during the 2009 recession, when I was working on clamping up sub panels on the first Ardents...
You know, back when I used to have weekends off, and could take vacation time...
I have over two months of accrued vacation time at this point... fortunately I don't hit the use it or lose it ceiling until the end of the year, when it will be about 3 months if things keep going like they have been...the AudioWorx
Natalie P
M8ta
Modula Neo DCC
Modula MT XE
Modula Xtreme
Isiris
Wavecor Ardent
SMJ
Minerva Monitor
Calliope
Ardent D
In Development...
Isiris Mk II updates- in final test stage!
Obi-Wan
Saint-Saƫns Symphonique/AKA SMJ-40
Modula PWB
Calliope CC Supreme
Natalie P Ultra
Natalie P Supreme
Janus BP1 Sub
Resistance is not futile, it is Volts divided by Amperes...
Just ask Mr. Ohm....- Bottom
Comment
-
- Bottom
Comment
-
I wonder why the C2C spacing is so large with this design. You could easily get the two drivers much closer together and improve the vertical off axis performance.What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
5th Element, otherwise known as Matt.
Now with website. www.5een.co.uk Still under construction.- Bottom
Comment
-
Hey Matt ..... I could so be a smart ass here but I won't. This was touched on a little in the New Adventure post somewhere around entry 173.- Bottom
Comment
-
I was wondering if there was a technical reason and reading the thread seems not. It mostly seems that the C2C has been chosen, at least I'm assuming, for aesthetics. Based on the rather tall cabinet a wider spacing does look better.What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
5th Element, otherwise known as Matt.
Now with website. www.5een.co.uk Still under construction.- Bottom
Comment
-
Steve- Bottom
Comment
-
I find the discussion over drivers' center to center (C2C) distances interesting. Obviously the question arises from the rule of thumb to minimize C2C. However. In engineering and industrial design rules are made to be broken. One of the champions of this is Avalon Acoustics who have inspired some of the top designs here on Mission Possible especially the faceted enclosure. Avalon's two ways seem rather prone to larger that typical C2C distances.
- Bottom
Comment
-
Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
Let's review the numbers I ran...
Without seeing your Unibox model, it's hard to comment- I use both Unibox and VituixCAD, and generally they agree very closely if the input information is the same...
Now, for a target Qtb of 0.707 (the classic maximally flat butterworts alignment) the sealed enclosure volume is 16.8L, with an Fb of 63.3 Hz and F3 of 61.7Hz, and F6 of 49Hz. That's a pretty typical alignment that might be desired for a sealed 5.5" alignment, so as to limit LF excursion and not overpower the driver. Kind of conservative for a 7.5" driver.
In a 28L ported alignment, with Fb of 33Hz, the F3 is 43.6Hz, and the F6 is 34.6 Hz, near the box Fb. This might be considered a "suboptimal" design as regards low frequency extension and output, but it's a decent balance between size and extension, and it keeps the driver loaded decently fairly low, so over excursion isn't likely. Pretty much a classic QB3 alignment, with slow roll off and smaller enclosure.
An alignment for the MW19P optimized for LF extension and output would be 50L tuned to around 30Hz; then you have an F3 of 29Hz, and an F6 of 25.6 Hz. Yes, TWO of these woofers would take an enclosure for the same tuning about the same size as the Isis clone.
And at 50L, you have some genuine "Monkey Coffins". That's bigger than the Wavecor Ardent LF enclosure volume. But it's got a midrange enclosure in there, too.
So, imagine a Wavecor Ardent size cabinet with 1 - 7.5" woofer and a tweeter... that would be the "optimized" Calliope for LF output. Just like the Avalon Eclipses in days of yore using 8" Eon woofers...
These were back in the days when Charlie Hansen would introduce me as the grandfather of Avalon...
I'm pretty sure we/Steve could do something like this for the existing design if they potential constructor can handle the size... with this panel width, it's pretty much a minor tweak for the crossover.the AudioWorx
Natalie P
M8ta
Modula Neo DCC
Modula MT XE
Modula Xtreme
Isiris
Wavecor Ardent
SMJ
Minerva Monitor
Calliope
Ardent D
In Development...
Isiris Mk II updates- in final test stage!
Obi-Wan
Saint-Saƫns Symphonique/AKA SMJ-40
Modula PWB
Calliope CC Supreme
Natalie P Ultra
Natalie P Supreme
Janus BP1 Sub
Resistance is not futile, it is Volts divided by Amperes...
Just ask Mr. Ohm....- Bottom
Comment
-
Without seeing your Unibox model, it's hard to comment- I use both Unibox and VituixCAD, and generally they agree very closely if the input information is the same...
Now, for a target Qtb of 0.707 (the classic maximally flat butterworts alignment) the sealed enclosure volume is 16.8L, with an Fb of 63.3 Hz and F3 of 61.7Hz, and F6 of 49Hz. That's a pretty typical alignment that might be desired for a sealed 5.5" alignment, so as to limit LF excursion and not overpower the driver. Kind of conservative for a 7.5" driver.
Fs 32.00
Re 3.40
Qms 4.11
Qes 0.30
Sd 158.0
Vas 48.0
Xmax 6.80
Le 0.14
With 1 ohm series resistance and stuffing it to the gills:
Interestingly, I can match the SW223 factory alignments pretty closely, so I wonder if it's a problem with the factory data itself. In either case, it seems that I need to crack open VituixCAD again. It's been over a year since I last touched it, and I did my last few designs/modeling scenarios in PCD (I spend way too much time in Excel....). I have had problems professionally with issues like rounding errors at the 14th decimal place and different results for exponent/root calculations when mixing native 32-bit and 64-bit files, so I can't eliminate a platform problem just yet, but I'd prefer to keep it in the realm of "operator error" as long as possible.
I'm pretty sure we/Steve could do something like this for the existing design if they potential constructor can handle the size... with this panel width, it's pretty much a minor tweak for the crossover.Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
I thought that I had reached a point from planning to building, but recent discoveries/events have me back looking at options. I am focusing on Hoffman's Iron Law to see how more power and more cone area can replace box volume (hence my acquisition of 8 SW223s in various tranches). This is where something like the Dutch & Dutch 8c or a variant on the latest Ardent D might be more fitting to my needs. As you know, California real estate prices put a premium on floor area, and sub-optimal is just my starting point for almost everything...
One understands these realities of life quite well... one might say that the Wavecor Ardent was a specific effort to straddle the difference between the original Ardent concept and the Isis, increasing LF capability and playback volume while minimizing changes in the enclosure size...DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
the AudioWorx
Natalie P
M8ta
Modula Neo DCC
Modula MT XE
Modula Xtreme
Isiris
Wavecor Ardent
SMJ
Minerva Monitor
Calliope
Ardent D
In Development...
Isiris Mk II updates- in final test stage!
Obi-Wan
Saint-Saƫns Symphonique/AKA SMJ-40
Modula PWB
Calliope CC Supreme
Natalie P Ultra
Natalie P Supreme
Janus BP1 Sub
Resistance is not futile, it is Volts divided by Amperes...
Just ask Mr. Ohm....- Bottom
Comment
-
Thanks! I was looking at the 4-ohm part, but this also helped me calibrate the factory data on the MW19P-8. That Qms number is, um, a touch higher than the spec sheet (not a huge effect, but...).Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
Hmmm.... I'm getting half of your volume with no fill and minimal leaks using the factory data.
Fs 32.00
Re 3.40
Qms 4.11
Qes 0.30
Sd 158.0
Vas 48.0
Xmax 6.80
Le 0.14
[ATTACH=CONFIG]29921[/ATTACH]
With 1 ohm series resistance and stuffing it to the gills:
[ATTACH=CONFIG]29922[/ATTACH]
Interestingly, I can match the SW223 factory alignments pretty closely, so I wonder if it's a problem with the factory data itself. In either case, it seems that I need to crack open VituixCAD again. It's been over a year since I last touched it, and I did my last few designs/modeling scenarios in PCD (I spend way too much time in Excel....). I have had problems professionally with issues like rounding errors at the 14th decimal place and different results for exponent/root calculations when mixing native 32-bit and 64-bit files, so I can't eliminate a platform problem just yet, but I'd prefer to keep it in the realm of "operator error" as long as possible.
I thought that I had reached a point from planning to building, but recent discoveries/events have me back looking at options. I am focusing on Hoffman's Iron Law to see how more power and more cone area can replace box volume (hence my acquisition of 8 SW223s in various tranches). This is where something like the Dutch & Dutch 8c or a variant on the latest Ardent D might be more fitting to my needs. As you know, California real estate prices put a premium on floor area, and sub-optimal is just my starting point for almost everything...- Bottom
Comment
-
As far as difference, from what I can tell, most 4ohm parts tend to require somewhere between a little to a lot less cabinet volume, depending upon how the manufacturer differentiates between 4ohm and 8ohm nominal parts. In some cases, it is literally 2x between the two. Of course, a lot of 4ohm parts have an EBP that put them squarely in the "vented" category, as well. However, I'm looking just at performance down to ~80Hz (smooth integration with a large handful of subwoofer drivers).Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
- Bottom
Comment
-
Thanks Heli-Tim ...... It's actually 3/8" threaded rod that holds it all together. With the spacers being slightly smaller there is no need for super critical alignment.- Bottom
Comment
-
When you have two identical drivers, with one 4 ohm and the other 8, the 4 ohm variant typically has lower Qts lending itself to smaller cabinets.
Box size is always a peeve of mine. I hate big boxes but love what they can do for bass extension and efficiency.
I'm always wrestling with trying to squeeze as much potential out of a driver in as little space as possible. It's rare that I manage to get something that optimises well into a ported box as usually I need cabinet to be smaller than that would allow.What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
5th Element, otherwise known as Matt.
Now with website. www.5een.co.uk Still under construction.- Bottom
Comment
-
When you have two identical drivers, with one 4 ohm and the other 8, the 4 ohm variant typically has lower Qts lending itself to smaller cabinets.
Box size is always a peeve of mine. I hate big boxes but love what they can do for bass extension and efficiency.
I'm always wrestling with trying to squeeze as much potential out of a driver in as little space as possible. It's rare that I manage to get something that optimises well into a ported box as usually I need cabinet to be smaller than that would allow.Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
This is the money quote, especially if you intend to handle a wide range of movies with extreme special effects, as high output to below 20Hz can be necessary.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Exactly I love subs. Inefficient things that will take power but go low and loud enough with it in small box.
Then you can keep the mains and the subs smaller whilst putting the subs somewhere inconspicuous.
I guess that sums up the Wavecor Ardents bass section nicely and I wasn't even trying. Easy to do these days when watts are relatively cheap. Still, I enjoy high efficiency stuff for the other drivers.What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
5th Element, otherwise known as Matt.
Now with website. www.5een.co.uk Still under construction.- Bottom
Comment
-
- Bottom
Comment
-
Very nice stack of material and will be very interesting to see how the end result looks.
What is the approximate speed you can use for the bamboo - ie how long it takes to cut out one of those pieces? I guess for other layers it will take about 2x as much as there will be the inner cut as well.- Bottom
Comment
-
Very nice stack of material and will be very interesting to see how the end result looks.
What is the approximate speed you can use for the bamboo - ie how long it takes to cut out one of those pieces? I guess for other layers it will take about 2x as much as there will be the inner cut as well.
I'm still getting a feel for speeds and feeds in general, but I'm finding the manufacturers recommendations for the bits, work pretty well. If you look at the second picture, the pieces to the left are the cabinet tops, which I set up to cut two of at the same time. For the insert holes I'm using a 1/4" bit running at 225 IPM. It takes 1.5 minutes to cut the holes. To cut the outer profiles, I'm using a 3/8" diameter bit running at 200 IPM taking 5 passes to cut the full 1.5" thickness. Cut time is a little under 4 1/2 minutes. Though I did program a couple of pauses in there to remove chip build up in the cuts.
Steve- Bottom
Comment
-
Slow work takes time, even with CNC! But the accuracy and consistency of results is another key benefit!the AudioWorx
Natalie P
M8ta
Modula Neo DCC
Modula MT XE
Modula Xtreme
Isiris
Wavecor Ardent
SMJ
Minerva Monitor
Calliope
Ardent D
In Development...
Isiris Mk II updates- in final test stage!
Obi-Wan
Saint-Saƫns Symphonique/AKA SMJ-40
Modula PWB
Calliope CC Supreme
Natalie P Ultra
Natalie P Supreme
Janus BP1 Sub
Resistance is not futile, it is Volts divided by Amperes...
Just ask Mr. Ohm....- Bottom
Comment
-
Those speeds are pretty good and until Steve has a need to push through tens and tens of these per day, using moderate speeds makes total sense.. I've been even more careful sometimes, but in milling case too slow is also bad as the bit starts to heat and will be ruined in no time (been there and done that too).
So I find this CNC feed and speed topic a very interesting multi variable challenge. Goal is to find some sort of optimum... one has to force himself to be very systematic and only change one (or max 2) parameters at each iteration and mark down the outcome... otherwise it's just shooting in dark.- Bottom
Comment
Comment