The only gain is probably that the horn is bigger so directivity should be able to be extended a little lower in frequency. Vertical pattern should be better, but that is still TBD, it also might not be. The new horn is able to use (and seems to like using) the DNA360 driver, so the the whole speaker should be able to tolerate significantly more power (only relevent for damaging ears, IMO... the SmallSyns play plenty loud for me).
This is a very cool project Bill. It's a lot of fun watching your progress on this.
On a side note (though tangentially related to the above), I found this Tom Danley post regarding speaker efficiency and amplifier wattage on some other forum and was curious as to your thoughts about it.
Makes me consider if I'm wasting my time with high efficiency speakers and 30wpc tube amps!
I dunno, that might've been a characteristic of that particular Threshold amp. I used to do signal subtraction tests (record the input and output of an amp, and scale, shift and subtract them with a program I wrote called "Diffmaker", and playing musical tracks. With better amps (best were a modified Hafler amp and an Icepower module), the null was basically down to the background noise. If any of this small peak limiting were happening, it would've certainly shown up.
I also play with amp design (look for my Second Spring amp article in the Solid State section here, for instance) and have put clip detection circuits in some of them. The detectors can see even 10 microsecond peaks being clipped. It's surprising how loud typical recordings can be played without triggering the clip detector (probably thanks to peak limiting already done before the music is released -- I'm sure Tom's fireworks recording would clip it out at low average volume levels, though!)
I also play with amp design (look for my Second Spring amp article in the Solid State section here, for instance) and have put clip detection circuits in some of them. The detectors can see even 10 microsecond peaks being clipped. It's surprising how loud typical recordings can be played without triggering the clip detector (probably thanks to peak limiting already done before the music is released -- I'm sure Tom's fireworks recording would clip it out at low average volume levels, though!)
Sorry if this is common knowledge but TLDR whole SmallSyns thread.
How loud can they play?
Taking nothing from Tom's contribution, but it does smell like a very aggressive soft clipping/SOA protection circuit was playing part. In either case, seems he found a workable solution.
There are a lot of factors that influence how much power you need, but 30w with 95+ db/1m speakers in a reasonably sized will get loud, even 4m away from your speakers and still give you plenty of room. Tom might be one of those guys who really likes to crank it, whereas someone like me listens nearfield and barely needs a watt.
Apropos the title of the thread -- Bill, have you thought about making each print slightly over half the part (talking 3-4 mm) and then lapping (hairspray sandpaper onto a glass plate) the halves back to exactly half? This way you have some overprovisioning to manage the printing deformation and the lapping ensures you have a perfect mating surface. Great work as usual.
old boat building trick "kerfing in" - fixture the halves together and make a sawcut centered on the split
https://books.google.com/books?id=u...AEIMTAC#v=onepage&q="kerfing in" logs&f=false
hard to search for examples/explainations due to popularity of "kerfing" for bending wood
https://books.google.com/books?id=u...AEIMTAC#v=onepage&q="kerfing in" logs&f=false
hard to search for examples/explainations due to popularity of "kerfing" for bending wood
Last edited:
Bill, I know you chose the dna 360 over the Tymphany due to the better low end gain on this waveguide. But what about subjective sound quality? The Tymphany has a very refined and transparent high that puts it close to the best hifi domes. I have not heard the dna 360, but I have the fusion 8 which uses the dna 205, which is supposed to be more refined than the dna360.
I used to think the F8 dna205 highs sound good, but after using the Tymphany, it sounds grainy and harsh in comparison.
I'm using the Tymphany and PRV waveguide in a 2 way Ewave on your strong endorsement of it. Hehe. Appreciate your thoughts.
Sent from my X9009 using Tapatalk
That's not hard to do, but still, if you want to do a complex horn like this the 3D printer option certainly looks attractive. Kudos to Bill.
I once tried to do a 3D drawing of a complex device (the first step in any complex prototyping operation) and found that the learning curve was pretty steep. Good solids modeling software isn't cheap either so this is still a very serious DIY venture. Once you have the drawing file, getting the 3D printing done is pretty easy online.
Thanks, I get confused by initializations and familiarities on most sites, as I hang out in a variety of places.
Gedlee, do you ever have a booth at the trade shows? I like going to them and hearing peoples speakers, but don't remember hearing yours.
Hi Earl,
Yeah, the learning curve beyond simple boxes and cylinders and the like takes a fair bit of effort, but the bigger problem is some of the quirks of using software like this. I spent over a day finally getting the program to agree that a point of a spline curve that was exactly at the same coordinates as a point on a circle actually contacted each other! Also, as the model gets more complicated, some weird things can happen and it's easy to get into a pattern of what is basically trial-and-error to try to make the weirdness go away. The last big problem I had finally got 'fixed' by my exporting the file, closing the program, rebooting the computer, restarting the program and re-importing the program --- big ugly blob on the model disappeared finally
I'm sure it gets better as you learn what kind of things to avoid, though, a lot of companies are using the program (Autodesk Fusion 360).
As far as cost of powerful programs, things have changed in recent years. Software like Solidworks is still out of my league by a large margin, but Fusion360 (link) is free for any hobbyist or even for any business making less than $100k per year. That's what I did this model in. My "training" on it has consisted entirely of watching You-tube video tutorials (which probably explains some of why I kept getting stuck on silly-seeming things).
There is also a user support group at the Autodesk website for Fusion 360, and I was able to get help and even some fixes in a matter of hours in some cases.
rhapsodee,
I'm mostly going with the dna360 because it's response was smoother on the horn (even though the horn was designed originally for the Typhany driver), and the midrange porting on this horn still worked with the DNA360. It's actually the first time I've gotten good unity horn operation with the DNA360 (it's length from throat-to-diaphragm is longer than the Tymphany or the CDX1-1445, which can complicate things with a Unity horn).
As far as sound, I've worked with the DNA360 (a quasi-clone of the B&C DE250) and have liked how it sounded, its long inside length was my only real problem with it. I believe the DNA205 (a quasi-clone of B&C DE10) is actually a less capable driver than the DNA360. I had tried DE10 on a unity horn a few years ago, couldn't get it to work out (haven't tried it on this horn, though).
For a while Erich had another driver he called "FL-450", which used a neo magnet and had very smooth response at least on the SEOS horns. But I think he is phasing the FL-450 out, "too expensive". I couldn't talk him into letting me try it on these horns (but I may still try).
Random observations:
I've tried Sketchup, Fusion, Blender, Rhino and 123D. My favs are :
5) Fusion
4) Blender
3) Sketchup
2) Rhino
1) 123D
About my only real gripe with 123D is that it crashes quite a bit, and I haven't found any easy way to make threads. The interface of Fusion was maddening to me, even moreso than Blender (which is famously non-intuitive.) But get it while you can - 123D has been discontinued.
If you want to simplify or fix your models, check out Meshlab:
Polygon Reduction with Meshlab - Shapeways
Microsoft used to have an awesome tool for doing this, but they removed support for 'open' file formats
I've tried Sketchup, Fusion, Blender, Rhino and 123D. My favs are :
5) Fusion
4) Blender
3) Sketchup
2) Rhino
1) 123D
About my only real gripe with 123D is that it crashes quite a bit, and I haven't found any easy way to make threads. The interface of Fusion was maddening to me, even moreso than Blender (which is famously non-intuitive.) But get it while you can - 123D has been discontinued.
If you want to simplify or fix your models, check out Meshlab:
Polygon Reduction with Meshlab - Shapeways
Microsoft used to have an awesome tool for doing this, but they removed support for 'open' file formats
I don't know how to answer this kind of question --- what would an answer look like? Eleven?
For what it's worth, it uses two 6 inch woofers which have xmax of 5.25mm, so from that you could determine how far they can move linearly at some frequency. Each is rated at "200W" though it will be smashing into the magnet at low bass frequencies with that much sinewave power.
The tweeter is rated at 30W and 107dBSPL/1W (somewhere around 2kHz, less at higher frequency), so that would be about 120dB SPL 3 feet away. Though that still doesn't say how loud they'd play and still sound good.
Last edited: