M2 Synergy Horn

Last week I did some work with Voishvillo type phase plugs, and it seemed to work out well.

I was curious to see how that would work on a Synergy Horn, so I thought I'd take a crack at it.

I think there's a fairly good chance this won't work, but the only way to find out is to build it.

Here's some pics of the model.

Je8M694.png


3AgMxki.png


MTUDRV6.png


This waveguide looks admittedly bizarre, but there's a method to my madness. As detailed here*, it is possible to 'mask' off a great deal of a radiator and still have wide bandwidth. For instance, JBL 'masks off' about 50% of the diaphragm of their EON drivers.

It occurred to me that it might be possible to do the exact same thing in a waveguide, basically take a conventional wave guide and 'fill' half of it's volume.

An externally hosted image should be here but it was not working when we last tested it.

The final shape winds up very reminiscent of the JBL M2

I did something similar two years ago : http://www.diyaudio.com/forums/multi-way/267083-synergy-horn-3d-printing-entry-20.html#post4211484


* Car Audio | DiyMobileAudio.com | Car Stereo Forum - View Single Post - Imitation
 
  • Like
Reactions: 1 user
I think JBL is using that weird shape for a couple reasons:

1) it looks neat

2) The reason that narrow angle horns have higher output is because they're radiating into a smaller area. Conversely, wide angle horns have lower output because they're radiating into a wider angle. The M2 style waveguide combines both; it has a wide beamwidth, but it's on-axis SPL should be higher than a conventional wide angle waveguide because half of the M2 waveguide is 'filled.' IMHO the shape is reminiscent of the Voishvillo phase plug on the M2's driver (JBL D2430K)


In my application, I am after a similar goal. I'm trying to make a waveguide that's as shallow as humanly possible, but with a decent amount of gain. I am hoping that I can sacrifice some frequency response smoothness for higher efficiency.


By far, the main reason I'm using this shape is that it allows me to 'push' the Synergy midranges right into the 'knuckles' on the waveguide.

This thing is insanely small, about three liters in volume. It's approximately 8" x 6" x 4". The idea is to 'push' it back into a corner, as far as possible.
 
The entry holes are not neatly hidden near the corners anymore... hmmm...

Yep.

This is a very specialized solution, designed to do one thing only:

I want it to fit as tightly into a corner as possible.

zq9xczV.png


SxUFyEy.png


Cum3M5T.png


In these pics, I've added a 60x40 pyramid to illustrate how the midranges and the tweeters fit inside the space. Basically I curved the walls and added those 'knuckles' to get everything inside of that envelope.
 
  • Like
Reactions: 1 user
I need to give a quick 'shout out' to Monoprice for my new 3D printer. It's 2/3rds the price of my other printer*, so I'd expected it to suck. And I was honestly prepared for it to be junk, but I figured I'd rather risk $400 than spend $1000 at Frys.

But lo and behold, it's light years ahead of my other printer.

Whenever I printed something on my other printer, I'd say there was a 25% chance that the print would fail IF it was about half a liter in volume. But the bigger the volume got, the more likely it would fail. It reached a point where just about anything larger than a liter in size would fail. Generally the way it would fail is that the print head would slam into the print, and knock it right off the print bed.

That printer was LITERALLY making me insane, because I'd invest a whole weekend designing something in 3D, only to watch the print fail. Week in, week out, for the better part of the last two years.

But this new printer JUST WORKS.

If I had to speculate why, I believe the print head and the fans work better.

I'm going to feel dumb if the printer heads downhill from here, but so far, so good.

https://www.monoprice.com/Product?p_id=15711

157112.jpg


* my original 3D printer is Printrbot Simple Metal (now discontinued)
 
ZVAEQIL.png

The printed waveguide. Note the quality is light years ahead of my other 3D prints over the past couple years

nPVB2je.png

This waveguide simply doesn't work with dome tweeters. There was a distinct 'sawtooth' pattern in the frequency response. My work with Hornresp has taught me that the 'sawtooth' pattern in horns is caused when you have a driver on the horn that doesn't have enough BL.

Based on that, I slapped a BMS 4552 on there that I bought from Danley Sound Labs

That helped quite a bit, the peaks and dips were reduced by about 50%

I'll post polars and frequency response later

p5yUuHI.png

I used a soldering iron and a file to enlarge the throat from 19mm to 25mm (The waveguide was intended to be used with a SB19)

k6RoMcB.png

While unintentional, my new trike is the exact same color as this waveguide. Neat.
 
  • Like
Reactions: 1 users
TBH, I've never had issues with complex parts, on either printer.

Oddly enough, the most difficult thing to print (on my old printer at least) is anything flat.

IE, I could print something like this easily:

3DMaze.jpg


But printing a 6"x6"x 1/2" piece of plastic would fail.

This is because the plastic tends to 'curl up' on the edges. If you're printing something that's 2"x2"x1/2", the 'curl' might just be a fraction of an inch. But if you're printing something that's 6"x6"x1/2", the 'curl' will be magnified by the size, and the 'curl' could be 1/4"-1/2" at the edges.

2013-12-21-135127_med.jpeg

This pic shows what I mean

But the Monoprice printer seems to have solved that. I believe it's due to three things:

1) a heated bed, to fight the cooling that causes the curling
2) a beefy print head, to keep the plastic hot, all the way through the printing process
3) multiple fans, to cool the newly printed layer

Seems to be an art and a science, but this one is way better than my old one


One 'neat' thing that people may not realized about 3D printers, there's no real penalty for complexity. IE, you could print something uber complex, like a phase plug, and the print time is simply based on volume. A cube of the same size would take just as long, for the most part.
 
The complexity I was referring to involved the bottom level being narrower or smaller than higher levels, and the design necessities to keep it standing while printing the upper sections -- the printers I saw running regularly screwed up when doing something like a toy car shape (maybe the guys at the hobby club just didn't know what they were doing?)
 
John, what midranges is that molded for?
Seems like a lot of front volume on them.

How much others stuff has to be bought with that 3D printer to make is useful?

It's for an AuraSound whisper. And I agree, now that it's printed, the front chambers are definitely deeper than they need to be.

To make speakers in 3D, you need the following:

1) a 3D printer that can build a reasonably large volume. Mine is $399, builds 8"x8"x7"
2) Something that can make the model. I use 123D, costs $0
3) Once you make the model, you have to 'slice' it into "G Code" that the printer understands. Cura is the most popular, cost is $0
4) Filament is remarkably inexpensive. The filament for this print cost me about five bucks.
5) Time is very expensive. It's not unheard for me to spend 10-20 hours on a single design. Synergy Horns are painfully time consuming, because you have to check and double check that EVERYTHING is correct. 'Drilling' sixteen screw holes in 3D is surprisingly time consuming. (You have to make the hole, flatten out a spot for the head of the bolt, flare the entrance to the hole, etc etc etc) Obviously, this pales in comparison to doing it by hand, and it's way more fun to sit in front of the computer for twenty hours than work in the garage for 20 hours.
6) My old printer was tethered to a PC. This meant that I had to keep a PC attached, which is kind of a waste of space and power. My new printer simply prints off an SD card, which is way more elegant IMHO. You just save the file onto an SD card and plug it into the printer.
 
John,

Any comments (have you heard/read anything in comparison) about the fancier $600 model? It has 5x the resolution (which I assume probably takes 125x as long to print if you use the highest). And capability of putting plexiglas windows (to keep cats out of it... I'm sure our white cat will go nuts when he sees the little gizmos scooting around on the printer).

I think the trick to using a dome tweeter (low magnetic force) is to not compression load it -- just put the whole dome into the throat. No efficiency benefit then, but most of us don't use waveguides to maximize efficiency (power is CHEAP) but for radiation pattern control.
 
John,

Any comments (have you heard/read anything in comparison) about the fancier $600 model? It has 5x the resolution (which I assume probably takes 125x as long to print if you use the highest). And capability of putting plexiglas windows (to keep cats out of it... I'm sure our white cat will go nuts when he sees the little gizmos scooting around on the printer).

I think the trick to using a dome tweeter (low magnetic force) is to not compression load it -- just put the whole dome into the throat. No efficiency benefit then, but most of us don't use waveguides to maximize efficiency (power is CHEAP) but for radiation pattern control.

The waveguide pictured in this thread was printed with a layer height of 0.28mm and it took thirty two hours to print

If you went down to 0.1mm you'd be looking at a four day print

The $399 machine can go down to 0.1mm iirc

If I wanted a super smooth finish I'd probably print in PVC instead of PLA and simply rub the waveguide with acetone to make the surface smooth

Note that this won't make an audible difference of course; even at 20khz the wavefronts won't "see" a 1mm ridge

And note that there's a relationship between cost and strength:

For instance, imagine that you have two models:
Model one uses a 4mm nozzle and a print height of 0.28mm and takes 32 hours to print
Model two uses a 2mm nozzle and a print height of 0.14mm and takes 64 hours to print
Model one will be STRONGER than model two because the extruded plastic has four times the diameter. This was something I noticed when I used a 7mm nozzle; my models were ridiculously strong and they printed super-fast. On the downside, they didn't look as good and they used up a lot of PLA. (Which is admittedly quite cheap, but it's still annoying to blow ten bucks of PLA especially when prints fail all the time.)

I actually had the $600 model in my cart and went back and opted for the $400 model

If I'm not mistaken, the Monoprice is a clone of some printer out of the Czech Republic, so there's ten or twenty variations on this exact same printer out there. Most of the clones are for sale on Aliexpress.

I learned a lot about what to buy over at the Reddit 3D printing forum.
 
Last edited: