3D printed waveguides

[augerpro]

I can incorporate any phase shield you can describe to me, or any other throat geometry for that matter.

Dan

I say copy SB more or less. A .47" diameter disc suspended not more than an 1/8" from the dome (had to eyeball that last measurement so this will probably require tweaking) Just noticed did not get the dome height of the TL so you know where to put the PS, I'll get it tomorrow. We'll probably want to try with and without the PS on both tweeters. The TL doesn't come with one in the first place, maybe it doesn't need it being so flat.

EDIT: of course if you had a scanner stroll down to a Kef dealer and scan the LS50

[ani_101]

i have one the transducer labs tweeters and am interested in the waveguide for it.

[Wolf]

Brandon- In my measurements, I tried with and without the felt dot on the attached grills that are available, and it did help the response as well as make it a bit less spitty, subjectively. It's a 3/8" diameter disc, and flattened out the 10-20k region. I just stuck it to the outside of the grill.

Later,
Wolf

[TN Allen]

Regarding the small oval WG I mentioned, the throat diameter is 1.25, and as it is oval, reducing the thickness to open it up, will produce an oval throat. I'll try to mill another next week. I'll need to do this quickly as I leave next week for 3 weeks. Other than the thickness, I can vary the dimensions. The thickness will be stacked MDF and a layer of Formica. It will be 1.385" deep, 4" high X 6" long, with a 1.3" diameter throat.

The image shows how the WG shapes are generated. The oval requires generating 4 arcs, and breaking the ellipse and throat circle into two pieces, which are used with the arcs to generate the surfaces.

The round guides are simpler. The mouth and throat circles are located on planes separated the guide thickness, a line is generated through the large circle and trimmed at the circumference. I added a point where that line would intersect the throat circle. An arc is then generated through the point and tangent to the end of the line that intersects the mouth circle. The nine that runs vertical through the centers of the circles is used as an axis around which a surface is generated using the arc.

The small hollow cylinders at the throats of the guides are necessary because I use a fillet cutter, a straight cutter with a radius on the tips. In order to form a full throat at the correct depth, I generate a tool path that ends at the bottom of the lowest edge of the cylinder. In other words, the milling penetrates a sacrificial substrate on the CNC table, leaving the correct throat circumference.

[augerpro]

Brandon- In my measurements, I tried with and without the felt dot on the attached grills that are available, and it did help the response as well as make it a bit less spitty, subjectively. It's a 3/8" diameter disc, and flattened out the 10-20k region. I just stuck it to the outside of the grill.

Later,
Wolf

Was that stock or in a waveguide?

[augerpro]

Regarding the small oval WG I mentioned

Can you make it 1" deep? The arc intersects the throat circle at 90 degrees? Can you change that?

[TN Allen]

Next time around, at present I have very little free time before leaving. Good bad or otherwise, It might be informative to try it as is.

[TN Allen]

I should have included that at the top and bottom, yes the angle is ~90°, but at the sides, the guide surface opens out much quicker.

Ultimately, listening to the effect of guides is probably as important as testing; are there any ideas for a practical way to include impressions based on listening in a comprehensive evaluation of the wgs?

[Wolf]

Was that stock or in a waveguide?

Stock.

[TN Allen]

I'd like to suggest making the test WGs that may be milled in MDF blanks so they can be easily mounted in a 1/2" thick test baffle, making the area around the guide an 8" diameter .75" deep rabbet. The rest of the guide blank can remain square. This would make it easy for a person to make a test baffle using a router on a circle jig to cut an 8" diameter hole accurately. If the baffle is 1/2", and the rabbet around the guide mouth surface .75" this would leave space to adjust for panel thicknesses that vary. Perhaps 4 1/4"-20 studs could be located in the back of the baffle and nuts used to adjust and hold the guide and baffle to be in the same plane. That is, there would be two nuts per corner stud.

3D printed guides could also be mounted in an 8" diameter MDF flange that could be mounted in the same baffle.

[augerpro]

The problem for me is that I no longer live in a house and all my tools are in storage. And I don't really know where I'd keep a test baffle. Let me chew on it.

I think you would need some sort of GTG to evaluate the subjective impressions. I've done it before with one of Dave P's waveguides and the Vifa NE19 tweeter, but that is only one person's opinion. After some designs are finalized it would be interesting to do at one of the regional GTG's.

[TN Allen]

Sounds like it may be a while before you test guides, which may be a good opportunity to further discuss a set of contours others may want to test. If you need a test baffle, I can mill a 2' square for you, that probably can be slipped behind a couch or chair for storage. 2' can probably be shipped, and is probably adequate for tweeter guides, though larger is probably preferable for midrange guides. It seems likely to me though that a methodical, uniform testing and data regime will provide the most useful data, and ultimately the most pleasant listening experience.

There is a Cheskey CD that might be useful for evaluating guides by listening. I'm not certain how this might work, given these would be evaluated in various listening environments, but it might provide at least a common set of tracks that were assembled to test different characteristics. https://www.amazon.com/Ultimate-Demo...rds=cheskey+cd Other tracks could be added to a test specific characteristics. This approach might be a start toward developing a way to produce a somewhat uniform method for reporting subjective results by various people in different locations.

Each track is prefaced with an explanation of the characteristic to be evaluated. I've listened to this CD once trying to evaluate a set of speakers with 20" diameter combined tweeter and midrange guides. It was an interesting, somewhat informative experience. This CD and a series of questions with answers that rate the perceived efficacy of the guide might be useful. I suppose before there can be a way to evaluate, there needs to be a consensus regarding what the effects of guides are and ought to be. Personally, I have only a vague sense of what guides do, and have designed all I have milled and listened to, making improvements intuitively. Various other people have listened to the results as I have gone along, reporting favorably and taking away the extras. This is a time consuming process, and I have been fortunate with access to the Haas and having already learned to use CAD. As I mentioned in an earlier post, I am willing to assist others with ideas they want to try.

[Neuro]

Really interested to see what comes of this.
I think I have the distinction of the largest 3D printed DIY horns to date. Not that they are overly large, 270mm diameter mouth. There are mounted to an SB65 on a spiral enclosure that I also printed. I would look forward to trying to print whatever you all come up with.
I might also be able to help with design depending upon what is desired and finding some time...

[TN Allen]

Interesting work, I'm interested in more design and printing information, as well as your sense of how well the combination works acoustically.

Any photos of the printing as it progressed would also be interesting as it looks as though there would have been temporary supports for the spiral. How long did these take to print? You must have had a fairly reliable printer.

[augerpro]

I keep eyeballing a Festool track saw so maybe I will build a small-ish baffle out of some 1/2" ply, something that can be atttached to my lazy susan measurement stand and put away behind a couch.