Dimensions clarifications [message #87742] |
Sat, 07 April 2018 14:03 ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
joshua43214
Messages: 35 Registered: October 2016 Location: USA
|
Baron |
|
|
I was drawing up the 2Pi Towers, and noticed the note about changing the port diameter if 3/4" material is used. Looking closer, it was apparent that the dimensions on the plans are all external dimensions.
This caused me to double check the 6Pi dimensions and I found some inconsistencies. I went and reread the white paper on uniform directivity, which led to even more confusion over dimensions.
Someone asked in a previous thread if the horn dimensions are internal - and the answer was yes.
When I made my drawings, I worked from an internal mouth opening of 12"x24" tapering to 4.5"x4.5", then found a depth for the horn that averaged out the error between the 11" and 14.5" dimensions given in the instructions. The error here is pretty big, something like 3/8". The resulting horn depth is 10.25" not including the driver mount.
I referred to the white paper for help. The white paper is calling for 45x90 degrees, but the horn as drawn is closer to 40x87 degrees.
So what exactly are the critical dimensions here? The white paper says 90degrees by approximately 45degrees. 87 seems close enough, but 40 is pretty far off.
I am assuming the actual horn depth is not terribly critical, but the 4.5"x4.5" opening is important due to horn loading.
Would I be better off starting with tapers at 90 and 45ish from the 4.5"x4.5" opening and let the rest of the dimensions just follow so it fits into the cabinet?
The images below should help. The image is the space inside the horn flare itself.
This image is the flare using the dimensions as best as possible from the instructions.
This image uses flare angles of 45x90 and a 4.5"x4.5" opening.
Note the second image works out pretty nicely, except the flare mouth is almost a full inch larger than specified, though after fitting the cabinet will only be a bit oversized.
Thanks for you help,
-Josh
|
|
|
Re: Dimensions clarifications [message #87747 is a reply to message #87742] |
Sat, 07 April 2018 18:54 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18757 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
You're right that the horn angles are approximately 85° x 40°. And that's the way the tweeter waveguide is too. That's what we want - We want a horizontal pattern that's a smidge less than the expansion of the walls from the corner. So your top drawing describes best the dimensions of the midhorn.
The 45° figure is actually our target to exceed for the nulls. We want the nulls out at +/-20° to +/-25°, something like that. It's a reasonable goal that gives us plenty of vertical coverage. We definitely don't want nulls too close to the vertical centerline. So to have vertical coverage in the 40°-45° range is perfect. It keeps the HF attenuated at large vertical angles - to limit ceiling slap - and it gives us plenty of coverage, a nice tall "strata" of good sound. The nulls are set just outside the vertical pattern and sort of punctuate it.
Don't get too wrapped around the axle about differences of 5° or even 10°. For one thing the beamwidth "edge" we're talking about here is defined as a -6dB point, and outside this angle isn't a brick-wall sound void but rather a gradual falloff. And for another thing, the pattern is not perfectly constant, even from a waveguide or CD device. Directivity is closer to a constant than the pattern from an exponential or tractrix horn, but it varies with frequency, especially at the low end of the passband. So an 85° waveguide is very similar to a 90° waveguide, and we'll see much more variance than that from the environment.
For example, consider the difference in the radiation pattern from the midhorn when it is placed in open freespace compared with the radiation pattern when it's placed in the corner of the room, as designed. In freespace, the pattern is approximately 85° x 40° at 500Hz upwards, but below that it begins to waver. It actually narrows before widening way up. But in the corner, the walls constrain the pattern and make it much more constant. The pattern can never exceed 90° even at low frequencies where pattern control is lost, because the walls constrain the radiation angle. And the corner loading even limits the narrowing that occurs just before pattern control is lost.
|
|
|
|
|
|
|
|
|