ZilchLab said:
1) No one is running a 12" at 800hz crossover.
2) Similar to a 12" round waveguide I imagine. Coincidence?
The wavelength of 800hz is 16.88" and the wavelength of 1600hz is 8.44". That means going from 800hz to 1600hz you would have to decrease driver spacing by 8.44" just to get the same lobing. Both of those lengths are reasonable for the Summa and your speaker respectively. So tell me, how much MORE than 8.44" did you gain by going to a small rectangular horn over something like the 15" round one in the Summa? Because it better be considerable to make running the horn past its cutoff seem like a good trade off in comparison.
catapult said:
That might be be a good way to use those speakers. On the floor, the lower null will possibly be lost in the carpeting.
However, I've always seen Summas on stands and I think that's the usual recommendation. On stands, I'd probably angle them downward, as I mentioned before.
Paul W said:
This is the whole thrust of my design approach. Apparently, it's the direction Earl has chosen when going asymmetrical too. I've written dozens of posts describing this very thing. The compromise I choose is to use a horn that loses pattern control in the vertical a little bit above cutoff, and to allow the nulls to punctuate that. This approach uses a smaller horn that allows closer CTC spacing.
To me, this is all very old news. I've studied this design philosophy for a long time, and I think I've optimized it very well. I'm not saying there's nothing new to learn or that no improvements can be made. But I am saying that I think round waveguides combine a step forward (improved HOM) and a step backward (poorer verticals). I personally would not go that direction. That doesn't mean I disregard the importance of reducing diffraction inside the horn. I have always thought horns with sharp edges sounded harsh, and I think the HOM matter may very well be the reason. I think Earl is right on that. No problem giving credit where credit is due.
Now, instead of one step forward and one step back, I suggest a low-diffraction asymmetrical horn / waveguide with mouth dimensions that are just large enough to maintain horizontal control down to crossover. This would not control the vertical pattern down to the crossover band, but it would allow tighter CTC spacing and would reduce the vertical coverage angle at HF. That has worked very well for me, and based on my experience, I think it will also work well as implemented with an elliptical PS waveguide.
Wayne Parham said:
I hope you're not suggesting that the stands in this pic you took raise the woofer to ear level.
augerpro said:
3) If so, a 12" waveguide won't cross that low, alas.
Note: The sim producing an 80° lobe width @ 992 is for a 13" C/C distance @ 800 Hz. That smells like 12" woofer + 12" waveguide to me.
augerpro said:
Déjà vu here. Because the 30° null angle is within the vertical pattern of the axisymmetric waveguide's vertical pattern generating the issues described above, but OUTSIDE that of a 90° x 50° rectangular one and therefore not exhibiting them.
Your 8.44" C/C distance works perfectly with the 6.5" high JBL waveguide, not "my" speaker per se, but subject of this thread, and a 10" woofer, crossed at 1.6 kHz, much as Wayne just prescribed....
ZilchLab said:
So what? It's not like the sound suddenly quits outside the nominal coverage. It slowly rolls off as you go off axis. It's a distinction without a difference they way I see it.
Hey, there's no perfect speaker. It's all about tradeoffs. I think way too much is being made of this particular tree of engineering design and we're losing sight of the forest.
catapult said:
Uhmm, so what the last three pages of this thread have been about, is what.
catapult said:
So, let's ignore the nulls and forget about the holes in the vertical off-axis response of the "superior" product offerings now disclosed here for the first time?
How 'bout we just call them "dips...." 😉
"Uhmm, so what the last three pages of this thread have been about, is what."
That makes it valid?
That makes it valid?
noah katz said:
I'd say that makes it worthy of consideration, and suggests Earl has some measurements to do, the ones YOU said you want to see, as consequence....
catapult said:
Actually, yes, when seated the top edge of the woofer was at ear level. The listener was below the centerline of the speaker. They were also very close, just a few feet back.
To help you gain perspective, the ports in that photo were approximately three feet up from the ground. As I said, listeners had to be very close. So because of the geometry particular to that loudspeaker, the listeners were more than likely sitting in a null at that show.
Wayne Parham said:
Well, if you say so. Those guys must be really tall and the listeners must be really short if that's the case. 😉
catapult said:
Well, it's hard to say really. But the three foot mark is somewhere between the top edge of the woofer and the bottom edge of the horn foam. The seated listener ear level was somewhere in that range, and the listeners were only a couple feet back.
It really isn't fair to get hung up on that particular show, or any others for that mattter. Shows aren't ideal conditions. The point is it is impoirtant to be careful when setting up a speaker that you know has a vertical null so close to the forward axis.
In this case, your idea of setting them directly on the ground might be a good idea. Or when setting up a home theater with these speakers mounted up higher on stands, be sure to angle them downward, even if the stands are relatively short. I would probably point them a few degrees downward of the target listening area so the woofer and tweeter were phased best.
There is another potential solution to the vertical behavior problem: use a steeper crossover. The dips off-axis are then confined to a smaller frequency region. I'm curious what you all think of this option.
The only downfall I see is the increased group delay at the crossover, but isn't this (within reason- say an LR8 or LR16 crossover) far less offensive than a huge trough in the amplitude response?
The only downfall I see is the increased group delay at the crossover, but isn't this (within reason- say an LR8 or LR16 crossover) far less offensive than a huge trough in the amplitude response?
Wayne,
"In this case, your idea of setting them directly on the ground might be a good idea. Or when setting up a home theater with these speakers mounted up higher on stands, be sure to angle them downward, even if the stands are relatively short. "
Is there some reason you won't answer the question as to whether angling them in accomplishes the same thing?
"In this case, your idea of setting them directly on the ground might be a good idea. Or when setting up a home theater with these speakers mounted up higher on stands, be sure to angle them downward, even if the stands are relatively short. "
Is there some reason you won't answer the question as to whether angling them in accomplishes the same thing?
Rybaudio said:
I thought about this too. What about a steep linear phase digital filter? No group delay, but steep filtering. This all adds cost and complexity, but may be a good option for the DIYer who builds new stuff all the time.
Rybaudio said:
breez said:
Took the words out of my mouth! Digital delay also means you can 'point' the lobe any where you like.....
Rybaudio said:
Crossover slope is important, and can help mitigate the nulls, yes. The crossover can also be used to set the position of the nulls.
noah katz said:
Because it doesn't. Horizontal orientation does nothing to affect the position of vertical lobes and nulls.
ZilchLab said:
The top or the bottom of it?
I just measured my own, and it doesn't seem that there's any problem of perspective there, though 36" would be slightly above woofer frame and vents, if my own legs are accurate.
I'm 6'4 but carry more of my height in my torso, so my legs should mostly be a pretty good indication of a 6' man's.
When I listened to the Summas in Earl's living room they were on the same stands. I was sitting on his couch and my ears were about the middle of the WG in level. I am 5'9".
badman said:
The top or the bottom of it?
Thanks!! I needed a good laugh this morning!
I really think that this discussion has degraded down to a fairly small point in a vastly more complex environment of "good sound". I am well aware of all the "solutions" and "heights" and "holes" and "lobes"- at least as savy on these points as anyone else here. The whole discussion is a non-issue to me, which is one reason that I don't participate. I've made my design criteria very clear and I don't consider vertical polar response to be important. If you don't agree with that then fine, design your own speakers around YOUR priorities and tradeoffs. But to banter on and on about how to "Fix" my designs or what kinds of vertical lobes they have is pointless. These designs aren't going to change - they work just fine the way they are.
If you want to discuss the needs and desires of vertical polar response thats fine, but using my designs as a "target" seems completely pointless to me as they are not designed with a very high priority placed on this particular aspect of a very complex problem.
- Status
- Not open for further replies.