You know, Art - you're opening up one hell of a can of worms with this. Been thinking about a new more modern replacement for an aging stack of lab honrs for a while now. Once I started looking at extenders instead of simply stacking more boxes, all my hornresp sims tell me the 21" B&C is a better choice than the 18's. Limiting to just 800 liters a box the (cheaper) 18's were preferred - and I have some on hand.
Damned and determined to make me spend $700 a driver.... just to try it out.
Damned and determined to make me spend $700 a driver.... just to try it out.
I built a 21" keystone sub, I do have the drawings, it's for a bicycle soundsystem, so it's playing into the ground below the cargo bike, this works like a weightless extension. Some directivity might be lost, but sub is pretty omnidirectional and a gain of around 3db is pretty nice, plus the fact that it might even go a bit deeper.
I'll post pictures soon
I'll post pictures soon
Rowan,
Probably would, in front of a sealed front loaded 4 x10" the level seemed to be about +3 dB, but have no actual measurements of that 2004 era worth posting.
Art
Great, my mid amplifier is occasionally limiting when I'm running the system, rather than replacing it I thought these wave extenders would provide the gain I require. I'll try and build some in the month or so. I measure both the before and after response and post the result here for people.
I know you had the extenders at 45degrees left and right and 15 inch in depth, but the top and bottom planes sat at the same height as the top and bottom of the speaker (no flare), if I'm sitting my midbass on the ground (to eliminate floor bounce) would it be best to put the top and bottom panels at 45degrees also, so there is a flare left-right-top-bottom??
Regards,
Rowan
Rowan,
Left-right-top-bottom with a throat the size (or smaller) than the cone would be a bit more effective than the simple wave guides I used.
I'm thinking about building wave guides to maximize the performance of my two Keystones recently built.
I like center stacking subs to avoid cancellation issues compared to LR stacked subs, but I'm thinking if the dispersion of a pair of centerstacked Keystones with a single waveguide for the two Keystones would be too narrow if they are located between two tops distanced 3-4 meters from each other with a 90° horizontal dispersion, would the sub be weak at the edge of the top cabinets horizontal coverage angle?
Would it be preferable to separate the subs and use separate extenders(and use them as stands for the tops) or would this yield undesierable cancellations between the subs? I figured if they are equipped with extenders the directivity might be high enough for cancellations to not be that prominent. Or would the stacks have to be distanced more than 3-4 meters from each other in this case?
I like center stacking subs to avoid cancellation issues compared to LR stacked subs, but I'm thinking if the dispersion of a pair of centerstacked Keystones with a single waveguide for the two Keystones would be too narrow if they are located between two tops distanced 3-4 meters from each other with a 90° horizontal dispersion, would the sub be weak at the edge of the top cabinets horizontal coverage angle?
Would it be preferable to separate the subs and use separate extenders(and use them as stands for the tops) or would this yield undesierable cancellations between the subs? I figured if they are equipped with extenders the directivity might be high enough for cancellations to not be that prominent. Or would the stacks have to be distanced more than 3-4 meters from each other in this case?
Last edited:
1)The waveguides I used increased level by about 3 dB, not much impact on bass directivity.
2)Locating the tops above the subs gives better integration in the crossover region regardless of the horizontal spacing, and using the subs for stands is more stable than an unweighted speaker stand, so is my preference. In any array, location of speakers more than 1/4 wavelength distant from each other will cause off-axis cancellations.
3)The distance the subs are apart (and distance to boundaries) determines the cancellation/reinforcement radiation pattern.
So after having used our horn extenders for a few years, a couple of things to watch out for.
Make sure you have REALLY good bracing. Preferably metal or metal edged wood with decent fixing systems. Our system is getting old now and the constant rattling / shaking takes it toll on all the fixtures! We only have vertical bracing right now but will be adding some horizontal at some point ... because of..
Watch out for other resonant frequencies. So recently I have noticed that putting a live band in between our two stacks can kick up some odd resonant freqs in the wave guides. Especially low mids ? Seems that having stage monitors close enough to the stacks can kick them off. With our system the bigger horn extension can point on stage or off stage..I'm going to experiment either way to see which is better.
Beer and dust seem naturally gravitate towards the extensions... make sure you have a brush and cloth !Also small children seem to love climbing in them...take earplugs to hand out !
Next plan is some keystones with extenders as well ... our 4 x SS15's run out of steam way before our Unity horns. Time for something new.
.p.
Make sure you have REALLY good bracing. Preferably metal or metal edged wood with decent fixing systems. Our system is getting old now and the constant rattling / shaking takes it toll on all the fixtures! We only have vertical bracing right now but will be adding some horizontal at some point ... because of..
Watch out for other resonant frequencies. So recently I have noticed that putting a live band in between our two stacks can kick up some odd resonant freqs in the wave guides. Especially low mids ? Seems that having stage monitors close enough to the stacks can kick them off. With our system the bigger horn extension can point on stage or off stage..I'm going to experiment either way to see which is better.
Beer and dust seem naturally gravitate towards the extensions... make sure you have a brush and cloth !Also small children seem to love climbing in them...take earplugs to hand out !
Next plan is some keystones with extenders as well ... our 4 x SS15's run out of steam way before our Unity horns. Time for something new.
.p.
Osse,
Sorry I missed your question, been quite busy with cross country move-in details for the last couple months.
The increased frontal area of two Keystones next to each other raises frontal gain by about 1 dB compared to a single, so adding the equivalent area that I used for one cabinet for two should end up with about a 2 dB gain for the pair compared to the 3 dB gain noted for the waveguide doubling of frontal area.
Art
Quote:
Originally Posted by weltersys
Sam,
To bring you back to reality may require you do some simulations, though one thing I have brought to David McBean's attention is that Hornresp does no account for the forward directivity a simple "barn door" provides- when you simulate an additional frontal area, IIRC there is no gain change, while in reality there is a fairly large measured increase.
1. No, the 15" deep 90 degree wave-guides that I initially used with FLH, then bass-reflex cabinets, and finally with the Keystones, did not extend the low frequency response on any of them, but did increase level on all of them.
2. The triangle depth, not the height, does increase path length.
3. Given a specific cabinet width, narrowing the inter-cabinet angle does increase the depth of the "V". Without simulation and/or testing the specific cabinets and angle of the "V" plate, as well as the specifics of how well braced and gasketed said device is, one can only speculate as to the results.
If you do do any simulation and testing, please share your results in the wave-guide thread:
Horn Extender/Wave-guide for TH
Thanks,
Art
1. I understand, and I guess in theory only a horn's response could be successfully extended.
2. I think we're on the same page and it's just semantics.
3. Now that I think about it, I think the angle is probably critical. I did a bit of research and from what I can gather horn loading is lost if too wide. You would still get an increase in output but then the extension is only acting as a reflector and not a horn. It’s getting beyond my understanding but I bet there is a way to calculate the best angle/flare rate. In my mind, the angle would be calculated as if one were to add another fold the KS. A quick look at the KS and it appears this angle would be more narrow than wide since the PL is nearly 90deg near exit. That might advantageous since the path length will be even longer and v plate would be smaller. I found this thread really informative and includes REW plots proving the extension does lower FR. Horn Extender/Wave-guide for TH
Curious if simming this in HR is would be to edit S4/S5/length accordingly. Maybe there is a feature I am overlooking.
You said this before: "a 3db gain from a piece of plywood is a big deal". Now if you simultaneously gain a lower frequency (potentially 10Hz) with the 3db, that is a HUGE deal.
Originally Posted by weltersys
Sam,
To bring you back to reality may require you do some simulations, though one thing I have brought to David McBean's attention is that Hornresp does no account for the forward directivity a simple "barn door" provides- when you simulate an additional frontal area, IIRC there is no gain change, while in reality there is a fairly large measured increase.
1. No, the 15" deep 90 degree wave-guides that I initially used with FLH, then bass-reflex cabinets, and finally with the Keystones, did not extend the low frequency response on any of them, but did increase level on all of them.
2. The triangle depth, not the height, does increase path length.
3. Given a specific cabinet width, narrowing the inter-cabinet angle does increase the depth of the "V". Without simulation and/or testing the specific cabinets and angle of the "V" plate, as well as the specifics of how well braced and gasketed said device is, one can only speculate as to the results.
If you do do any simulation and testing, please share your results in the wave-guide thread:
Horn Extender/Wave-guide for TH
Thanks,
Art
1. I understand, and I guess in theory only a horn's response could be successfully extended.
2. I think we're on the same page and it's just semantics.
3. Now that I think about it, I think the angle is probably critical. I did a bit of research and from what I can gather horn loading is lost if too wide. You would still get an increase in output but then the extension is only acting as a reflector and not a horn. It’s getting beyond my understanding but I bet there is a way to calculate the best angle/flare rate. In my mind, the angle would be calculated as if one were to add another fold the KS. A quick look at the KS and it appears this angle would be more narrow than wide since the PL is nearly 90deg near exit. That might advantageous since the path length will be even longer and v plate would be smaller. I found this thread really informative and includes REW plots proving the extension does lower FR. Horn Extender/Wave-guide for TH
Curious if simming this in HR is would be to edit S4/S5/length accordingly. Maybe there is a feature I am overlooking.
You said this before: "a 3db gain from a piece of plywood is a big deal". Now if you simultaneously gain a lower frequency (potentially 10Hz) with the 3db, that is a HUGE deal.
Can't seem to put this thing to rest. After some thought, I don't think a v plate could lower the system since the v would create a kink... But, if another panel was made to join the pair (on bottom) creating a "w-plate" maybe? I don't think it would too hard to build and install. I just eyeballed the angles and drew up my thoughts in paint, but I suspect this would really add to overall pathlength since it's so narrow, my guess its' like 15deg.
Attachments
Hi Zwiller,
Sam, if you want to keep the function of the keystone mouth as Art intended, then it might be better to move the keystone mouth to the extension.
For instance like this:
Cheers,
Djim
Hi Zwiller,
It works fine for almost any TH as long you don't want an extension that drops to low. In that case you need to redesign the complete thing and make a balance between best performance for both (both, with and without extender, will not be ideal).
Cheers,
Djim
Hi Djim,
Just pondering here, we are talking about two different things:
1. low frequency enxtension through horn path lengthening, as Art has suggested with his next version of the Keystone, and
2. broad-band level increase through using "barn doors" (large panel reflectors that are not designed to directly extend the horn path).
The first one requires redesign of the basic enclosure, and maybe of shape and size of the Keystone terminus (as that is the result of extensive experimentation). The second can be done simply by experiment?
Regards,
P.S.: Hadn't seen your Post #78.
Just pondering here, we are talking about two different things:
1. low frequency enxtension through horn path lengthening, as Art has suggested with his next version of the Keystone, and
2. broad-band level increase through using "barn doors" (large panel reflectors that are not designed to directly extend the horn path).
The first one requires redesign of the basic enclosure, and maybe of shape and size of the Keystone terminus (as that is the result of extensive experimentation). The second can be done simply by experiment?
Regards,
P.S.: Hadn't seen your Post #78.
Last edited:
Hi tb46,
Oliver, I think I agree with your thoughts and for sure with what you write
Experimenting with "barn doors" is pretty harmless. Even with an ("V") angle of 90 degrees, it still does not represent a theoretical corner since "barn doors" are not infinitely long.
Food for thought, wouldn't it be nice if simulators have the ability to 'unlock' the 'fixed' radiation angles?
Cheers,
Djim
Oliver, I think I agree with your thoughts and for sure with what you write
Experimenting with "barn doors" is pretty harmless. Even with an ("V") angle of 90 degrees, it still does not represent a theoretical corner since "barn doors" are not infinitely long.Food for thought, wouldn't it be nice if simulators have the ability to 'unlock' the 'fixed' radiation angles?
Cheers,
Djim