Rock Block - a space saving sub

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
As a ‘weekend warrior’ sound guy living in a smallish town in Europe I’ve always been struggling with space. Whether it is storage space or transport space, both are limited. Even the venue often has its limitations.

So I’ve been thinking about ways to slim down my hornloaded subwoofers. Since most of a horn consists of air, there surely must be a way to take advantage of this. I’ve tried a few approaches, but nothing really satisfying came out and I didn’t even want to post about it. But now I think I could be onto something and to be honest I am just wondering what you guys think :). So here goes…

Design name: Rock Block
Design purpose: “small" form factor subwoofer for use with band/live sound
Design goals:
  • Small pack space
  • Loud
  • Capable of at least 50Hz, preferably lower
Design compromises:
  • Longer setup time
  • Louder instead of lower
  • Not a run-of-the-mill design

That said, here are the Hornresp sims:

attachment.php


attachment.php


attachment.php


attachment.php


attachment.php


Based on this I designed a foldable prototype, measuring 76x66.5x48 cm (30x26x19 inch) on the outside.
And here are the measurements:

attachment.php

This is taken together with my Keystone loaded with an 18TBW100-8 for reference. Mind the 15TBX100-4 in my prototype is 4 ohms, effectively getting double the amount of watts. So no free lunch. But not bad either, right?

I'm not sure where the 52Hz dip is coming from. It was a ground plane measurement in open air on grass with no real buildings closer than 100 meters.

attachment.php


attachment.php

(some quick EQ I did based on measurements done inside combined with listening tests)

attachment.php

(Measured at 10 meters with limited power, otherwise my measurement system would clip)

This is just a prototype, built with some C class plywood/leftovers, so I expect this to be a worst case version...
 

Attachments

  • Rock Block proto 2 vs Keystone.png
    Rock Block proto 2 vs Keystone.png
    45 KB · Views: 644
  • Rock Block proto 2 impedance.png
    Rock Block proto 2 impedance.png
    24.5 KB · Views: 474
  • Rock Block+filter.png
    Rock Block+filter.png
    37.5 KB · Views: 457
  • Rock Block+filter THD @100W vs 310W.png
    Rock Block+filter THD @100W vs 310W.png
    35.3 KB · Views: 49
  • Rock Block+filter THD @310 Watts.png
    Rock Block+filter THD @310 Watts.png
    59.3 KB · Views: 458
  • HR max SPL.png
    HR max SPL.png
    102.3 KB · Views: 445
  • HR group delay.png
    HR group delay.png
    84 KB · Views: 439
  • HR impedance.png
    HR impedance.png
    73.1 KB · Views: 472
  • HR frequency curve.png
    HR frequency curve.png
    90.7 KB · Views: 460
  • HR Parameters.png
    HR Parameters.png
    103.2 KB · Views: 595
I'll be watching this thread closely.

My main issue is weight, but packing size is also a factor, for sure.

I had previously floated the idea here of building a "medium sized" TH horn, then cutting it into 2 pieces so each would be easier for 1 man to move, then latching them together at the stage. I got a number of responses, all strongly negative. The basic opinion was that there is too much pressure inside a horn, or that the internal structure is too complex.

Well, Inever agreed with either point, seeing as one simply had to find the right spot. It doesn't have to be cut exactly in half, and you would try to separate at the area of lowest pressure. The particular horn design would of course be very important for this to work.

It looks to me like you found a very good spot.
---------------------------------------------------------


But now, if I were you, I'd ditch the hinges and make them 2 separate pieces. (Maybe with some nylon screening to keep bugs & debris out?)

Uses strong latches and a neoprene o-ring type seal, maybe like a Pelican case.

You could still put one part into the other, for transport and storage, but you'd also have the option of separate, lighter boxes for when you need to go up stairs, etc.
 
Now that's some thinking outside the box..!

Very interesting idea. And I actually like the use of the hinges - should make setup time a little easier, amongst other things.

The hornresp sim should probably be set to use PAR not CON segments for a little more accuracy, and if that's a 15" driver, Vtc should probably be a little higher (Vtc should include the volume contained within driver's cone as well as the volume between it and S2).

Export the impedance data from Hornresp and use REW to compare the measured impedance against the predicted one - that should give you a pretty good idea of how close the build is to the sim.

The main issue is likely to be panel flex and stability, as the bottom section isn't properly braced. Any squiggles or smooshed sections in the measured impedance curve that don't have analogues in the predicted curve are signs of panel flex affecting the output of the system. You could use impedance curve measurements as a "guide" to finding the optimal locations for clamps (to clamp the top section to the bottom section) and maybe a removable cross brace (for the bottom section).

Really interesting idea though!

Did you do the FR measurements with the horns side by side like that? If so, that could affect the measurements and might explain that 52 Hz dip.
 
I have been thinking of a similar idea for a much smaller blue tooth connected full range sound system for beach usage. I never would have thought of trying for a horn, I was more thinking of 3-4 telescoping tubes that could screw or lock somehow into expanded position. I like the hinge idea. Do you have wing nuts or the like to lock it together in the expanded configuration?
 
Jared, I actually like your "tube" idea. Except, it COULD be a horn:

Imagine a 25' straight horn, just a continuous expanding diameter, made out of something very rigid. You could cut this into a bunch of smaller sections, and they would stack inside of each other quite nicely.

Once assembled, it could lay across the back (or front) of the stage, with the mouth on its side (towards the front.) Or heck, just add one 90º bend, and have the last third come forward, towards the audience. That way you could have one on each side.

Again, it would have to be REALLY stiff, so no internal bracing, or maybe run thin aluminum "box" channel down the length at a few points.
And if it were "round" like a large, straight trumpet, then that would add a lot of strength all by itself.

Hmmm .......
 
But now, if I were you, I'd ditch the hinges and make them 2 separate pieces. (Maybe with some nylon screening to keep bugs & debris out?)

Uses strong latches and a neoprene o-ring type seal, maybe like a Pelican case.

You could still put one part into the other, for transport and storage, but you'd also have the option of separate, lighter boxes for when you need to go up stairs, etc.
I have built a (way too big) prototype with exactly this idea, but the main problem was that it would take too long to put it together on the venue. With the hinges it takes about 30 seconds or even less. And also, I could not get the 'extender part' rigid enough without adding internal bracing. Believe me, I've tried lots of things and asked many people who I thought had a keen eye for constructions, but every solution was either not strong enough, too invasive (took too much space) or took too long to put together... I still have this prototype. It is really something else. The 15TBX100 is lower and louder in it than the 18NLW9600 is in my drumrizer sub.
 
Now that's some thinking outside the box..!
Pun intended? :smash::spin:

The hornresp sim should probably be set to use PAR not CON segments for a little more accuracy, and if that's a 15" driver, Vtc should probably be a little higher (Vtc should include the volume contained within driver's cone as well as the volume between it and S2).
I've implemented cone correction. If you leave it out, the sim is quite worse. Haven't built a proto without cone correction though...

The main issue is likely to be panel flex and stability, as the bottom section isn't properly braced. Any squiggles or smooshed sections in the measured impedance curve that don't have analogues in the predicted curve are signs of panel flex affecting the output of the system. You could use impedance curve measurements as a "guide" to finding the optimal locations for clamps (to clamp the top section to the bottom section) and maybe a removable cross brace (for the bottom section).
True that! At the mouth panel flex is about a cm each way at 2000 watts :eek:. I've clamped it down at the mouth with two times two opposing clamps (so basically I created two braces) and what it mainly does is bring THD down (by quite a lot, I must add), especially second order, but it doesn't really alter the frequency curve by much. Just listening to the difference I would say it plays louder WITHOUT the braces, that's probably due to the second order harmonics.
I'm still thinking about the best solution for this. Maybe some sort of broomstick with bolts on both ends and a sunken metal dish in the sides of the mouth... Of course any bracing should be quickly installable and removable. That's kind of a challenge!

Really interesting idea though!
Thanks!:blush:

Did you do the FR measurements with the horns side by side like that? If so, that could affect the measurements and might explain that 52 Hz dip.
No. The Keystone was about 3m back with it's mouth to the ground.
 
I have been thinking of a similar idea for a much smaller blue tooth connected full range sound system for beach usage. I never would have thought of trying for a horn, I was more thinking of 3-4 telescoping tubes that could screw or lock somehow into expanded position. I like the hinge idea. Do you have wing nuts or the like to lock it together in the expanded configuration?

My first ideas were based on telescoping tubes, but in practice (I've built a proto) it proved too cumbersome. You would need to use a very rigid sliding solution AND come up with a decent way to seal the whole thing airtight when extended AND how to keep it all in place.

The hinges makes it really fast to 'unfold' the internal/top part and position it exactly right. But they must be very sturdy. I've used el cheapo ones and they just bend...

No, I don't have any locking implemented between the two parts yet. The top part holds the speaker and proved heavy enough to press down on the double weatherstripping to prevent any airleaks, good enough for a prototype. But indeed some sort of butterfly lock will need to be implemented.
 
Yes, there are a few 'compromises', but for now I think it's a reasonable start. Excited for comments :wave:.
Thijs666,

Prototype seems to work well, especially considering the "C class plywood" ;^) !

Your BC18TBW100-8 loaded Keystone appears to roll off about 1/3 octave below my BC18SW115-4 loaded original version.

That said, using a BC18TBW100-4 in the "B-Low" version also did not perform as well as simulations predicted in the low end..

Back to the "Rock Block", other than the impedance response suggesting the horn being a bit shorter than the sim, it looks pretty close to the Hornresp sim.
Well done, very nice concept and execution.

Round the corners, fill in the voids, a few coats of "Turbo Blue", and off to the races!

Art
 

Attachments

  • Lower Art's BC18SW115-4.png
    Lower Art's BC18SW115-4.png
    209.9 KB · Views: 144
Last edited:
I'm still thinking about the best solution for this. Maybe some sort of broomstick with bolts on both ends and a sunken metal dish in the sides of the mouth... Of course any bracing should be quickly installable and removable. That's kind of a challenge!

One or two flat panels that span the width of the mouth and a few butterfly screws might do it... :)
 
The hornresp sim should probably be set to use PAR not CON segments for a little more accuracy.
Thanks. I just tried this and (luckily) there is hardly any difference. Nothing worthy of posting.

Export the impedance data from Hornresp and use REW to compare the measured impedance against the predicted one - that should give you a pretty good idea of how close the build is to the sim.

attachment.php


Could be better, could be worse...
 

Attachments

  • Impedance-sim vs measured.jpg
    Impedance-sim vs measured.jpg
    72.7 KB · Views: 401
True that! At the mouth panel flex is about a cm each way at 2000 watts :eek:. I've clamped it down at the mouth with two times two opposing clamps (so basically I created two braces) and what it mainly does is bring THD down (by quite a lot, I must add), especially second order, but it doesn't really alter the frequency curve by much. Just listening to the difference I would say it plays louder WITHOUT the braces, that's probably due to the second order harmonics.
I'm still thinking about the best solution for this.
A fairly quick install could use flat metal strips routed flush on the cabinet exterior, with ratchet straps pulling wood braces in place between them.
The example below used a single vertical ratchet strap without wood braces for reference- seems there are some neophytes who have never encountered their many uses.

Art
 

Attachments

  • Strapping.png
    Strapping.png
    713 KB · Views: 129
Last edited:
Thijs666,

Prototype seems to work well, especially considering the "C class plywood" ;^) !
Maybe even D class :boggled:, but it was thrown away at my brother in law's company and this way it's had some more useful purpose before going up in flames;)

Your BC18TBW100-8 loaded Keystone appears to roll off about 1/3 octave below my BC18SW115-4 loaded original version.

That said, using a BC18TBW100-4 in the "B-Low" version also did not perform as well as simulations predicted in the low end..

Sadly I don't have the funds to buy the SW115 just to try it out :cool:. I would love to be able to check this. Or be able to measure a well-known industry sub side to side, just as a sanity check for my measurements.

Back to the "Rock Block", other than the impedance response suggesting the horn being a bit shorter than the sim, it looks pretty close to the Hornresp sim.

I thought the sensitivity was quite a bit lower than predicted. Hopefully this is mostly due to losses due to the bad plywood and a final version with proper plywood will improve this a bit. As I said, I guess this is a worst case scenario.

Well done, very nice concept and execution.
Round the corners, fill in the voids, a few coats of "Turbo Blue", and off to the races!
Thanks :). Nah, this is just a proto. The OSB is really bad as cabinet material. It isn't even plywood. And it splinters badly :mad:. I was thinking to use Baltic Birch 12mm for the outside of the top part, poplar for the inside with a lot of bracing (to keep it light, as you will have to lift this part) and use 22 or 25 mm BB or douglas fir plywood for the bottom part, as there will be no internal bracing possible (only the front will probably be braced). But maybe glueing two different sorts of wood together isn't the greatest idea? :confused:
 
A fairly quick install could use flat metal strips routed flush on the cabinet exterior, with ratchet straps pulling wood braces in place between them.
The example below used a single vertical ratchet strap without wood braces for reference- seems there are some neophytes who have never encountered their many uses.

Art

I've tried straps like this on a different prototype, but I have a couple of issues in this:
- The sheer force/pressure of the sound waves is so intense at the mouth, that the straps would have to bend the plywood so far inward, that I would fear for it to deform permanently.
- The movement that has to be stopped is two-way. Straps only prevent one-way. So the best solution I see would be a rigid construction to keep the sides from moving inward or outward. Luckily the forces applied to the sides are opposite to each other, so a simple threaded bar could – in theory – do the trick. (in practice it's a different story; I've tried :()
 
A fairly quick install could use flat metal strips routed flush on the cabinet exterior, with ratchet straps pulling wood braces in place between them.
The example below used a single vertical ratchet strap without wood braces for reference- seems there are some neophytes who have never encountered their many uses.

Art

I've tried straps like this on a different prototype, but I have a couple of issues in this:
- The sheer force/pressure of the sound waves is so intense at the mouth, that the straps would have to bend the plywood so far inward, that I would fear for it to deform permanently.
- The movement that has to be stopped is two-way. Straps only prevent one-way. So the best solution I see would be a rigid construction to keep the sides from moving inward or outward. Luckily the forces applied to the sides are opposite to each other, so a simple threaded bar could – in theory – do the trick. (in practice it's a different story; I've tried :()

I'm sorry. I just now understood what you meant :blush:. With the wooden braces in between the sides and kept in place by the pressure of the straps... That might work. Though I've tried this on another sub with some serious wood clamps and they just weren't strong enough to keep the brace in place. :eek:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.