AVS Forum banner

DIY bass horn build...Up next

69K views 462 replies 50 participants last post by  wireburn 
#1 ·
EDIT: This is for PA, not home audio or HT.



Ok. I've never built a bass horn yet and after experiencing Labsubs and lately the DTS-10's I've gotten the itch to give it a go. The ported 21's are about a week away from being wrapped up and I've finally decided on a horn simulation that I'd like to turn into a pile of sawdust and wood...I think.
I figure I'd better get the ball rolling if I expect to get this built before weather turns bad for the winter.


I've run through a ton of simulations in the last 6 months and there are about 20 possible scenarios that I like and about 8 that it was really tough choosing between for the build. I weeded that down to about 3 possibles by eliminating the ones involving drivers I don't already own. Why buy if you don't need to? I ended up deciding on a 380L net, 30hz FLH using the LMS 5400, over a 380L 30hz B&C FLH and a 30hz 380L B&C TH.

The performance of each is close enough on paper to make it a toss up really.


For the curious here's my thought process. The other contenders were involving dual xxx12's in a TH, TC's PA5100 and the 18sound 21NLW9600. The 18sound 21 and the B&C are VERY similar in performance when using the same enclosure, basically interchangeable IMO with very small differences in modeled performance, so that eliminated the 18sound because I already own the B&C. The PA 5100 also models very well, but differently in the same basic horn parameters as the 21's but obviously needs some changes in the overall form factor due to being a smaller diameter deeper driver. Again since none of them really showed a notable performance advantage on paper and I already own the B&C's...The dual xxx 12 sub I would really like to build (sub bass gargantua), but it's the most complicated, expensive, heaviest and largest. Maybe later. The LMS I decided on over the B&C, because it's the less obvious choice and more of a known quantity at this point. Plus I'm already doing a new build for the B&C's. The build I am doing will be leaving some of the LMS's potential on the table. It'd be better in a larger 20hz FLH or an 18hz TH, but I don't want to build anything so large and difficult to move first off. If I can't easily fit it in the back of my Jeep with the help of 1 other guy it was eliminated this is also why I ended up with a 30hz corner. The LMS doesn't like TH's so small tuned so high.





Anyway. This is what I've got. Nothing is set in stone yet. I need help reviewing the model for mistakes and I could especially use some help with folding the horn up. I've never folded one, so help or comments from more experienced guys would be much appreciated. All opinions are welcome.
BTW the model is intentionally a very basic single expansion rate conical (parabolic now) type to hopefully keep the folding and build as simple and straight forward as possible.




Input parameters
Attachment 183905


Phase and delay
Attachment 183906


1W into the minumum impedance in half space
Attachment 183907


124v input in half space
Attachment 183908


124v driver excursion
Attachment 183909




 
See less See more
7
#403 ·

Quote:
Originally Posted by Ricci /forum/post/20572497


...

Okv you seem to have a good amount of knowledge and are quite the lurker I see (4 posts in one year). Glad to see I've goaded you into posting.

Thank you for noticing, I have found lots of inspiring information here.


There are many knowledgeable people around here, and when you present measurements that indicate something similar to what I have measured myself, and can not yet explain exactly, then I feel it is a good idea to participate and try to find out what is going on - may be someone else knows something I don't.


If the mechanisms behind distortion and high-level changes in frequency response is fully understood, then it may be possible to either make improved designs, or, live with it and utilise the designs differently.
 
#404 ·
Ricci,


I see the thread you started elsewhere, and I understand your posts a little better now.
It least it hasn't hit "Sound difference between horn and sealed/vented?" levels of absurdity yet.


I'm a little disappointed in the VI Box. Sorry, I ever brought it up in our emails. I was a little worried about it handling the loads you would be placing on it. From the details of it's circuit and components I could never figure out how it could handle 2500W at that price point without liquid immersion. A resistor that can handle that load (dry,) and stay in spec is not cheap.


Maybe you just got a bad box. Crossing my fingers.


They replaced this one for free, right?
 
#405 ·
No it was not free...They didn't offer and I didn't ask. I actually never saw any mention of any warranty on the site at all so I didn't bother.Talked to Chris at LinearX. He says this has never happened. I do not have time to wait around for discussions of what might of happened whether it will be replaced etc. I recieved the new one that I bought the same night today. I will be sending the burnt unit back for them to look at. Not much to say, the 4 first resistors on the high level input are toast. I could not have put anywhere near 2500w into it. I was running a single channel into a 5ohm minimum load and the amp was on a $10 powerstrip at the time. That amp could maybe muster all of 1000w into that and it was nowhere near clipping. I asked about duration and didn't get an answer on that. I had done 3 or 4 sweeps in a row at a good level when I saw an orange glow and the magic smoke. If I smoke the second I will have to figure something else out. Vance D. has been using one for years quite successfully. Getting ready to mess with the replacement now. Fingers crossed.




"yet"
No doubt.
 
#406 ·
How did the testing go with the new box?


Not hearing anything by now has me thinking things didn't go smoothly.
 
#409 ·
At 65Hz, Th S3, you have a particle velocity of 49m/s and the pressure equals >170dB.

Then it should not come as a surprise that the output is affected by nonlinear effects in the horn, this is a very high pressure and a very high velocity.


Considering these numbers, it is more surprising that it actually performs as well as it does.

Numbers from other horn designs show that very high pressures and high velocities are common, and a smaller horn may actually present numbers higher than a larger one.

Also, the area of the horn channel is important here, because a larger area can manage higher velocities (and obviously, higher volume velocities) before separation and turbulent flow occurs along the walls.


I find your Gjallerhorn very interesting; it is extremely loud and powerful, and the powerful driver really works the horn, so that it may be possible to actually measure and find physical limits for this kind of constructions.

Knowing that design of things, and perhaps loudspeakers in particular, are often based on experience and known good practices, one may find that what was valid for horns in the old days may not be true for this kind of design, because the driver is capable of producing a lot more low freq sound due to its displacement capacity.

PA horns can of course be even louder, but rarely goes as deep and the drivers usually have a lot less displacement.


Now it would have been very interesting to see some distortion plots of this one.

Most useful would be dist at different levels, and 2hd and 3hd separately plotted.

(I understand you are working on getting the impedance measurements.)


I have started to look at nonlinear effects in low frequency horns and distortion, and what I can see so far, is that it may not be a big concern, as the distortion from the driver itself will be quite large before the horn itself add significant levels of distortion.

That is, unless you make an unfavourable design - and that is a good reason to search for more knowledge about the subject, so that one can learn good design practices, to avoid building expensive piles of wood.

It also surprised me to see the actual numbers for pressures and velocities inside the horn - very high pressure, and velocities that also quickly enters the range where separation and turbulent flow occurs.


The kind of dist people are so afraid of in horns, often called throat distortion, is not a big problem in bass horns, as the bandwidth of the horn really is not wide enough to allow waves to propagate long enough (in wavelengths).


Talking about turbulence, one can imagine what happens through the horn bends..


I tend to believe some distortion at very high spl levels are not that important, as the speaker should only enter this kind of spl on top of transients, and then the dist will not be detectable as the time span is too short.

I am more concerned about compression.


I think that using practical designs and measurements of those are the way to go.

This makes it possible to verify theoretical hypotheses, and find out what actually works.


I shall not indulge in to this any more here, as the subject is rather comprehensive, and also I do not have much valuable information, at least not yet.
 
#410 ·
I appreciate your comments Okv. I already have all of the data collected including distortion measurements. The only thing left is high power impedance. Anyway I had not planned to post the rest of the data until the site update was done but I will post some of the distortion and also power compression results tomorrow.


You note that the internal pressure is reaching 170db at 63hz. (Btw horn segment 3 mouth is the end of the throat effectively and where all of the forces in the horn seem to be highest.) I am also interested in the area near 90hz as this is where the cabinet develops a notable buzz/ resonance at war volume giving the impression that the cabinet is in danger of popping a joint . It is hard to describe. I havent noted it with actual content but I dont normally dump >100v into cabs in real world use either. The charts indicate there is significant force involved at that area of response also.
 
#412 ·
1w/1m, 2 volts input.






Approximate input voltage for the following nominal levels.

90db 1.63v

95db 2.9v

100db 5.15v

105db 9.16v

110db 16.3v

115db 29v

120db 51.5v

125db 91.6v

128db 130v



Power compression sweeps.




THD and distortion by component measurements.












CEA2010 output. Measured at 2 meter, ground plane, rms.

10hz 90.2db

12.5hz 105.2db

16hz 118.3db

20hz 122db

25hz 124.9db

31.5hz 125.8db

40hz 128.4db

50hz 127.2db

63hz 130.2db

80hz 130.9db

100hz 131.1db

125hz 123.6db
 
#413 ·
Sorry if this was mentioned before - what proggy do you use to generate your distortion graphs? I tried using RTA in REW, but as far as I know, you can only do one frequency / gain at a time - no ability to do a sweep, which would sure save time.
 
#416 ·
Excellent, these distortion plots add very valuable information.


Measurements (still) at 2m?

(Meaning - add 6dB to get to 1m.)


Distortion is overall sufficiently low, considering the SPL and frequency range at work, I would say very low.

I would not worry too much about distortion at the limit, and it can also be seen that distortion falls rapidly as spl is reduced from the 128 max level.


What is interesting, though, is to observe how the distortion develops.

It first increases at higher frequencies, relative to the passband, and then narrow bands with much higher distortion appear.


Measurements of other designs shows similar characteristics, which is interesting, because that may indicate that the mechanisms behind generation of the distortion are the same.
 
#417 ·
Yes all output spl measurements are taken at 2 meters.



Some thoughts I have after looking at the data for a few days in different ways....

I think there are some trends apparent here if you consider the pressure, imp, velocity, current and acceleration data from Akabak in context with the actual compression and distortion measurements. The distortion performance I am pretty happy with really. The dramatic spikes up seen at high power seem to be inherent to TH's when driven hard. The DTS10 both in stock form and with alternate drivers exhibited similar spike in distortion. This cab is ridiculously clean from 15-45hz considering the power applied and the output. It gets rough above there but again given the levels used, I consider that it stays well under 15% THD from 15- 95hz even at the 130v level except for 2 up spikes at 50 and about 67hz to be great. The biggest spike in distortion near 100hz is also where I note that at the highest input levels the cab develops a buzz. Were I to build another pair I would beef up the internal bracing more. I plan to cross at 85hz or lower anyway.


Note that in the power compression test there is what appears to be an upper ceiling to how loud this sub will go at some frequencies no matter the increase in input power. There is a very large amount of compression of the 63hz responce peak and also the 85-110hz area where the sub also exhibits peaks in its response. Again this same type of compression behavior was noted with the DTS-10. A follow up test with EQ applied is planned. I have a feeling it will have some good effects on the distortion and high power compression performance in addition to giving the sub a smoother more linear response. For example the sub plateau's out at 63hz and ends up exhibiting nearly 12db of compression at the 130v input level. However looking at the basic response at the lowest 90db or 1.63v input level 63hz is boosted almost 11db relative to the surrounding frequencies to begin with. The same story is apparent at 85-110hz where the response is boosted up by 5 to 10db and again the compression and distortion performance is worst in this area. The output has already almost plateau'd at these frequency ranges by the 120db, 51.5v sweep. Further input does little more than add to the distortion. Once the response is EQ'd to be flatter at these frequencies by cutting some 5-12db of output, the compression performance, distortion and power demands on the amplifier should all improve quite a bit while the maximum output should be affected barely at all, plus the decay and response will be smoother. That's the theory anyway. Should be a win/win all around.


If you are running TH's I would consider EQing the major spikes in response down a must for better performance of the cabinet.
 
#418 ·

Quote:
Originally Posted by Ricci /forum/post/20624027


..

Some thoughts I have after looking at the data for a few days in different ways....

I think there are some trends apparent here if you consider the pressure, imp, velocity, current ...

Your excellent documentation is what makes it possible to study data that gives very useful information.


To actually come up with something theoretically valid for distortion in low-frequency horns may take some time and effort, though.

It might not even be worth the effort, seeing as distortion clearly is not that big an issue before the limits of the design is reached.

But even to come to that conclusion one needs measurement data from horns actually made.


To me it also seems very likely that the spl limit is actually set by the driver.


And, there are artifacts that seems to occur in many designs.

Like resonant spikes in the upper range of the passband (i.e. like your 100Hz-peak) with high distortion.


The distortion issue might perhaps deserve its own thread, where other as well could contribute with measurements and ideas.
 
#419 ·
Ricci could this cabinet design be used for live applications?


I always thought about having a PA style bass rig with a sub and top. The Top would probably be a pair of AE15's with some sort of horn to get up to atleast 16Khz. I would mainly just use the top for my standard tuned 6 string bass. Then when using the ERB bass I would use a sub like this one if it would work. 15hz is about as low as I would want to go myself. I figure a steep Hpas here would be perfect for me when using the Top/sub together.


Just a thought. I still liked your 24" cube, dual 21" B&C sealed sub also you did a while back when making your other pa subs.
 
#421 ·
Somehow I missed these questions a month ago...I just saw them when I came to update the CEA2010 data.




Quote:
Originally Posted by chrapladm /forum/post/20640804


Ricci could this cabinet design be used for live applications?.

I guess it should certainly be loud enough if you use 2-4, for most applications, but really the cab loading is lower than needed, and it is really big and heavy. You could gain a lot of top end output and efficiency in the meat of the live sound range if you moved the corner up a whole octave higher. You know like the Othorn. Basically the Gjallerhorn would be OK 40-90hz, but some smaller, lighter cabs can compare favorably over that range. It isn't until the octave and a half below that range that it starts to set itself apart from the normal live sound gear. In most live or SR instances that sort of content is pretty much non existent anyway though. Everybody usually filters below 30hz pretty heavily to protect their equipment and cabs and live music doesn't have much below 30hz anyway, so the extra extension is wasted in that application.


Don't get me wrong I am using them for live sound right now
but they are stationary, it is only for myself and I run the system.




Quote:
Originally Posted by 04FLHRCI /forum/post/20640854


Any estimations how one of these guys would perform in a ~3000 cube HT LPF @80Hz?


Violently. The pair that I have have been pushed in the >10,000cu ft very lossy space they are in a few times, just having fun with them and it is more than plenty for that space. They can shake the entire building and easily pressurize the space with Jurassic Lunch and I only have about 1300w of amplifier going to each one. They aren't even using all of that power. Some people say there is no such thing as overkill, but seriously the Gjallerhorn is simply not needed in any sort of domestic environment. Even one will do reference levels to 16hz in a normal room without breaking a sweat. I guess if you had a screen wall it might make some sense, but most of the time I doubt that you would use much of the available headroom. Most situations would be better served with multiple smaller, more modest subs scattered about the room to smooth the response.


I'd be lieing if I said that I wouldn't like to experience what they would do "opened up" in someone's HT room just once though. Maybe someone will have a GTG I can drag one to.
 
#422 ·
Here are the CEA2010 distortion results to go along with the basic output spl numbers.


Gjallerhorn CEA2010 output. Referenced to 2 meters, rms, outdoor groundplane.


10hz-----90.2db-----Distortion

12.5hz---105.2db----Distortion

16hz-----118.3db----Amplifier

20hz-----122db------Driver noise

25hz-----124.9db----Driver noise

31.5hz---125.8db----Driver noise

40hz-----128.4db----Amplifier

50hz-----127.2db----Amplifier

63hz-----130.2db----Distortion

80hz-----130.9db----Amplifier

100hz----131.1db----Amplifier

125hz----123.6db----Amplifier





















Here is 10hz distortion at just about maximum output. Obviously this is not clean or a passing result. This is way below the cab loading and the distortion is very high, but it shows what the Gjallerhorn can do dynamically down that low before getting the driver into mechanical trouble.

93.2db at 10hz
 
#423 ·
wow.


did you run out of amp at 100hz and 125hz? is that why the blue line never touches the red line distortion limit or is it something else?


also, it looks like your getting the hang of running these kinds of measurements. hopefully that is cutting down the time required.
 
#424 ·
also, what kind of signal is being used for these tests? is it a standard that you can link to or is it something else?


i'm particularly interested the crest factor. manipulating that seems to be increasing in popularity as a way to hit higher peak outputs and appear to have higher sensitivity.
 
#425 ·
The signal is the standard CEA2010, 6.5cycle duration, Hann shaped, 1/3rd octave bursts. 100 and 125hz are very short duration. Nothing more than a blip. The program usually just records them like that.


I ran out of amp at 16hz, 40, 50, 80, 100 ,125hz. There were driver distress noises that made me leery of extra level at 20,25,31.5hz despite low distortion results. Distortion limited 10 and 12.5hz and 63hz. That is if I remember all of this right.
 
#426 ·
if running out of amp all over the place, what is actually being measured? is it distortion caused by the driver/enclosure or by the amplifier as it is running into clipping?


edit: i see you updated the numbers with notes. that makes some sense and suggests that there is even more output potential of the subwoofer.
 
#427 ·

Quote:
Originally Posted by Ricci /forum/post/20788532


I'd be lieing if I said that I wouldn't like to experience what they would do "opened up" in someone's HT room just once though. Maybe someone will have a GTG I can drag one to.


*laughs* Just kidding. Well, sort of ... I'd go for it if I didn't already have too much on my plate.


Ricci, your Gjallerhorn is CRAZY STUPID CLEAN through the whole lower bass region. Frickin' insane output out of one. It's too bad it gets so 'meh' above 80hz. I bet if you were to cross a cluster of the Gjaller to some uber midbass array you would have the ultimate bass system for outdoor venues. As long as you never had to move them... ever again, anyway.




Love the work you're doing, Josh.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top