Hi,
I was reading through BLH and a thought came to me. What if you created a BLH horn by putting in a 2" PVC pipe inside 3" PVC pipe so the end of the 2" PVC pipe will now enter the 3" pipe at the bottom. The PVC pipe would then be fitted with another 6" PVC pipe on the outside. The top end of the 3" pipe will now enter the 6" pipe from the top. The exit of the horn is at the bottom. So effectively I would have created a 3 fold horn with the thinner pipe inside the thicker pipe, hence the coaxial part.
I have not done the calculations, neither have I built a horn yet. Just talking purely from scientific curiosity...
Oon
I was reading through BLH and a thought came to me. What if you created a BLH horn by putting in a 2" PVC pipe inside 3" PVC pipe so the end of the 2" PVC pipe will now enter the 3" pipe at the bottom. The PVC pipe would then be fitted with another 6" PVC pipe on the outside. The top end of the 3" pipe will now enter the 6" pipe from the top. The exit of the horn is at the bottom. So effectively I would have created a 3 fold horn with the thinner pipe inside the thicker pipe, hence the coaxial part.
I have not done the calculations, neither have I built a horn yet. Just talking purely from scientific curiosity...
Oon
For me it sounds more like a expanding TML but you are talking about a basic concept.
Of course it is possible to use PVC pipes and you can achieve some good results with it.
Depends on the speaker you wanna use.
Smaller eg. 3" i would say no problem but the larger the diameter the more air is pressed through your pipes
and maybe the thickness of the PVC can be a problem.
Those pipes are cheap, do some simulation with HornResp or LeonardTL and do some tests 🙂
Of course it is possible to use PVC pipes and you can achieve some good results with it.
Depends on the speaker you wanna use.
Smaller eg. 3" i would say no problem but the larger the diameter the more air is pressed through your pipes
and maybe the thickness of the PVC can be a problem.
Those pipes are cheap, do some simulation with HornResp or LeonardTL and do some tests 🙂
Just put a trapezoidal or rectangular baffle plate for driver though to avoid the worst kinf of 10dB dips/peaks due to round baffle diffraction artifacts. You can easily model in HR or other code using simple CSA changes and lengths.
I had just thought of this exact same thing a couple of months ago - http://www.diyaudio.com/forums/full...board-speaker-enclosures-283.html#post4969273 - and am still in the process of learning and simulating designs in Hornresp. My latest iteration, for a full range Jordan JX92S driver, is 1m in diameter and 30cm deep. This provides a mouth area of up to 0.94 square meters and around 1.3m of path length using an exponential expansion! Those parameters yield a lower cutoff frequency of 100Hz and the approximate total volume of the enclosure would only be about 0.25 cubic meters. The limitation is that this enclosure would have to be mounted on or very close to the wall, but that also means it only has to radiate into half space which should provide a nice boost at the low end.
I had not envisioned using PVC tube/pipe because it would really limit the design in terms of the expansion profile. A design like this is a segmented horn where the segments do not expand over their length, but where the segments connect (two 90 degree bends) there is a volume increase as you go around the bend and then the volume decreases back to that of the next segment. It's obviously not ideal but it can provide quite a bit of path length in a compact space.
I had not envisioned using PVC tube/pipe because it would really limit the design in terms of the expansion profile. A design like this is a segmented horn where the segments do not expand over their length, but where the segments connect (two 90 degree bends) there is a volume increase as you go around the bend and then the volume decreases back to that of the next segment. It's obviously not ideal but it can provide quite a bit of path length in a compact space.
Last edited:
Hi,
I tried reading about horns etc and TL. Just want a quick answer can't seem to find it round the usual article. If you were to make a TL or horn, what is the minimum size of the horn entrance. Is there a rule of thumb answer? Must it have a cross sectional area the same size as the speaker driver? Can it be smaller?
Thanks.
Oon
I tried reading about horns etc and TL. Just want a quick answer can't seem to find it round the usual article. If you were to make a TL or horn, what is the minimum size of the horn entrance. Is there a rule of thumb answer? Must it have a cross sectional area the same size as the speaker driver? Can it be smaller?
Thanks.
Oon
Most people start simulations with a throat area that is the same as the cone surface area. The ratio of cone area to throat area is part of what determines the efficiency gain. The throat area, mouth area, flare constant, and horn length all need to be considered together. It's difficult to grasp this without using a simulation program like hornresp to see how changing different parameters affect the response.
This coaxial path has a long history - do a Google Image search for "reentrant horn".
These narrow paths = good for passing HF through a bend. Note that front loaded reentrant horns are often used for the ~300-3kHz speech range.
This good HF means it is probably bad for a BLH, where you only want the horn to pass LF.
These narrow paths = good for passing HF through a bend. Note that front loaded reentrant horns are often used for the ~300-3kHz speech range.
This good HF means it is probably bad for a BLH, where you only want the horn to pass LF.
hollowboy, thank you for pointing that out. It hadn't registered as being similar to what we're talking about but it definitely is. Do you have any sources that explain sound wave propagation through a bend in more detail? I've been doing lots of searches about this recently and what I had found was more or less related to larger ducts/tubes and lower frequencies. The general results were that low frequencies were less attenuated than high frequencies through 90 degree bends.
Assuming a throat area equal to the cone surface area of a 4-5 inch driver and a lower cutoff between 100-200Hz, the length will need to be much longer than that of a bullhorn, and the exponential expansion would start around 70-80 square cm. The size of most bullhorns means that the throat area is much smaller than 70 sq.cm because they're obviously not typically meant to amplify lower frequencies.
Assuming a throat area equal to the cone surface area of a 4-5 inch driver and a lower cutoff between 100-200Hz, the length will need to be much longer than that of a bullhorn, and the exponential expansion would start around 70-80 square cm. The size of most bullhorns means that the throat area is much smaller than 70 sq.cm because they're obviously not typically meant to amplify lower frequencies.
Thanks Hollowboy, I noticed the Loudhailer of having that design when I looked up Horns in wikipedia.
I think it was written somewhere that most BLH are more like TQWT, since the mouth of the horn is normally smaller than the wavelength of the horn. So my interest at the moment is to use the coaxial pipe as a TL rather than a horn. However, you are right that it will be expanding TL. Since pipes typically come in a few sizes, it will more likely be a 4" in a 6" in a 8". Since the cross sectional area after minus the pipe that is inside will be in this sequence 16sqin, 20sqin, 28sq in. So it is slowly expanding. I have some questions which I am soliciting some answers
1) I was thinking of putting the all the pipes skewed to one side, to provide a bigger airway on one side, making it less narrow at 2" rather than 1". Would it be better or worse?
2) I think I would probably use equal length pipes. Making it easier to mount. Cutting out part of the sides of the pipe to be the link rather than using the bottom of the pipe.
3) Can the exit of the horn be at the same side as the speaker? Or must it always be a the far end of the speaker cabinet.
Thanks
Oon
I think it was written somewhere that most BLH are more like TQWT, since the mouth of the horn is normally smaller than the wavelength of the horn. So my interest at the moment is to use the coaxial pipe as a TL rather than a horn. However, you are right that it will be expanding TL. Since pipes typically come in a few sizes, it will more likely be a 4" in a 6" in a 8". Since the cross sectional area after minus the pipe that is inside will be in this sequence 16sqin, 20sqin, 28sq in. So it is slowly expanding. I have some questions which I am soliciting some answers
1) I was thinking of putting the all the pipes skewed to one side, to provide a bigger airway on one side, making it less narrow at 2" rather than 1". Would it be better or worse?
2) I think I would probably use equal length pipes. Making it easier to mount. Cutting out part of the sides of the pipe to be the link rather than using the bottom of the pipe.
3) Can the exit of the horn be at the same side as the speaker? Or must it always be a the far end of the speaker cabinet.
Thanks
Oon
Most are compromised horns because they don't strictly follow the expansion.
From a geometrical standpoint, this would introduce various path lengths in parallel, but it's still one air mass. It would probably act like it is tuned at different frequencies depending on the geometry. I have no idea how this would perform and if it would be better or worse!
That's a good idea. I was thinking of using dowels or threaded rods and nuts at various places.
It can go wherever you want - front, back, or side. In relation to the front wave, the back wave will be delayed by the length of the horn, so different mouth positions will have different effects on the summed frequency response.
Can't remember / find where I got the info*. But, if I recall correctly (and making the usual assumptions): there's an inverse linear relationship between how wide the horn is at a bend, and how much HF passes through.
*I found this review while looking. It contains a tale about too much midrange getting through a BLH.
6moons audio reviews: Atelier Audio First Horn
- Status
- Not open for further replies.