Single Driver Front and Rear Horn

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
There are quite a few RLH horn designs around, yet from what I understand loading only one side of the driver isn't optimal. I was reading some of Dinsdale's work and he suggested using 1 driver and covering 80hz up to 18kz with the front radiation loading an HF horn and the rear radiation loading a bass horn.

If I'm willing to use a sub to cover the bottom octaves, are there any reasons not to try to tackle such a project using a driver such as the FE206E ?
 

GM

Member
Joined 2003
Greets!

Right, the ideal BLH loading is hyperbolic since the front of the driver 'feels' a ~infinite 'compression' chamber, so to get an optimally loaded shorter, faster flaring BLH requires a front horn to provide the reactance annulling/mass loading to ensure a uniform acoustic load on the driver. Since this mass loading not only lowers the driver's effective Fs, but also raises its effective Qts, a low Qts driver such as the FE206E is the obvious choice.

The concept is as simple as the ubiquitous B*** bandpass 'sub', a successful execution of it is somewhat more elusive. :(

GM
 
I was wondering just how challenging this might be. I had the thought of roughing up a couple different short front horns and somehow attaching them to existing cabs just to see what they would do. Sounds easier than it probably is.

I did get the Nagaoka books and was hoping to find something of this sort in there, but didn't see anything. It seems that it wouldn't be too challenging to design a BLH with a front horn worked in. (If you're already knowledgable to design a BLH in the first place) However, it seems that must not be the case, as you certainly don't see it too often.

From GM's comments, I'd guess that to do this correctly, the front horn would somehow have to reflect the amount of loading the driver is getting from the BLH. However, I have to wonder just how vital this would be in the end. Figuring out the correct flair rate is challenging enough without throwing that variable in there.

I'd like to see where this goes...
 
I'd think the hardest part would be constructing a good front horn, but since it will only be about 5" of horn length it can't be too bad.

Just to verify that I'm on the right track:

1. I can go with expo flare rates on both horns.

2. The combined horn lengths is important as the output of both
horns needs to be in phase at the cutoff frequency between the 2 horns.

3. The throat and chamber sizes for the bass horn are more critical than I see in typical RLH's because this determines the upper cutoff of the bass horn.

4. I should plan on a 10% overlap in response of the 2 horns, which will be somewhere in the 1200hz range.

5. Since I'm going to take the bass section up so high, that horn will have no parallel sides and I will keep the pathway as close to square as possible.

As long as I do the math properly, using HornResp as a tool and construct what the math tells me, do I have a high probability of success or is this a speaker design with unforeseen pitfalls that are likely to reach out and bite me in the a**?

Also, with a lower cutoff of 80hz am I going to need a filter to protect the driver below that using an amp with only a few watts of power?
 
That is along the lines I was thinking and Dinsdale's minihorn was my inspiration, but I couldn't make heads or tails of the drawings. Alf's bass horn is only covering a little more than 2 octaves and I need 4 since I want true single driver, if it is reasonably possible to get right. With my bass horn going so high, I will definitely need front firing.
 
4. I should plan on a 10% overlap in response of the 2 horns, which will be somewhere in the 1200hz range.

It's just intuition, but I would think you would want the hand off to be quite a bit lower than this. The 206 looks to have a serious rising response. If my understanding is correct then loading this into something like a 400Hz front horn would bring the mid range up to meet the treble which will not be boosted by the front horn. I would also think that you might have trouble getting a horn to play from 80Hz thru 1200Hz. Maybe not though....
Joe
 
Joe,

That brings up a question that's been nagging at me. What happens to the frequencies above a straight front horn's upper cutoff? Are they attentuated? How much efficiency is gained in a front horn?

I guess my question is how to best account for the rising response of the 206? More efficient rear horn with a longer horn or less efficient front horn or leave the top end not loaded by a horn as Joe suggests.

If I can go with a mid front horn and stay single driver, that gives me the option of going with a much lower cutoff on the bass horn. My gut feeling is that doing so is going to require a tweeter.

Efficiency, SQ, and retaining a single driver speaker are my goals. I want to take this pair of drivers to their full potential. If they still end up with an uptilted response in the end, I'll just damp the room a bit.

Size is of little concern, but having the smaller size with a higher bass cutoff is attractive. If I do lower the bass cutoff, I want to keep any compromises to a minimum. I've got dipole bass with my current setup and I don't want to go thru all this work to end up with horn bass that sounds like a ported woofer.
 
diyAudio Editor
Joined 2001
Paid Member
The post #7 wit the square tractrix horn reminded me of my fav question:


Does this geometry have any afvantage over the tractrix?
http://aa.peavey.com/downloads/pdf/qwp1.pdf

I can think of one-its a whole lot easier to make using layers of thin plywood or fiberglass reinfored panels that you can buy at a home center. (restaurants use it to cover walls - also used to make cheap shower enclosures. Its very flexible with a white color. One could make the shape out of thin stuff, then goop on think goo -foam or plaster or something on the outside to make it non resonant.

John in cr :

I'm a horn newbie, but using the loading to control the exagerated mid response of some drivers sounds doable and smart. Probably lot of people do it :xeye:
 
Greets!

Right, at 1200Hz, the mouth's output will be hundreds of degrees out of phase with the front radiation and their acoustic centers will be much >1WL apart, so lots of comb filtering. Then there's the high compression ratio required which the driver probably isn't rigid enough to handle without a complementary load on its front side, i.e. has the same effective mass loading. This mass will dictate a large, long front horn since it will have the same size throat, ergo will roll off the top end somewhat.

From this we see that there's two ways to approach the design, build a BLH of 'x' gain/BW/'M' factor and use a WG to control the front's output in the acoustic XO BW and hope that the driver is rigid enough to handle the non-linear loading at a low enough distortion or assume the worst and design a low gain FLH/BLH that use the same 'M' factor for both and choose the flare frequency that best flattens the on-axis response.

An octave is a 2:1 ratio, so 100, 200, 400, 800, 1600, etc. would each be a one octave (3dB) spread.

2^n = Fh/Fl

n = (m*log(Fh/Fl))/3.01

Fh = Fl*2^n

Fl = Fh/2^n

dB (change) = m*LOG10(Fh/Fl)

where:

Fh = upper frequency
fl = lower frequency, or the XO point in this case
n = octave spread
m = log multipler:

m = ((log10(log base))*delta dB)/(log(log ratio))

10 = 3.01dB
20 = 6.02dB
30 = 9.03db
40 = 12.04dB
50 = 15.05dB
60 = 18.06dB
70 = 21.07dB
80 = 24.08dB

Tractrix is just another way of saying truncated hyperbolic, so better IMO to design with a gain appropriate for the BW/desired SQ and polar response and do whatever is required to control mouth reflections, whether by mouth shape or other means.

GM
 
GM,

First, thanks for the detailed info.

Regarding the comb filtering problem:
Since I'm used to OB's, I'm of the opinion that a ruler flat response is highly overrated. Would this comb filtering be very detrimental or just a compromise of the design.

Phase issue and horn length:
Is calculating the 2 horn lengths and ensuring the combined length is an odd multiple of 1/2 wavelength at the acoustic XO sufficient or do some other issues affect phase? eg Does the horn accelerate the wave or does the rear chamber size add some delay ?

"M" factor-
Do you mean the mass loading caused by the horn?

Which would you recommend in terms of SQ, an "assume the worst" design, or something more middle of the road?
 
Greets!

You're welcome!

OK, I guess we'll have to agree to disagree then since a ~flat in-room response is one of the prerequisites to being able to ~accurately reproduce a waveform.

If the acoustic XO point is low enough to be down in our falling hearing acuity BW then the comb filtering isn't an issue, but 1200Hz?! Back when my hearing was still pretty good I had a ~250Hz limit, though many others say up to 400Hz is acceptable, so as always YMMV.

WRT to phasing, yes you have to take pathlengths into account so that at the XO point both are in-phase, i.e both are pulsing together, but understand that the rear wave has started at some point behind the diaphragm and traveled an additional distance to get back to the front, so in time it's waaaaay off, increasing with increasing frequency.

BTW, the Fc is usually low enough that you have to take into account the room's effect on extending the acoustic length beyond the actual one, AKA end correction, otherwise it won't load as low as predicted.

Horns take a high pressure and let it expand to a low pressure, but on average it's at the SoS based on the air temp in the horn, which will be a bit higher due to friction and acting as a huge heatsink for the driver.

'M' factor, AKA 'T' factor defines the flare expansion rate, i.e. 0.5 = hyperbolic, 1.0 = exponential, etc..

If you read my recommendations on various subjects you'll see I'm very much a worst case scenario kind of guy since it's always easier to attenuate performance than add it after the fact, especially if low distortion is a prime performance goal. ;)

As always though, YMMV.

GM
 
GM,

Thanks again and I want to make sure I understand your recommendations:

Go with an acoustical XO in the say 300hz range. That will enable a lower possible cutoff for the bass horn and keep the potentially destructive comb filtering of the 2 sources out of the critical hearing range.

Stay with a low compression design that will help minimize the effects of differentials in the mass loading of the driver cone. While this sacrifices some potential efficiency, I think it makes wider operating bandwidths possible. It also increases the odds of achieving good results and will be more "tweakable" at the end because if I tried to go for maximum efficiency the end result could be bad and with way to fix it.

Now if the front horn is going to end up having an upper limit of 4-5kz, what will happen to the driver output above that? Is it going to be attentuated or just ignored by the horn and propagate the same as it would without a horn? ie would a super tweeter become mandatory or can I use the 2 horns efficiencies to offset the rising response inherent of the 206 ?
 
johninCR said:


Now if the front horn is going to end up having an upper limit of 4-5kz, what will happen to the driver output above that? Is it going to be attentuated or just ignored by the horn and propagate the same as it would without a horn? ie would a super tweeter become mandatory or can I use the 2 horns efficiencies to offset the rising response inherent of the 206 ?

I believe the treble will be ignored by the horn, the horn of course will be more directional, but I don't think it will effect the treble otherwise. I would think adding a super tweeter would widen your sweet spot, but wouldn't be nessesary.
Joe
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.