Collaborative Tapped horn project

[Patrick Bateman]

Originally posted by GM
Greets!

Didn't have any, I mean I haven't heard the CB in years and my acoustic space is maybe 1/20 what the TOPs are in. WRT SQ, I was very disappointed, but I'm not familiar with either of the pieces used to demo plus there were lots of other speakers, tapped horns scattered around in close proximity to act as various BW passive radiators, so I'm sure the system has much more potential than I heard.

Isn't anyone going to comment on this?

In my personal experience, dual reflex bandpass is very VERY hard to get correct. And the tapped horn has a lot in common with dual reflex bandpass. Subjectively I've found that dual-reflex designs sound 'sluggish' and 'out-of-step' with the mains. Whereas a single reflex design can be integrated without much fuss, as long as the levels, the response curve, and the timing is fairly accurate.

Has anyone heard a lab horn AND a tapped horn? Is a lab horn more 'musical?'

[Patrick Bateman]

Originally posted by cowanaudio
G'day again rick

This plot gives you an idea of what some acoustically absorbant material will do at the velocity maxima at 75Hz. Red is undamped. You reduce the dip to just 8dB, but lose 6dB of sensitivity in the process. My conclusion was that it is not worth doing.

Thanks for posting graphs! I'm personally interested in using the "tapped horn" concept for the midranges in a small unity horn. So your experiments in damping and controlling the dip will save me some time

[rick57]

"tapped horn has a lot in common with dual reflex bandpass"

Agreed bandpass can be out of time, didn’t like the one I heard

I thought tapped horns are more like a Tapered quarter-wave pipe?

[Patrick Bateman]

Both a dual reflex bandpass and a tapped horn utilize the rear wave to augment the overall response. A front loaded horn, like the lab horn, does not.

A transmission line uses the rear wave to augment the front wave, but the rear wave is significantly damped. So it's contribution to the overall response is much less dramatic.

In a tapped horn, it appears that the rear wave is just as prominent as the front wave, which is why the efficiency is so high.

[paulspencer]

Patrick, it does concern me a little that GM wasn't impressed with the SQ, but not being there or seeing the setup and the source material, it's hard to make a secondhand judgement. What concerns me more is that there is a dip on the midbass that seems unavoidable, although Danley seems to have pulled it off.

Could it be that there is some special trick to do it that William didn't find? Or could it be some unobtainium driver?!

GM, wow!

40kw x 40 drivers and 105 db @ 250m!

I wonder what they actually use it for!

One thing I've noticed is that the midbass dip actually moves higher as you reduce the length. One possibility would be to make two tapped horns of different tunings and attempt to get them to further balance out the other dips. I wonder if this could work.

Another idea - what if a second driver was used and placed higher up - it would be like a second tapped horn tuned higher.

William did you try anything like this and do you think it could prove an answer?

Ideally I'd like to get this thing to be useable up to 80 Hz.

Ok so here's the idea:

We have a 1.8m and a 2.4m tapped horn - this is what it looks like:



Notice that the upper bass dips are pretty well cancelled.

[rick57]

> it does concern me a little that GM wasn't impressed with the SQ

That was the opposite to WC’s more extensive experience . .

> I'd like to get this thing to be useable up to 80 Hz.

Me too originally, not rally now; what are you likely to want to integrate it with, that 15 Hz is significant?

OTOH, it was suggested to me that
“as there is no damping material, I would expect midrange sound waves from the back of the cone to reflect back from the untreated and parallel surface of the box onto the cone and be heard at the front of the cone.
At high volume levels one might experience standing waves.

Also, the driver being buried, as it were, inside the box would give rise to a beaming effect as the frequencies increase.”

Operating below 100 Hz, I think these are both non-issues, anyone think they are?

Cheers

[cowanaudio]

G'day Paul

>Another idea - what if a second driver was used and placed higher up - it would be like a second tapped horn tuned higher.

>William did you try anything like this and do you think it could prove an answer?

We tried this with two drivers in the 30 Hz horn, and it didn't help. Here's a picture of one of the spaced driver test boxes, in the other the top driver was 1/2 way up the line. I'll have to look for the plots. They were nothing special.

Cheers

William Cowan

[paulspencer]

Oops, here's the pic properly linked:



Beaming?!!!!!!!
This is a subwoofer! Now if we could get more directivity in its bandwidth, that could in fact be an advantage rather than a problem. Of course, dipole is the only way this is going to happen.

The other issue is also irrelevant to a subwoofer. This would be no different to any other subwoofer. Sound inside any box speaker will get out through the cone. In most speakers this is undesirable. In this one, the sound getting through the cone is a much lesser proportion. We are talking about a subwoofer here, which makes it pretty much irrelevant anyway.

In my personal experience, dual reflex bandpass is very VERY hard to get correct. And the tapped horn has a lot in common with dual reflex bandpass. Subjectively I've found that dual-reflex designs sound 'sluggish' and 'out-of-step' with the mains. Whereas a single reflex design can be integrated without much fuss, as long as the levels, the response curve, and the timing is fairly accurate.

I don't think there is a lot of similarity with an 6th or 8th order bandpass. Tuning is based on line length like a TL, hence it should not be difficult to get the right tuning. I believe it isn't finicky in that way.

Some argue that undersized horns behave like bandpass more so than true horns. Still, from a design point of view I think a tapped horn is very different.

Has anyone heard a lab horn AND a tapped horn? Is a lab horn more 'musical?'

I'd also like to know if anyone has ...

A transmission line uses the rear wave to augment the front wave, but the rear wave is significantly damped. So it's contribution to the overall response is much less dramatic.

The damping is to tame the resonances and TLs are normally used for midrange as well. This damping should not reduce bass output unless overdone. As I recall from playing around with Martin Kings simulations, if the density isn't too high, the resonances will be damped without reducing bass reinforcement from the line.

In a tapped horn, it appears that the rear wave is just as prominent as the front wave, which is why the efficiency is so high.

I believe it also allows it to behave as a true horn with a much smaller size by operating as a 1/2 wave horn one octave higher than its 1/4 wave operation which determines low frequency extension and output.

I think the reason for the upper/mid bass dip is the same reason you get this problem with a BLH. I wonder if I should actually be modelling these as a BLH instead.

William ....

Damn, it didn't work!!!!!

Ok but could that be because of two drivers in one box creating other problems? I'm curious what would happen if you were driving the offset driver with the bigger tapped horn both measured at the same time. My chart does suggest that if the tuning is different enough that the other peaks and dips should cancel.

One more thought. In the Danley version, does anyone know how the driver is actually mounted? It has an access panel in the middle - could it be that the driver is actually in there? It would seem a strange thing to do ... or does that panel serve another function?

[GM]

Originally posted by paulspencer
One thing I've noticed is that the midbass dip actually moves higher as you reduce the length.

Greets!

Right, this is the pipe's acoustic 3rd harmonic, so as the acoustic path-length is shortened, the dip must rise in frequency and if I understand TD's white paper correctly, it's what gets filled in by the driver's anti-phase output, putting the first major dip at the 5th harmonic somewhere above a sub's passband.

GM

[rick57]

Originally posted by paul

One thing I've noticed is that the midbass dip actually moves higher as you reduce the length.

Originally posted by GM

as the acoustic path-length is shortened, the dip must rise in frequency and if I understand TD's white paper correctly, it's what gets filled in by the driver's anti-phase output, putting the first major dip at the 5th harmonic somewhere above a sub's passband.

So the trade-off is upper vs lower cutoff?

GM

IIRC you said the ToP is effectively a 6th order BP.
Would you also sat that with Patrick’s view that it’s *dual reflex bandpass; and that
“dual-reflex designs sound 'sluggish' and 'out-of-step' with the mains.
Whereas a single reflex design can be integrated without much fuss, as long as the levels, the response curve, and the timing is fairly accurate.”

Cheers