Collaborative Tapped horn project

G'day Paul

What a great idea! It's fun playing with tapped horns, and not too hard coming up with a good result. I'd join you in a heartbeat if I didn't have to drive 16 hours.

This will certainly save having to purchase the drivers yourself. All the drivers I tested were my own. I assume the baffle the driver mounts on is removable, so you can modify it or replace it to adapt different drivers? The bottom layer of the baffle could have a clearance hole for a 15" driver and remain fixed and the driver baffle could then be screwed to that.

What is the length of the line? Also it might pay to mount the drivers the other way around for more space behind the magnet.

The 830515 was very poor in my 30Hz tapped horn. The 830500 was a good performer.

Measurements are good if you can do them.

Cheers

William Cowan
 
Hi Paul
The PHL I measured IIRC as 390 mm centre – centre, can check on the weekend.

William’s suggestions were in one of the threads here, I search I think on ‘tapped’ and ‘Vas’.


Hi William,
Peerless– on the 5th graph at www.diy.cowanaudio.com/th.html you got more deep bass with the 830515 than the 830500 - -did you prefer the ‘top end’ of the 830500?
Is the left hand vertical line on those graphs, 10 or 20 Hz?

Thanks
 
G'day Rick

I didn't notice this until now, but those graphs are labeled incorrectly. The red trace is the 830500. Many people must have thought I was crazy preferring the brown trace! The red 830500 trace is very easy to eq flat, with only one channel of PEQ. I will change the page ASAP. That plot is 20-200Hz.

Moray, I'm not sure how the Clayman design would work, I havn't played with one.

Cheers

William Cowan.
 
I tried to look up Klayman’s 1962 patent
http://patimg2.uspto.gov/.piw?Docid...&SectionNum=&idkey=NONE&Input=View+first+page
but it doesn’t allow access or display for me.

Hi Moray
Would the extra half octave be at the top or boo tom end? Can you post the drawing?

Hi William
At one stage you suggested Qt of 0.2 - 0.4, FS near LF cutoff, and
VAS low. We mightn’t have time to properly test all - any further thoughts on optimum TS specs?

The 830500 is Qt of 0.17, FS of 18, and VAS of 139.

Any thoughts on how Qt of 0.35, FS of 23, and VAS of 144 the much cheaper but well regarded CSX might be?

Cheers
 
The plot is 10 to 100Hz.
 

Attachments

  • 18hzth.gif
    18hzth.gif
    12.5 KB · Views: 25,428
To keep the other thread focused on the day I'm organising, I'm starting this thread to talk about the technical aspects of this tapped horn project. The thread to discuss the day is here:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=97620

Here is a simulation:

An externally hosted image should be here but it was not working when we last tested it.


Yes, it looks bad for the tapped horn, but hornresp is not simulating the full story. A tapped horn such as this is a 1/4 wave horn at 18 Hz which is what hornresp shows, but at the same time it is a 1/2 wave horn at 36 Hz. This is what hornresp doesn't show. With the correct driver, the result is that the response is flat and the dips are filled in.

The tapped horn has only one driver while the other horn has two. Note that the voltage input is the same for both at 30V.

How can it be both a 1/4 and 1/2 wave horn? The front of the driver loading into the throat gives the 1/4 wave output, while the rear of the driver located in the mouth actually sends its output towards the throat and then back again.

I'm far from understanding it fully, hence this project to try out different drivers and see how what we can measure is different from the simulation.
 
compression ratio

Hi William

When philpope in the discussion at http://www.speakerplans.com/forum/f...?TID=8329&PN=10
said

“I think when William said 2:1 compression ratio, he meant the ratio of Sd to the end of the horn, NOT the cross section where the tap is.
I assumed that the lowest bass response would be obtained by tapping the driver into the horn where the cross section is the same as Sd so I make the throat of the second section in the model equal to Sd.”

Was that what you meant?

Cheers
 
G'day Paul

Looking at your sim, I'd guess you'd be about 4dB down at 22Hz. That is a good compromise for reduced size. The tap will smooth out the dip above the low cutoff. How about you sim the different drivers being offered for the comparison?

Rick

The compression ratio is the area ratio between the driver and the throat of the horn. If your driver Sd is 800CM^2 then if the throat is 400CM^2, you have a 2:1 compression ratio.

Cheers

William Cowan
 
I'm collecting driver data so that I can simluate them all.

Drivers:
AE speakers AV12 2nd run
Rythmik DS12TC
Peerless CSX 10 & XLS12
PHL 5230
Eminence Kappa pro 12
Eminence Alpha 15
Lambda 15TDM
JBL 2035

For the last 2 drives, I don't have parameters for CMS, RMS and MMS which are necessary for hornresp. A quick search didn't turn up anything. Does anyone have data on them?

So far the AV12 parameters look very good according to the target range. It meets all criteria except being slightly under recommended SD (not a big deal). I have found it also models very well in a bass horn generally. Not surprising since it is similar to the Lab driver which was designed for bass horn use.
 
Hornresp simulation

An externally hosted image should be here but it was not working when we last tested it.


AV12 - red
PHL - green

The other 2 drivers are the Alpha 15 and the Kappa Pro 12. The kappa is very similar to the PHL except that it has a peak at the bottom end due to the much stronger magnet. The alpha has the extreme low peak due to the crazy high Qts and very weak magnet, but the dip is the biggest. The XLS not shown as it is fairly similar to the AV12. The Rythmik is similar to the AV12 except that its bottom peak and dip and less. CSX is similar to the Rythmik but less efficient.

Of them all, I suspect the AV12 will work out the best. My gues so far is that if the magnet is too strong, and hence the Qts is too low, the end result will be drooping bottom end.

Simulation is for 1w of input in a corner.
 
G'day Paul

Hornresponse will let you back calculate the values you are missing. Enter all the values you have and double click on the empty field where the unkown value should go. It will work it out for you.

It looks like the AV12 will work well in this tapped horn. I hope you have enough power available to explore it's limits. An Aussie Monitor AM1600 could not give an 830500 any grief before going into clipping.

Cheers

William Cowan
 
William,

I seem to find the AV12 to be very good for bass horn use. I have an Europower EP2500 to power it, which has 2 x 650w into 4 ohms. To drive just one in a tapped horn, I have 2.4kw bridged into 4 ohms (yes it's stable down to 2 ohms). Do ya think that's enough?! :D

What REALLY surprised me the other day - I saw it clip when driving them in OPEN BAFFLE! :bigeyes:

It was just once for a brief moment. That's only the 2nd time I recall in a few years that I've had it. Before that amp I had tried out two 2x450w pro amps and both of them clipped before the subs ran out of steam for music. Of course I use a rumble filter ... when I upgrade my mains to keep up, then they will have more of a challenge.