Wolfhorn SDX Design Finished - Page 3

Did the DIY spud thread get nuked? There were a bunch of new posts in it this AM and I was going to read them over lunch... then it was just... "gone".

I'm curious what you mean by this since there's no sealed rear chamber per se in a TH and AFAIK there's only one way to get any measure of reactance annulling in what is otherwise a BLH which design characteristics I see none of in yours or any other DIYers designs to date.

GM

Hmm, while moving in the right direction TH design wise, I don't believe this conforms to any sort of reactance annulling, at least not as I understand it.

Here's the sort of TH response (1 W/2pi) I consider at least somewhat reactance annulled using a theoretical 12" driver for a stillborn PA app that Eminence or similar could easily make, i.e. no 'unobtanium' or high $$$ component spec requirements. Indeed, for all I know there's now a driver with close enough specs as this is a several year old design now:

GM

OK, I feel I should expand on the Throat(acoustic) Impedance a little here.

There are two forces at work here, and the problem one is Reactance. This is the mass of the air in the horn working against the driver. In HR it is the red trace. The second effect is Resistance, which simply means the horn and driver are working together. Black trace.

It may be best to think of it in terms of a Slinky. Most everyone has bridged one across their hands at some point it time. Remember how you had to move your hands at just the right speed to get it to send the ripple back and forth from hand to hand? That is like resistance. The crap you got when you messed your timing up is like Reactance. The weight/momentum of the spring was not liking all that bouncing you were doing.

Ok, with that out of the way...

Here is a pic showing the throat impedance of an "ideal" hyperbolic horn.
Notice the scale.

The horn was setup with RA the Leach way. This involves tuning the rear chambers volume. The idea is to match the Reactance with Resistance at the cut off, to cancel the Reactance with Resistance.

Now an issue is that at these sizes both effects are so tame that it doesn't really matter. Lets' take 90% of the throat size away, and re-adjust the cutoff to match to see something a little more familiar.


Now a horn without a rear chamber can't really work the same way, as it has no rear chamber. Easy enough. The trick is to find where the drivers suspension does as similar job.

Here is a TH straight off my sheet where I have matched things up with the right horn/driver combo.


Here is the same TH, after a very minor tweak in HR.
Of course I designed the sheet around my methods.

Here is the results of the Nd 1/4WL horn straight of the sheet.


I am no expert, and I'm sure GM will correct me if I have misspoken, but this is how I see it right now. As GM said though, the DIY horns that mimic this (that I know of) can be counted on one hand... two.

Hope some of this helps.

EDIT: make that three. two were mine.

OK, still don't believe what you're doing is reactance annulling, at least not enough to call it as such, but I'm no expert either.

Don't recall seeing any of your designs posted with measurements. If so, links please.

GM

Reactance annulling is a three dollar word used to describe a simple phenomenon. When you horn load a driver the acoustic load of the horn adds to the Mms of the driver, resulting in system resonance far lower than the original Fs of the driver, sometimes as much as an octave lower. A horn works best near Fc if the system resonance is close to Fc. With a rear chambered horn you can make the chamber small enough to push the system resonance back up to where you want it. The results can be most easily be seen on the electrical impedance chart. With no rear chamber you can't reactance annul per se.

It might be a little too much.


I was in no way implying that my examples were best case scenarios.

I picked a random driver known to be used in a DSL TH, and the sims were based on real world possible uses in a sub. As a result the HC needed to be pulled down to ~220 to give a survivable compression ratio for the "ideal" horn. I also used the standard OD position for the front tap, and ran it down until the low 20s.

The idea is the get the resistance, and reactance to cancel out at the cut off. If I put the front tap at the throat, and altered the compression and flare it will spit out a graph similar to yours.



It's semantics really, as TD says otherwise.

Agreed.

No, not if we think more in terms of how we do it instead of why we do it. From this POV we have to find a way to mimic the rear chamber's function of balancing out the acoustic loading on both sides of the diaphragm and at least in a sim using Prof. Leach's math I've accomplished this, so lacking any more technically correct term for this type of BLH, TH alignment, 'reactance annulled' works for me.

GM

And I in no way implied they were, only that the plots/explanation you presented to support OW's understanding that one only has the driver's suspension to work with doesn't meet my understanding/criteria of a BLH or TH equivalent of a reactance annulled FLH, so not sure what your point is, especially now that you've agreed with me that to get reactance annulling in a BLH, TH requires a considerably different alignment than conventional horn design wisdom dictates.

GM

Yeah, especially for designs that are intended for deep down in the room dominated BW and especially down below our acute hearing BW and if you don't have a map of the room's acoustics, then it's a real crapshoot as to whether what you design can be made to perform well; though in recent years the huge advances in relatively inexpensive digital room correction has made it mostly moot beyond ensuring the design has a low enough cut off and power handling for the app.

With today's single digit BW and need for up to near live dynamics capability, multiple critically damped subs spread around the room in 3D and tweaked digitally is where it's at for me if extensive passive room correction isn't an option.

Yeah, horn design is 'simple' like a fully articulated/adjustable independent suspension is 'simple', the basic concept is simple, but the 'Devil's in the details' to get it to perform at peak efficiency for the intended app.

GM

Ok, I just finished this, and am afraid it reads like a argument. I'm going to go back through, and try to work that out, but I wanted everyone to be sure that this is not my intent. I am not the best at getting what is in my head across in print. My intent is to throw as much out there, so a common point can be found.


That was not exactly my intent, it was more along the lines of I could make a bunch of graphs that looked like yours if I didn't do a realistically sized sub horn, and built a midbass horn instead with something like an EPS15-500. It's all about relative size.
See example:


As the size of the horn shrinks compared to the ideal, the throat peaks will get larger, and the electrical ones will be harder to suppress as well. It doesn't means that reactance annullance isn't taking place anymore, to me. If you look at the Nd horn traces from the other post you can see this pretty clearly.

I don't think a TH requires a considerably different alignment than conventional horn design wisdom dictates to get RA going. Now you can do it unconventionally, and I think I have figured out what you are doing, but playing with things myself, it seems like you can do it with normal designs, but I haven't worked it out 100% yet.

Here is an EPS15-500 straight off my sheet with only L12 reduced to .10, at 443l.


Here is the sim altered up to 750l, and 2.5:1 comp. At this point the mouth has become large enough that the tap really isn't necessary anymore, but it is there for them all.


Here it is again at 1500l, and 2.5:1 comp.

I could not get it any better from here without raising the compression a good bit.

It's all about scale, to me. Here is the same horn crammed into 150l.


If you use the HR sample tool on the Acoustic Impedance the relative location of the Reactance & Resistance are pretty close to each other, except for the peaks.

So at what point is it not reactance annullance anymore? I would really like to nail this down. If possible.

cz, how deep?

Wolf,

How are you calculating path length error?

JSS

How do you go about measuring to determine the resonance frequency of a flh's rear chamber? I just finished mine but don't know how to actually do the measurement. thanks

Any updates on this endeavor?