New Frugel Horns: Tuning Problems

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Just completed my Frugel Horn beta kits...first horns I have ever built. Put in a brand new pair of FE126e, stock...added 1/4" wool felt to the wall directly behind the driver, no other driver mods yet. Driving them with a signal source and SE 2A3 amp of known high quality.

First impression was not very good...male vocals sounded very "boxy," or hollow...as if singing through cupped hands. I pulled the baffles out and put a handful of Dacron polyfill in just behind the driver, in the compression chamber but not in the throat. OK...MUCH better now, but still sounds "boxy" compared to my FE-103s in a BR cabinets. Still, the horns sound promising...there is something"magical" about them.

What am I witnessing here? Is this a typical problem before tweaking a BLH? I am unfamiliar with horns, but I'll guess that baffle step is causing a dip before the low-pass of the horn starts kicking in, and this "dip" in response is giving me the "boxy" sound. Adding polyfill must have raised the LP frequency of the horn and filled in the dip? Am I on track here?

I ordered a pair of the "SupraBaffles," which in theory should mitigate some of the baffle-step effect. Will be trying them next week to see if it clears up the problem.

Cna anyone more experienced please comment on BLH horn tuning, baffle step effect and how one goes about getting all this tuned properly?

Thanks,
Dennis
San Diego
 
The baffle-step point of the Frugalhorn, sans suprabaffle, is 658Hz with 18mm side-panels. So you're right in guessing that there's a dip in the response before horn-gain kicks in. The baffles will solve that problem.

Give the drivers about 100 hours to loosen up. You'll be surprised at the difference it can make. The Frugalhorn controls the driver excursion far better than a reflex etc box, so it'll take longer for the suspension to break in.

Depends what volume chamber you've selected as it's adjustable in the Frugalhorn. The larger the chamber volume, the lower the horn's upper-cut-off will occur. So if you're running the full back-chamber volume, try reducing it by adding something (the plans suggest wooden blocks) to adjust the upper cut-off frequency of the horn.

While you're at it, I'd pull out the stuffing from the chamber, and try line the back, top, bottom and one wall of it instead with damping material like carpet underlay, which should kill any unwanted resonances in the chamber (adding blocks should help break them up too). If you feel it needs more damping, add some of the polyfil back into the chamber. If you find it over-damped, remove some of the lining material. And so on.
 
Hello,

that is a problem of material,
the same effect by Lowther and other horns they sound coloured,
because also the frequencies you don´t want getting
louder ~150 Hz, the bass gets weak because it is
covered by the freq. over 100 Hz.

Therefor I use WEICHFASER softfibre the first m of the horn to reduce the "Schallschnellen" over 4/lamda of 100 Hz.

http://www.hm-moreart.de/11.htm in the middle
and look my horns.

One desisive point is for a horn you need min. 1 mm Xmax,
I write it several times. Therefor is a simulation with AJ horn
helpful.
 

Attachments

  • material.jpg
    material.jpg
    46.1 KB · Views: 872
Interesting material. Thanks for the heads-up Horst. I'm not sure how easy it is to obtain outside Germany -I've never seen it, but I'll now be looking. I don't like too much damping in cabinets -loads of damping is just a lash-up fix for sloppy design IMO, so this looks ideal.

It's been a while since I looked at AJ Horn. Does the latest version now have facilities for simulating corner-loading, wavelaunch from the baffle, curved mouths and deflectors?

BTW -could you clearly set out your reasoning why a stroke 0.0394 inches is an decisive requirement of every driver in every cabinet? I've asked before, but you've simply pointed me to your Saxophon page, which frankly wasn't much use. I'm not trying to be unpleasent, just curious as to why you claim this figure to be universally applicable, and everything else is completely wrong.

Best
Scott
 
Scottmoose,

the material is made of the rest of woodworking,
very cheap, a sound reducing board, may be you get
it not in 12-13 mm with paper, but because it´s made by
waste, it must got everywhere.
Ask you joiner.

Please ask mr. Jost, I use AJ 4.0.

Remember we analysed this several times,
I make around 100 simulations in different
horns with different drivers, most need 0,8 mm
at 70 Hz and 0,5 mm ~150 Hz stroke for 1 Watt.
 
Cheers for that Horst, I'll keep an eye open for it.

Ah, now I see where the reasoning comes from. Fair enough for those boxes -it's not universally applicable though. Depends on the design and the room influence. I've done more sims than I care to remember in MathCad and not every box makes those demands. Here's a quick example I've got on the PC at the moment. 0.2mm 150Hz, 0.5mm 70Hz, total excursion (/2 for one way) at 1w input. It's for a BVR type of horn actually, so a long-path version with hyperbolic expansion should beat it.

To be honest, I reckon people get a bit hung up over Xmax sometimes. Best not to pass it if possible, I wholeheartedly agree -but so long as something really daft isn't being attempted, breaching it mildly isn't automatically going to ruin the sound. That depends on the distortion characteristics of the driver of course. Some are worse than others. The Fostex units are pretty forgiving in this respect.
 

Attachments

  • excursion example.gif
    excursion example.gif
    6.2 KB · Views: 775
Surely. The driver is the Eminence Beta12LTA 12in wide-band unit. The cabinet is a double-mouth (over-under layout) BVR style horn called Maria, that I'm working on with Dave at the moment, intended for for semi PA purposes. The plans aren't uploaded on the 'net yet, but it's about as simple as it gets, so shouldn't take too long. Great unit for the price actually. Not quite 'hifi' but a lot of drive-unit for the money.

The room has an influence on driver excursion if it forms part of the horn expansion (corner horns for example). So it needs to be taken into account if possible, or as close an approximation as can be done, as a 1/2 space or anechoic sim effectively is only simulating a part of the enclosure.
 
More Questions on Frugal Horn Tuning

Scott and others:

Thank you very much for your comments. I will withhold final judgement on the Frugals until my drivers break in and I install the SupraBaffles. They are already sounding rather nice at this early point, so I look forward to getting it all sorted out.

A few additional comments, questions:

1) My compression chamber is the smallest one, 2.356 liters.

2) Scott, I understand why you recommend lining the CC with felt, but why should I leave ONE side unlined? What does this achieve?

3) I have my horns placed about 1-1/2 feet in from a back wall, but far away from any corners. Bass extension appears to be less than my FE103s in a vented cab. I have my sub rolled in at approx 80 Hz, and it might need to be rolled in even higher. In short, the horns are basically unlistenable (in my opinion) without some form of additional bottom help from a sub.

Have others found bass extension to be rather minimal? I wonder what the addition of the rear reflectors might do...does it provide an considerable, easily audible extension, or should I not bother? I mean, it is easy enough to adjust the crossover knob on my subwoofers to give me what I want.

4) Scott...if you are willing to please do another simulaton, I'd sure like to see the frequency response of the Frugal Horn/FE126e combo...with and without SupraBaffle, and with and without reflector if that is possible to model in your program.

Again, thanks everybody. I will be sharing all of my future experiences with the tuning of this speaker.

Dennis
San Diego, CA
 
hm said:
I didn´t have the data of the horn.

Scottmoose,
I need
Horn throat,
horn mouth,
length
Your explanation show your horn is very very large?
I remember Freddy done a simulation.
The constructor must have a real measurement,
if he don´t, think about why not.

I agree about the measurements. Thing is, not all of us are in the happy position of having the time, facilities and money to be able to build & measure everything we design. I had to sell almost all my kit to fund my PhD. :sad: Fortunately, Martin's MathCad worksheets are proven frighteningly accurate, so this is as good as I can manage at present.

For the Eminence box, it rather depends on what you call 'very large'. ;) Actually, it's tall, but the rest isn't as big as you might think (as I say, keep in mind that this just happened to be a box I had to hand & it's partly intended for PA).

OK, what you need to simulate is a large BVR style enclosure 320mm wide and 362mm deep. The chamber is 1219mm tall. IIRC, AJ Horn isn't capable of simulating the two horn vents used, so sim them as a lumped parameter, throat CSA 147.2cm^2, length of 40cm, mouth CSA of 2438.4cm^3

Remember when simulating the Frugalhorn that anechoic or 1/2 space simulations are of only limited value, mainly for development purposes, because they don't account for one of the main parts of the enclosure and / or it's expansion (namely the room). Also, AFAIK, the only software that can model the wavefront pressure, expansion & thermal issues, and the effect of the curved mouth is Ron Clarke's.
 
Scottmoose,

you kiding us, we speak about a 5" driver and the
nessesairy stroke at 1 Watt and you show us a
12 " driver in comparison, this is like
compaire apples with pears.

I get this data for the F-horn:
PChamber 5,4 L
L = 2 m than the expansion is to large to take it with the horn
Throat 0,00266 qm
mouth at 2 m : 0,0224 qm

please proof this from the constructor.

The simu looks worse, sorry, the movement between
45-80 Hz is ~0,8 mm + 100-200 Hz from 0,7 down to 0.4 mm,
the driver has a Xmax of 0,35 mm,
I never will construct it like that, so better ask the constructor
for the right data and measurements.

The simu is near the Stereophile measurement.
 
Horst:

Thanks for the sim information. Can you please re-run sim with compression chamber volume at 2.35 L? The F-Horn designer is no longer recommending the large chamber, but to use the smallest (2.35 L) chamber...this is what most of us have in our F-Horns.

Will this change the Xmax?

-Dennis
 
I can't read your mind Horst (wouldn't life be easier if we all could? ;) ). Your statement read: "One desisive point is for a horn you need min. 1 mm Xmax, I write it several times." You didn't make any stipulation about driver size, and knowing that you regularly repeat the same thing (or similar) for drivers of 8in too, I assumed fairly naturally that you applied it to everything, and gave a quick example which, as I said, I simply happened to have to hand. I see now that you obviously don't apply 1mm stroke critera to every driver after all. However, this has gone far enough for the moment. We're mustn't hijack poor Dennis's thread in a discussion about Xmax, or about the design minuitae of the Frugalhorn.

To answer your question Dennis, it comes from what I do with TQWT, pipe-horns and MLTLs -a very effective proceedure I learned from GM. One side remains un-lined, to give a bit more scope for tuning. Most of the time, it's not necessary for both to be damped. It should help to supress resonances -cuppy effects on vocals are also often associated with a chamber that's a little too small, so it'll help with that too.

If you can, get your cabinets into the corners. They're corner horns; the room actually forms part of the cabinet design & the last stage in the horn's expansion. This is a big reason they simulate and measure in anechoic conditions so poorly: the final, and major stage of the horn isn't included in the sims or measurements. Sadly, this fact is rarely appreciated, both for the Frugalhorn, and many other corner-horn cabinets, and the implications of this are missed. What difference does corner positioning make? The resistive portion of the acoustic impedence will increase by a factor of 8, as will mouth size, and the wavefront expands conically as it moves from the corner into the room.

The deflectors will help, in corners, or if corners are not available, near a wall. Again, they don't just improve LF extension and gain, but also in continuing the expansion of the cabinet, they improving the wavefront linearity. The Frugalhorns are never going to be bass champions -a 4 1/2in cone can't be, so the subs are helpful, but this will improve things greatly, both in quantity and quality.

Hope that's of use
Scott
 
Diogenio said:
Horst:

Thanks for the sim information. Can you please re-run sim with compression chamber volume at 2.35 L? The F-Horn designer is no longer recommending the large chamber, but to use the smallest (2.35 L) chamber...this is what most of us have in our F-Horns.

Will this change the Xmax?

-Dennis


One of the original design intentions of the F-Horn design team was to allow for flexibility of experimentation with a wide range of drivers - attendant to that would be the requirement to adjust volume of the chamber.

There is of course much debate as the correct methodology for determining exactly how to determine that, and the degree to which X-Max may be a "crucial" factor in a design of this type. For example, of the 3 systems in my home, only one enjoys a room of over 2700 ft^3 (76m^3), and with sensitivity of 93dB @1W@1M and my personal listening habits of under 86dB and listening position of 2M ... well I think you can see where I'm going - this ain't no disco, and I plan on retaining what's left of my hearing for at least another decade.

Back to the matter of adjust the CC - during our experiments, one discovery was that using irregular shaped solid objects (blocks of wood or a small can of paint!) seemed to help break up some of the standing waves that can occur between parallel walls of the CC.


and of course ditto to everything Scott stated above - in corners, with the rear deflectors and supra baffles, these are very musically satisfying, but will never blow out any candles - for that you'd need a pair of Bruce or Loreena (actually, she's more likely to light my candle than put it out!)
.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.