Stage 2

[Buckle-meister]

Some of you will know that I recently purchased RealTrap acoustic absorption and reported my findings in a review.

For the treated room's low-frequency plot, it seems clear that a room mode is wreaking havok at about 38Hz (a quick calculation returns the first axial mode for the room as being centred at 40Hz). To make matters worse, the room is square in plan so the first axial mode ocurrs in two directions simultaneously!

Anyway, before I run ETF to properly setup my sub (used in conjunction with the front towers) to obtain the best low-frequency plot I can, I thought it'd be best first to construct a Helmholtz resonator (or two) to alleviate the damage being caused by the aforementioned mode and then place the sub accordingly; my logic being that building the resonator first will let me see 'what I've got' to work with, with the sub.

After all that, my question is:

I have read this paper by Audioholics and understand that varying any two (or more) of the variables in the equation to calculate the tuning frequency would still allow the same tuning frequency to be obtained (analagous to 2+3=5 vs. 6-1=5). What then would be the difference between having a certain sized absorber tuned to 38Hz and a larger absorber, but that also was tuned to 38Hz? Is it simply that the larger the size of the absorber, the more effective at absorbing it becomes?

 

[mtrycrafts]

What then would be the difference between having a certain sized absorber tuned to 38Hz and a larger absorber, but that also was tuned to 38Hz? Is it simply that the larger the size of the absorber, the more effective at absorbing it becomes?

That would be my input, more effective. Low frequency is not easy to control. In anechoic chambers that can handle the lows well, you would see wedges several feet long, if not more.

 

[WmAx]

I thought it'd be best first to construct a Helmholtz resonator (or two) to alleviate the damage being caused by the aforementioned mode and then place the sub accordingly; my logic being that building the resonator first will let me see 'what I've got' to work with, with the sub.

A real stage 2 should consist of parameteric equalization. At this low frequency, it's acting almost entirely as a minimum phase phenomena in relation to the room dimensions. Nothing need be constructed or even moved. A properly set up parameteric filter will remove your resonance, at the listening position, completely. Since you are apparently used to/comfortable with using ETF, I would expect you would now be more confident in using a parameteric equalizer(which you did not seem comfortable with in the past, based on my memory).

-Chris

 

[Buckle-meister]

At this low frequency, it's acting almost entirely as a minimum phase phenomena in relation to the room dimensions.

Can you elaborate please?

Nothing need be constructed or even moved. A properly set up parameteric filter will remove your resonance, at the listening position, completely.

Everything hinges on that innocent little word 'properly'. If a PEQ is capable of removing the resonance completely, are Helmholtz resonators never used nowadays? Are they obsolete?

Since you are apparently used to/comfortable with using ETF, I would expect you would now be more confident in using a parameteric equalizer (which you did not seem comfortable with in the past, based on my memory).

Your memory serves you well. I am comfortable using ETF, though freely admit that I don't understand all its functions in depth. My reluctance to using a PEQ can best be summed up by...you:

...products such as the the Behringer DCX2496 allow advanced crossovers, room correction and even driver dynamic behavior correction, within an affordable price range... with a catch. The disadvantage is that one must know how to measure, analyze and configure the setup manually and have the appropriate measurement equipment and analysis software to accomplish this feat. This, along with some other minor set up quirks [such as reducing the signal voltage output to feed standard -10dbU inputs, for example] that must be addressed when using professional audio equipment in a consumer application seem to alienate most people from the idea when I suggest it.

To which Sleestack replied:

That's a pretty big catch. I'm definitely not up for the learning curve involved with that, although it seems like it would make for a fun project from someone more technically inclined.

I do consider myself technically minded (at least in my own field), and would certainly consider implementing a PEQ a 'fun project', but although treating my room has changed the frequency responce at the listening position, with the solution being passive, there's never been any chance of damaging my equipment. With parametric equalisation being an active solution, I'm just a little nervous about the (I feel) real risk of my inadvertantly damaging either my amp or speakers (or both) from messing with settings. Especially at the bottom end of the frequency range, where speaker drivers are working the hardest anyway.

 

[WmAx]

Can you elaborate please?

38 Hz is about 30 feet in length in air. The half wavelength is roughly 15 feet in length. This frequency is not going to have much complex interaction in the room because of it's large size, it's going to basicly be a simple resonance based on a major room dimension. It is described as minimum phase because it will, for the most part(it won't be entirely minimum phase, because some multi-point behaviour is still an issue, but it is not to a large effect), be a resonance that's amplitude magnitude is directly related to it's phase angle. This is because that essentially[and for the purposes here] the resonance is coming from a single source(1 large room dimension, the parrallel wall space). If the frequency was much shorter, thus also had the opportunity to reflect from multiple intersecting points that exceeded a 1/4 wavelength difference of the frequency, then it would no longer be essentially a single point of source resonance. Equalization would be very difficult in this case and cause as many problems as it solved(as an extreme example, a 100Hz resonance in your room would have substantial non-minimum phase behaviour).

This being said, it is critical to properly identify a resonance's center frequency and bandwidth(Q), otherwise the equalization will worsen the response. It is usually easy to measure and calcuate these, but it can be further complicated if you have, for example, significant null affecting one side of the resonant peak, thus making calculation more difficult. But in this case, you can use simple trial and error, where you change the value slightly in increments and remeasure after every change until you find a suitable value. Nothing difficult.

Everything hinges on that innocent little word 'properly'. If a PEQ is capable of removing the resonance completely, are Helmholtz resonators never used nowadays? Are they obsolete?

Certainly Hemholtz resonators are still used. But I don't see a dinstict advantage to one in your particular case at the very low frequency resonance that you specified. The equalizer will easily handle this frequency, and in addition, if you change rooms in the future, the equalizer can be re-accomodated with a quick reset to the new target frequencies. The Helmholtz resonator works over a very narrow band and will not likely be useful in the next room/set-up.

but although treating my room has changed the frequency responce at the listening position, with the solution being passive, there's never been any chance of damaging my equipment. With parametric equalisation being an active solution, I'm just a little nervous about the (I feel) real risk of my inadvertantly damaging either my amp or speakers (or both) from messing with settings. Especially at the bottom end of the frequency range, where speaker drivers are working the hardest anyway.

For one, it's easy to insure that you never cause damage by simply not ever using much in the way of boost, which you should not be doing anyways in corrections.

But let's assume you ignore this, and add significant boost levels to your bass. To damage the speakers, you would have to meet a complete list of critierion. You would have to meet ALL of these qualifications to risk damage:

1. Use a substantial boost of equalization.
2. Have this boost coincide with a bandwidth where the speakers were operating near their limits initially.
3. Use the speakers at high SPL continuously playing content that has significant energy in the frequencies that were boosted.
4. Ignore the obvious audible distortions that will be created in this case when you are bottoming out your woofer(s) continuously or clipping your amplifiers continuously.

It seems highly impropable that you would fulfill all of the above conditions by 'accident'.

-Chris

 

[Buckle-meister]

Certainly Hemholtz resonators are still used. But I don't see a distinct advantage to one in your particular case at the very low frequency resonance that you specified. The equalizer will easily handle this frequency, and in addition, if you change rooms in the future, the equalizer can be re-accomodated with a quick reset to the new target frequencies. The Helmholtz resonator works over a very narrow band and will not likely be useful in the next room/set-up.

The major advantage I see of the home-built Helmoltz resonator vs. the PEQ is the cost.

I understand about the Helmholtz resonator being room specific (unless of course I move to an exact sized and furnished room, but that is hardly likely), but I had vague plans if I went ahead and built one of allowing the port portion of the absorber to be replaceable. As far as I can see, the equation in the paper I linked to allows one simply to adjust the length of the port to suit whatever frequency is problematic in a new room in the future.

I'm a bit dubious about my reasoning though. I mean, if I am right, then one could make an infinite number of differently sized Helmholtz resonators (but of identical volume) with corresspondingly different diameters and lengths of ports etc, all tuned to an identical frequency and all would be equally as good as the other. I find this hard to believe, hence the questions.

To damage the speakers, you would have to meet a complete list of critierion. You would have to meet ALL of these qualifications to risk damage:

1. Use a substantial boost of equalization.
2. Have this boost coincide with a bandwidth where the speakers were operating near their limits initially.
3. Use the speakers at high SPL continuously playing content that has significant energy in the frequencies that were boosted.
4. Ignore the obvious audible distortions that will be created in this case when you are bottoming out your woofer(s) continuously or clipping your amplifiers continuously.

You see, the way I see things:

1. I have no idea (apart from by the sound which by then would be too late) what the limits of my speakers drivers are.
2. Don't want to worry every time I play music 'are there parts of this song which are coinciding with that boost I injected at x-Hz'

 

[Doug917]

Robbie,

Have you given any thought to adding the Velodyne SMS-1 PEQ to your setup? The SMS-1 allows you to view the settings it/you make via a video out on the unit to make sure you are not boosting anything too much. It has 8 bands of highly adjustable PEQ. It will also allow you to adjust the phase of your sub as well in increments of 15 degrees. You might want to give it a look here:

http://www.velodyne.com/velodyne/products/product.aspx?ID=15&sid=530i227f

Good luck!

 

[Buckle-meister]

The SMS-1 allows you to view the settings it/you make via a video out on the unit to make sure you are not boosting anything too much...

Thanks Doug, but I realised that one of my own comments is incorrect; I'd not be boosting anything period, so I guess the potential of inadvertantly damaging the amp and/or speakers couldn't occur as WmAx states.

I'll think on the matter, but it'd be nice for my questions about the absorbers to be explained anyway...

 

[Buckle-meister]

Apart from Mtrycrafts (thank you ), I've had no other responses that answered my question specifically (no disrespect intended to the other posters ).

Does nobody know the answer? Can this be so?

 

[JohnPM]

On your original question, the larger you make the resonator the greater its absorption. Basically analogous to adjusting the gain of a parametric filter, if somewhat less convenient The effect that is achieved by a Helmholtz resonator can be achieved with a single parametric filter, but the filter is considerably easier to tune. The main difficulty with the resonator is getting the Q right (though getting the port corrections and volume calculations just right to get the frequency you want can be a tad frustrating also). If the resonator's Q is too high it will add ringing of its own, too low and its effect will be too broad and the reduction in ringing less than desired. Ideally the Q must be matched to that of the mode it is correcting. I'm a bit dubious about your cost argument against PEQ as BFD Pro's are very cheap. Even if you consider the time you would have to spend building a resonator to come for free, materials are not, and there's bound to be some tool or other that you end up buying to make the job go a bit more easily

P.S. I was in Paisley last weekend, small world, eh?

 

[Buckle-meister]

The main difficulty with the resonator is getting the Q right (though getting the port corrections and volume calculations just right to get the frequency you want can be a tad frustrating also). If the resonator's Q is too high it will add ringing of its own, too low and its effect will be too broad and the reduction in ringing less than desired. Ideally the Q must be matched to that of the mode it is correcting.

Understood, and thanks. Disregarding the hours spent building a resonator (you can't assign a cost to something you enjoy), I still reckon the cost of the PEQ I'd be considering would far outweigh the cost of any materials/tools (which I may well have been able to pick up for free anyway). However, I appreciate that even playing about with port length etc to tune the resonator to the correct frequency would still leave me with having to sort out the Q with, as I understand things, the placement of a degree of absorption inside the absorber. For this reason, I must admit that I'm beginning to lean towards a PEQ. Unfortunately, unanticipated funding may temporarily prevent me taking this route.

P.S. I was in Paisley last weekend, small world, eh?

Indeed. Glad to see you made it out alive.

 

[WmAx]

Understood, and thanks. I still reckon the cost of the PEQ I'd be considering would far outweigh the cost of any materials/tools (which

Curiously, what is the identity of the particular PEQ that is prohibitive in cost that you refer to in the above statement?

-Chris

 

[Buckle-meister]

...what is the identity of the particular PEQ that is prohibitive in cost that you refer to in the above statement?

Why, the DCX 2496 of course. The unit is £165 in Britain. Assuming that I cannot obtain the necessary materials to construct an absorber for free, I suspect my outlay would be no more than approximately £20. Tools would not be a problem. Clearly the unit is more than eight times the cost of the absorber or, to make it sound much much worse, 825%.

Two things:

1. When I wrote the previous post, for some strange reason, I had it in my mind that I'd read that the DCX was around £400 ('twas around the time of our previous discussion that I last checked).
2. I am back-peddling because even if the DCX is eight times the cost of constructing an absorber, at £165, it is still a pittance compared to what I've spent on other equipment, offers greater flexibility than the absorber and can do much more besides.

Ok ok! Let me think on the matter. January is a bad month for me (lots of bills to pay). I do have one question though: Can the 'lugs' that allow the DCX to be mounted in a rack be removed?

 

[WmAx]

Can the 'lugs' that allow the DCX to be mounted in a rack be removed?

Yes. These are like L-brackets. 2 screws hold it on from the side. But if you remove the rack mount brackets, this will leave an opening on the sides of the DCX chassis. But it's easy to get rid of the tabs and cover the chassis entrance. Remove the L-brackets and cut off the part that mounts to rack and then re-install it on the DCX. It's a soft aluminum, so it's easy to cut/file.

-Chris

 

[Ethan Winer]

Chris,

> A properly set up parameteric filter will remove your resonance, at the listening position, completely. <

I'd say this has been disproven conclusively over at the AVS forum in the ongoing thread "¿Que es mas macho?" While it may be possible in theory for EQ to reduce ringing, in practice it doesn't appear to work that way. Here's the report on my company's web site, and there's a link at the top of the page to the AVS thread:

www.realtraps.com/eq-traps.htm

JohnPM chimed in on the thread over at AVS to point out that the fellow who tweaked the EQ did a less than stellar job. But if a guy who does this for a living was unable to reduce ringing, what chance do most people have? More to the point, what has never been established - even in theory - is over how wide a physical span EQ could reduce ringing.

Let's say for the sake of argument that someone adjusts their EQ correctly and manages to eliminate all ringing below, say, 200 Hz. What happens six inches away? (The distance from one ear to the other.) Is the ringing now even worse than without EQ? If so, how much worse? Or is it still present but better than before EQ? The burden of proof is now squarely on the shoulders of those who propose that EQ can reduce ringing. I will gladly accept that EQ really can reduce ringing as soon as someone shows a waterfall plot of one location where the EQ was optimized for, and also several other locations nearby. Until then I don't believe it, and others here shouldn't either.

--Ethan

 

[WmAx]

I'd say this has been disproven conclusively over at the AVS forum in the ongoing thread "¿Que es mas macho?" While it may be possible in theory for EQ to reduce ringing, in practice it doesn't appear to work that way. Here's the report on my company's web site, and there's a link at the top of the page to the AVS thread:

www.realtraps.com/eq-traps.htm[/url[/QUOTE] I have read the link on your website. I don't like to have to say this, but, your [I] *failure to analyze your own data properly[/i], combined with your strong apparent bias against electronic correction is alarming. *It is noted that you talk about change in position, but at frequencies <70Hz in your own graphs, located at your link, resonances are reduced at all of the measured positions with equalization, where as 17 mondo traps did very little in any position <70Hz, yet you still appear to categorically deny the usefulness of equalization in your forum posts. [quote]Let's say for the sake of argument that someone adjusts their EQ correctly and manages to eliminate all ringing below, say, 200 Hz. What happens six inches away?[/quote] I specifically and consistantly recommend equalization correction for resonances that are primarily minimum phase in nature. A 200Hz, or even 100Hz, a resonance would not behave sufficiently as minimum phase in any normal size room. At frequencies where the resonance is not directly related to parrallel dimension(s), the efficiency of equalization to tame resonances so far as listening position location is exponentially reduced as frequency rises. -Chris

 

[JohnPM]

Let's say for the sake of argument that someone adjusts their EQ correctly and manages to eliminate all ringing below, say, 200 Hz. What happens six inches away? (The distance from one ear to the other.) Is the ringing now even worse than without EQ? If so, how much worse? Or is it still present but better than before EQ? The burden of proof is now squarely on the shoulders of those who propose that EQ can reduce ringing. I will gladly accept that EQ really can reduce ringing as soon as someone shows a waterfall plot of one location where the EQ was optimized for, and also several other locations nearby. Until then I don't believe it, and others here shouldn't either.

I'll be happy to put that together Ethan, I've been meaning to produce a case study to walk people through using my measurement app anyway. The dimensions of the room I'll use are different to the one you used (longer but narrower), but I can make the reference point the same proportions as you used - I guess that would be best? I can measure over the same 2' grid you used, and I'll do some over smaller distances as well. Will be interesting to see how the effects vary with position, but give me a few weeks to get it all done.

Regards,

John

 

[sploo]

Gents,

Can I stop a potential fight here before one starts. Both of you are highly knowledgeable in your areas, and have given useful info to BM.

The way I see it, BM's done his room treatment, and now has a peak at a frequency that's probably going to require a fairly hefty resonator to fix.

The DCX 2496 should help this greatly, and for a reasonable price (and take up far less space too).

The last thing I want to see on this forum is a group of people, who are all opposed to audio snakeoil, starting to tear strips off one another. Leaving this forum open to the believers of 'magic beans' and cable cookers would be a great shame.

 

[Buckle-meister]

You got there just before me Sploo.

Sploo is correct. Remember; I am caught in the middle between two knowledgeable people and looking for guidance. For my sake, please keep it civil.

I understand that a PEQ can do much more than a Helmholtz absorber and that so long as no gain is applied, I cannot damage my equipment (though for all I know, not doing so maybe unrealistic). But Ethan has raised a point which for me is the crux of the matter and is exactly why I have always been a little wary of PEQ's; If someone who supposedly knew what they were doing didn't do that well, what chance have I?

Remember WmAx, the word 'properly' carries a lot of weight. I am not underestimating my intelligence, but as I have noted before, you need to bear in mind that I have nothing like the depth of audio knowledge that you possess.

 

[WmAx]

But Ethan has raised a point which for me is the crux of the matter and is exactly why I have always been a little wary of PEQ's; If someone who supposedly knew what they were doing didn't do that well, what chance have I?

Remember WmAx, the word 'properly' carries a lot of weight. I am not underestimating my intelligence, but as I have noted before, you need to bear in mind that I have nothing like the depth of audio knowledge that you possess.

It is true that it is critical to have it adusted 'properly'. But as I explained before, even if you don't know precisely what value you need to use, it is a simple matter to fine tune and re-measure until you centralize and correct specific resonances optimally. Good ol' trial and error. If you use the DCX GUI software with your computer, it's very easy to change the parameters on the fly with a mouse click or two. It might take you an hour for one resonant frequency, assuming the worst case scenario, but you'll get it right eventually.

-Chris