QTS = closed or vented?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2002
Paid Member
The following excerpt from a previous post makes sense to me:

Closed box : use speaker QTS in the 0.35-0.55 range, for QTC between 0.65 and 0.9

Vented Box : best QTS are around 0.27 - 0.35 for decently damped QB3 alignements


But, Dickason's LDC clouds this issue for me with it's quote of Small's EBP formulae. It suggests that you divide Fs by Qes and any values of 50 and under suggest closed while values of about 100 suggest a vented one.

The confusion lies in the fact that these two methods don't always correlate.

Dan
 
Disabled Account
Joined 2001
That was a generalization, and is typically accurate. As you will notice, the eficiency bandwidth product does not have iron clad rules either, but in either case.....the higher resonance magnification(overall Q) the more suited it is for a sealed, and then infinite baffle use, likewise, a better self-damped speaker(low Q) is more suited to ported enclosure. This is because a high Q speaker has to be used in an application that allows for the high levels of energy stored in the speaker at resonance to be dissipated over a broad and relatively linear frequency range(centered at fb). Using a high Q speaker in vented box is not satisfactory because it does not offer broad and linear damping curve to counteract the effect of high Q, resulting in a non linear, and peaky amplitude response and very poor transient decay. Likewise, using a speaker with a low Q in a sealed box is not optimal becuase it is, in actuallity overdamped in these alignments. And while not hurting transient decay as when using a high Q driver in a vented, this will lead to a relatively high F3. However, if the F3 is suitable for your purposes their is absolutey no reason not to use it in a sealed alignment because the amplitude response does remain linear..

The LDC, while accurate and an excellent source of information...simply does not have room to expand on every single issue in detail. If it did, it would be a mult-volume reference series, not a single book.

-Chris
 
diyAudio Moderator Emeritus
Joined 2001
I have never read Vance Dickason's book. However, both of David Weems' books give you a chart, made up by AN Thiele himself and modified by Small, of useful "Qts/box size combos" for each Qts in vented systems. The original chart is also available from Thiele's article on vented loudspeakers in the Journal of the Audio Engineering Society, Loudspeakers in Vented Boxes, Part I. This is available online from the Journal of the Audio Engineering Society for something like 5 bucks.

It should be noted that Small found that Thiele's alignments work well in the real world if you simply increase Vb 35% over that which Thiele's chart states. Don't modify anything else, just increase the real-world volume 35% over the stated formula. Small and Weems modified their charts accordingly, so that the Vb they state is the real-world volume you need.

Weems also tells you how to modify tuning, etc. to get good results when you are using a driver/box combination that is NOT ideal according to the chart. In point of fact, you have wide variation that is possible.

To briefly summarize, the "classic" bass reflex consists of a driver with a Qts of .38, (or .4), the Vb, (box volume) is equal to Vas, and the Fb, (frequency of the box tuning) is equal to Fs. This will yield a box where the F3 is equal to Fs as well.

If your Qts is BELOW .38, then your ideal box volume will be smaller than Vas, and your Fb and F3 will be ABOVE Fs.

If your driver's Qts is .3, for example, then your ideal box volume is .5 Vas, your Fb will be 1.2 Fs, and your F3 will be 1.45 Fs.

If your driver's Qts is ABOVE .4, (up to .56), then your ideal vented box size will be larger than Vas, your Fb and F3 will be BELOW Fs.

A driver with a Qts of .52 will have an ideal box volume of 2.4 Vas, and have an Fb of .75 Fs, and an F3 cutoff of .64 Fs.

The most desirable alignments are the "classic" .4 Qts on down to those requiring smaller Q's. The alignments requiring larger Qts will work, but will not have the same "snap" as the smaller Qts alignments.

As for closed box systems, you want a Qtc, (final Q of the closed box/driver combination), of between .7 and 1.0. Purists like a final Qtc of .5, but your output is 6 dB down at resonance in that alignment.

Fc is the resonance frequency of the closed box/driver combination.

When a driver is placed into a closed box, BOTH it's Qts and Fs are raised according to the same formula. For the raise in Q, the following formula is used:

Qtc = the square root of [(Vas/Vb) + 1] times Qts.

And applying the same formula to the resonance frequency:

Fc = the square root of [(Vas/Vb) + 1] times Fs.

So, a driver with a Qts of .5, if placed into a box that is one third it's Vas, will yield a driver/box combination that has a Qtc of 1.0 and an Fc that is twice the drivers Fs.

Any number of programs will give you these answers, but I thought I would give you some reasons as to why they arrive at these numbers. You can make your design decisions based on these principles.
 
Member
Joined 2002
Paid Member
To briefly summarize, the "classic" bass reflex consists of a driver with a Qts of .38, (or .4), the Vb, (box volume) is equal to Vas, and the Fb, (frequency of the box tuning) is equal to Fs. This will yield a box where the F3 is equal to Fs as well.

If your Qts is BELOW .38, then your ideal box volume will be smaller than Vas, and your Fb and F3 will be ABOVE Fs.

If your driver's Qts is .3, for example, then your ideal box volume is .5 Vas, your Fb will be 1.2 Fs, and your F3 will be 1.45 Fs.

---------------

The driver I'm using is an 8" woofer with a QTS=.272. I'm using LspCAD to design a ported enclosure and when you use it's 'optimize' function it calculates a volume that alway produces Fb being equal to Fs. The only downside to this is that this also produces an impractically long port length. As an example my info is :

Qts=.272
Vas=144L/5.09 cu.ft.
FS=20.5=FB
SBB4 volume is approx. 47L/1.66 cu. ft.
Port dia. is 3" and length is 22.7"

Something that puzzles me is that I quite often see people shortening their port length to 'fit the box' It appears to only raise Fb. I'm not experienced enough yet to calculate the true impact of this rise in Fb over Fs. Hmmmmmm...

Dan
 
diyAudio Moderator Emeritus
Joined 2001
I do not have LSPCAD, though I have heard it praised in this forum. However, if it is telling you to tune a box that is one third of Vas to Fs, then something is wrong somewhere. Download and use another program, such as BoxPlot, QUICK until you get that situation resolved.

If you do download BoxPlot, I find it a good idea to deduct 10% from the box volume when inserting your numbers. Simply multiply the box volume by .9. Otherwise, the curves tend to come out a teensy bit optimistic.

According to Boxplot, you should tune the enclosure to 29 Hz, and expect an F3 of around 32 Hz or so. Using the port designer on Steve Ekblad's site, your port should be a 3 inch pipe that is 11 and half inches long. 8 inch speakers can frequently get away with a 2 inch port, which would be 4 and a half inches long in this case, but I am guessing that any 8 inch speaker with a 20 Hz Fs is probably a "long-throw" model and moves more than the average amount of air.

If you have difficulty fitting the pipe into your box, PVC elbows are a available at most building stores.

In addition to BoxPlot, there is UniBox, which requires Excel on your computer, and WinISD, which I haven't used but is supposed to be good, and BoxModel, a simple but effective DOS program.

PS: If you do download BoxPlot, be aware that "Alpha" means (Vas/Vb). In your case, "Alpha" is about 3 (5.1/1.66). Also, "h" is (Fb/Fs). In your case, "h" is about 1.5-(29/20). "h" is often referred to in technical articles as "the tuning ratio".
 
diyAudio Moderator Emeritus
Joined 2001
dantwomey said:


Something that puzzles me is that I quite often see people shortening their port length to 'fit the box' It appears to only raise Fb. I'm not experienced enough yet to calculate the true impact of this rise in Fb over Fs. Hmmmmmm...


Steve Ekblad's site has a nifty "port designer" that will quickly tell you how much shorteneing the port will raise Fb. Also lots and lots of other important stuff.
http://www.wssh.net/~wattsup/audio/

As to how this impacts the curve, it is as follows. If the tuning is raised to 1.4 it's original value, (half an octave), then the new tuning frequency will be 6 dB above what the old tuning frequency was. If the new frequency is 2 times the old frequency, (a full octave), then the new frequency will be 12 dB above what the original frequency was.

A speaker that normally plays at 90 dB @ 1M/1W is 3 dB down at 20 Hz. That means the speaker is 87 dB at 20 Hz. The vent is shortened so the box is now tuned to 28 Hz. The speaker is now playing at 93 dB at 28 Hz.

The vent is shortened again so the box is tuned to 40 Hz. That is a full octave above 20 Hz. The speaker will be playing at 99 dB at 40 Hz.

All this works the same way in reverse when you LOWER the tuning frequency. :cool:

Try it on one of the new programs you are about to download and see for yourself.
 
dantwomey said:

---------------

The driver I'm using is an 8" woofer with a QTS=.272. I'm using LspCAD to design a ported enclosure and when you use it's 'optimize' function it calculates a volume that alway produces Fb being equal to Fs. The only downside to this is that this also produces an impractically long port length. As an example my info is :

Qts=.272
Vas=144L/5.09 cu.ft.
FS=20.5=FB
SBB4 volume is approx. 47L/1.66 cu. ft.
Port dia. is 3" and length is 22.7"

Something that puzzles me is that I quite often see people shortening their port length to 'fit the box' It appears to only raise Fb. I'm not experienced enough yet to calculate the true impact of this rise in Fb over Fs. Hmmmmmm...

Dan

You might want to pose these questions directly to the author of LSPCAD, either at his site or on the Madisound forum (he often replies there). I think there is also an LSPCAD user's group somewhere at Yahoo!

By the way, are you trying to design this box based on published or measured specs. I wasted much time trying to optimize a box for a rather expensive woofer, only to find when I measured it's T/S parameters they were way different than advertised...and in fact my box would need to be nearly 50% larger than planned. What a disappointment!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.