What's all this "Q" stuff??

[ThomasW]

People always seem to be confused regarding "Q", particularily when we're talking about IB subs. First, don't worry about trying to understand what all the various 'Q' numbers are or mean. For now consider them reference numbers we'll use for comparison.

Qts is the 'Q' of the raw loudspeaker. Qtc is the 'Q' of the 'system' (system meaning the operational subwoofer)

People design box subs for various Qtc. One of the more popular is 0.707. This is known as 'maximally flat' and is a popular 'Q' for 'HT' subs. A Qtc of 0.50 is called 'critically damped' and is frequently call a 'musical' sub.

What many 'box' sub designers fail to take into consideration, is that room gain will change the characteristic of these 'Q's. So starting off with a 0.707 sounds good in theory. But when placed in the room, the sub may tend to sound a bit bloated in the midbass given the boost from room gain.

So if we start of with a 'critically damped' Qtc 0.5 and add room gain we obtain a 'better' in-room response.

People get confused regarding the usual recommended "Q" for OB/IB sub drivers compared to what I recommend. So here's the beef....

The standard thinking is that open baffle/dipole/IB drivers should have a Qts around 0.69. And companies make drivers targeting that specific Qts. My thinking is that for IB subs and for box subs, this is too high a 'Q', unless one want's the added 'boom' when room gain is factored into the picture.

With a 10X Vas or larger IB sub, the drivers Qts = the Qtc, since the influences of the box are mimimal. So I recommend starting off with drivers having a Qts as close to 0.5 as possible.

One can change the Qtc. One method is RDO, where an L-pad is placed on one VC of a DVC driver. With that in place one can dial-in the 'Q' one desires, note it will always be higher than the Qts of the driver

What people tend to forget or don't know, is that using narrow band EQ filters raises the 'Q' of the sytem.

This is why I think it's better to start with a lower 'Q' driver. It's very to easy to raise the 'Q', but it's physically impossible to lower it.

Finally, people ask about 'music' only subs vs HT subs (more confusion generated mostly by marketing). I don't think this differentation should exist. A good sub is a good sub. It's no problem to take a 'music' sub (typically low Qtc), dial in a slightly different EQ curve and have a great sub for HT use....

This is of course my particular opinion, so the usual disclaimers apply YMMV....

Here's a nifty little online calculator for raising the 'Q' of a driver..
www.mhsoft.nl/spk_calc.asp#newqts

[chrisbee]

Thanks for the clear explanation.

Are there any other particular factors (or parameters) which one can use to pre-judge a driver's (likely) SQ in an IB?

[ThomasW]

Most of the T/S parameter are inter-related. Changing one changes others. As a result one can't arbitrarily pick and choose a bunch of random 'ideal' parameters to create a driver.

For an elementary real world example, one can use Unibox to see some of these interactions. If ones enters a known Fs, Re, Qms, Qes, Sd, Vas, Xmax, and Le, The program generates the Qts, Mms, Cms, Rms, BL. etc. Now change some of the manually entered parameters, then see what impact that has on the others..

Also there are many things we can't see using only T/S parameters, one is how the cone itself behaves. We want a stiff linear (pistonic) cone motion. Cms tells us about the suspension, but not the actual cone.

[jman]

Thanks Thomas, very informative.

I checked the TS specs for my Avalanche 18's and they claim a QTS of 0.400 so I guess they should have no problem sounding "musical" as well as "theatrical"

[chrisbee]

I've often wondered what would happen if you load the cone of a driver with something very light and stiff when Fs is higher than one might like but Xmax is acceptable.

I vaguely remember a DIY subwoofer design (a couple of decades ago) where the driver cone was filled with a disk of polystyrene foam glued in.

Would this be a cheap way of making a car or PA driver into a better IB candidate?

[Darrel]

Hey Guys,

"Most of the T/S parameter are inter-related. Changing one changes others. As a result one can't arbitrarily pick and choose a bunch of random 'ideal' parameters to create a driver. "

This is an important viewpoint and one that was hard for me to understand when first starting to work with the engineers while developing our drivers.
It was explained to me that each of these numbers are like ingredients in a recipe... nothing more.
The quality of the ingredients as well as the skill of the chef is likely to determine the final taste.

Take QT for example..... Not all 0.50 QT drivers will sound exactly the same .... It depends on how the QT was achieved as well as how this number interacts with several of the other numbers when viewed holistically ( As in the whole being greater than the sum of it's individual component parts).
Unfortunately we cannot design or purchase "by the numbers" . High performance is largely based on design experience and trials. Also remember that all drivers have compromises and we can only minimize the ones we feel are the most important for a particular task.

Using numbers is not too unlike using a dating service...... They will get you close to finding a perfect mate but are a poor substitute for actually meeting a person and letting your soul tell you what it feels......
Darrel

[ThomasW]

The typical thing is to mass load the cone with a compound to lower the Fs. Any benefits from the lower Fs are off-set from a resulting lowering of efficiency and an extra workload on the motor assembly and suspension from the added weight.

As we've seen it's not a big deal to have drivers play quite a bit below their rated Fs. And it's not a problem to use EQ to obtain the same output benefit from a lowered Fs.

So it ends up being a "why bother?" situation when it comes to lowering the Fs.

[chrisbee]

Thanks Thomas.

[aarond]

One can change the Qtc. One method is RDO, where an L-pad is placed on one VC of a DVC driver. With that in place one can dial-in the 'Q' one desires, note it will always be higher than the Qts of the driver

]

Where can I find some details of this practice?

Thanks,
Aaron

[cameron]

aarond said:

Where can I find some details of this practice?

This link is from #7 in the IB FAQ: www.adireaudio.com/Files/TechPapers/RDOOperation.pdf