• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

How to identify a good/great speaker driver?

Joined
Oct 4, 2018
Messages
49
Likes
18
#1
Leaving aside the quality of the enclosure (for which there are some known parameters to look for, i.e. sturdiness, correct volume, etc...), how can you separate a great speaker driver from a mediocre one?

For electronic devices we have some well known measurements (SINAD, IMD, THD, noise, etc...): are some of these measurements applicable also to speaker drivers?

For example I guess that measures like THD and linearity would be very applicable, but usually they are not published by manufacturers, and so we are left with frequency response graphs as the one and only evaluation parameter.

Is there any way to correlate some other quality related measures to Thiele/Small parameters maybe? Some rules of thumb like "all other things being equal, drivers having low/high Qms/Qes/Qts/Vas/etc... should have lower/higher THD"?
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,635
Likes
1,722
#2
It’s hard to generalise because different woofer designs will result in very different T-S parameters (and yet not necessarily great differences in quality of performance). Another problem is that all else is almost never equal, as changes to a woofer's design that affect one T-S parameter will tend to affect other T-S parameters too.

Having said that, here are a few rules of thumb:
  • Woofers with lower EBP (= Fs/Qes) tend to be better suited to sealed enclosures than to vented enclosures (and vice versa).
  • Woofers with a higher Vas will be better suited to larger enclosures (and vice versa).
  • Woofers with a lower Le tend to have lower nonlinear distortion and a more extended high frequency response.
  • Woofers with an underhung voice coil should behave more linearly within Xmax and less linearly beyond Xmax than woofers with an overhung voice coil.
  • Woofers with a high gap depth/Xmax ratio should behave more linearly beyond Xmax.
I know the last two points don't strictly relate to T-S Parameters but many manufacturers provide data nonetheless. There are probably a few other rules of thumb that haven't come to mind...

It's often more fruitful look at a woofer's design and its T-S parameters as a whole when trying to make an educated guess about its performance. For example:
  • What's the diaphragm material?
  • Is there a shorting cap?
  • Is the voice coil former conductive?
  • Is the voice coil overhung or underhung?
  • Is the pole piece vented?
  • Is the surround lossy?
  • Is the dust cap porous?
  • Is the suspension system linear or progressive?
  • etc...
Some manufacturers publish a wealth of information about the design of their woofers (e.g. B&C). Others publish very little. In the end, there's no substitute for actually measuring the woofer yourself...
 

Shadrach

Active Member
Joined
Feb 24, 2019
Messages
100
Likes
127
#4
Different drive units work better in one construction than another.
A decent drive unit producer should be able to supply all the measurements you need to design a loudspeaker.
There are better constructions than others and some materials work better in some applications than others.
My loudspeakers use ScanSpeak units and units from the people below.
The problem imo with many loudspeakers is they use cheap drive units and then try to design an enclosure to minimize the drive units shortcomings; porting to achieve the effect of having greater bass for example. Truth is a speaker with a port is just pressurizing air through a hole.
When I do that it's called a fart!
Same often applies to the crossovers, properly laid out, preferable not on a circuit board but hard wired with decent components.
There is no need to go the boutique route, just decent quality and the right type for the job.

https://voltloudspeakers.co.uk/
http://www.audio-constructor.com/index.php?main_page=index&cPath=1_345
 

March Audio

Major Contributor
Manufacturer
Patreon Donor
Joined
Mar 1, 2016
Messages
3,202
Likes
2,459
Location
Perth Western Australia
#5
It's a difficult question as the enclosure forms an intrinsic part and effect on the driver performance. You have the fundamental choice between sealed and ported and require a driver with appropriate TS parameters for that choice.

Just one observation, when I was researching drivers for my speaker design I saw many which had extremely variable frequency responses. I saw designs that needed the crossovers to counteract these deficiencies. I made two specific design choices. Drivers that were as inherently flat response as possible and wide even dispersion.

I found the SB Acoustics woofers seem to perform better than many of the usual Scandinavian suspects in this respect.
 

Another Bob

Member
Patreon Donor
Joined
Nov 28, 2018
Messages
5
Likes
2
Location
Madison, WI
#6
In addition to being "inherently flat", how smoothly the driver rolls off on the upper end will have a big impact on how easy or difficult the crossover design will be - unless, of course, you have the luxury of being able to place the crossover point well below the roll-off. Also look at the off-axis response. If that differs significantly from the on-axis response (except for the expected HF roll-off as the wavelength approaches the driver diameter), it could be an indication of diaphragm break-up or other undesirable behaviors.
 
Joined
Oct 4, 2018
Messages
49
Likes
18
#8
It's a difficult question as the enclosure forms an intrinsic part and effect on the driver performance.
Yes, absolutely right. My assumption here is that the enclusure needs to be the correct one for the selected woofer. But we don't want to build a perfect enclosure for a bad driver :)
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,635
Likes
1,722
#9
Truth is a speaker with a port is just pressurizing air through a hole.
When I do that it's called a fart!
I have the same feeling about ported speakers. I only have sealed ones, including the subwoofer.
For various reasons I've explained elsewhere, my view is that ported speakers are nowhere near as problematic as the conventional audiophile wisdom would have you believe. I won't bother with the details here, but suffice it to say that a port allows the same SPL output with far lower voice coil excursion and therefore lower distortion and higher maximum SPL. Moreover, like sealed boxes, ported boxes are essentially minimum-phase* systems (there is a widespread mistaken belief that they are not).

*Of course, in reality no driver in a box is actually a minimum-phase system.
 
Joined
Jun 20, 2018
Messages
330
Likes
120
#12
Leaving aside the quality of the enclosure (for which there are some known parameters to look for, i.e. sturdiness, correct volume, etc...), how can you separate a great speaker driver from a mediocre one?

For electronic devices we have some well known measurements (SINAD, IMD, THD, noise, etc...): are some of these measurements applicable also to speaker drivers?

For example I guess that measures like THD and linearity would be very applicable, but usually they are not published by manufacturers, and so we are left with frequency response graphs as the one and only evaluation parameter.
Measurements do exist for many hi-fi drivers.
https://hificompass.com/en/speakers/measurements
http://www.audioexcite.com/?page_id=178
bit older: http://www.zaphaudio.com/
More measurements by users often in DIY forums.
And there are magazines.
https://www.audioxpress.com/tags/Test-Bench
And many drivers tested by http://www.hobby-hifi.de/ (in German and have to look up catalog for which driver you want to see measurements and then backorder the actual magazine)
 

watchnerd

Major Contributor
Beer Hero
Joined
Dec 8, 2016
Messages
4,280
Likes
1,362
Location
Seattle Area
#16
Identifying great drivers is like identifying great tires. Racing slicks are horrible in the snow, and vice versa.

Without information on the use case, it's a fairly nonsensical discussion.
 
Joined
May 30, 2018
Messages
76
Likes
56
#19
I've got a interesting set up where I have some 6.5 cubic foot enclosures (originally built for stage use) but early on I modified the interior so I could adjust volume with a series of cleats where I can slide a partition, caulk it, and be able to variably adjust from 6.5 to 3 cubic feet with another interior enclosure for a tweeter. Midrange boxes I just build to suit. This has allowed me to change out drivers over the last 26 years with relative ease by just remaking front baffles.

My experience is that factory specs are very useful and a necessary part of the selection process. But, and this is true of showroom speakers too, the measurements you get from your build and ultimately your listening environment will be different by a fair margin - mostly because of your room. I have an active XO so it's fun to try to coerce even a mediocre driver into behaving when I find something interesting and cheap. With an active 3.2 way XO I also have the luxury to easily in real time limit frequency ranges applied to each driver and the type and strength of their slopes.

If the OP is inquiring as they are on a quest to build something for their listening pleasure in a dedicated space - my advice would be to look at the space first and find a speaker design that compliments your intended space and SPL needs. Then start researching for the specific types of drivers your going to need for that build. If you decide to go with active XO you would also have the ability to use a selection with a greater difference in sensitivity and many other benefits as well. Perhaps you are already at this stage! :)

I have a less than ideal space which I try to design my speakers around and aim for wide dispersion more than for a sweet spot and these things are hung from an 11' ceiling. Here are some of the incarnations over the years. Obviously, my in room measurements of individual drivers would be pretty useless to most people.

speaker.jpg
after.jpg
current.jpg
 
Top Bottom