What makes a speaker driver ‘good’?

A driver is good(for you) if it fits your needs without making life difficult for you.

Frame your needs in the best detail, and you will narrow your choices.

Like a google search, generalise and get 44,566,908,423 answers, be specific, and get eight.

I'm sorry, but the hard part is figuring out how to find what works easily.
 
Thats an incredibly general question. Its like asking what is the best tasting food.

Some of my preferences are -

stay away from stamped steel baskets

look for a high Qms with a paper cone and cloth surround in bass/midrange drivers

Qts around .38 sounds best (to me) for ported applications - best transient response

Qts around .40 to.55 for larger sealed boxes and aperiodic dampened enclosures

straight cone profile for low crossover points (tends to break up worse in midrange but usually more accurate bass)

curve linear cone profile and lower Mms for higher crossover points ie. 2 way systems

smooth climbing response preferred in bass midrange drivers

under spider ventilation is a nice design feature
 
What makes a driver good can vary for a number of reasons.

A drivers main feature could simply be its price. It might not have the best performance but if it's usable, and it's extremely cheap, you'll often have a lot of fun experimenting with it.

One of the main downfalls of speaker design is that the drivers can get very expensive very fast and if you want to try out certain configurations, with multiple mid/bass units, price simply beats everything.

In a nutshell though you want a driver to have a flat frequency response over the range you intend it to be used, this includes the part where the driver is rolling off because of the crossover, in the transition-band of the driver+crossover. The same is true for it's harmonic distortion. You want this to be low over the frequency range you intend the driver to be used.

You want the speaker to posses T/S parameters that will give you the kind of bass response you are after, in the kind of box you desire, whilst reaching the SPL you require, without exceeding the drivers thermal or mechanical limits.

You may also wish the speaker to be of a particular sensitivity.
 
Last edited:
When looking at specs, what should I look for? Flat response? What can the t/s tell?

As per science and engineering principles, what you should be looking at, is your application. Now, since engineering is world of compromises, use the T/S parameters, curves, data (and wisdom) to arrive at a modest one.

"Non-professional" drivers usually have little to no data available for assessing their performance etc. and are therefore difficult to assess qualitatively. However, in the professional segment, there are two distinct categories namely studio (also known as near-field monitor, Hi-Fi, reference speaker etc.) and sound reinforcement (PA and high power). Cinema sound that often requires studio quality and PA power, is sometimes considered a third category and is a bleed-over between the two.

Using physical aspects:

To identify a driver's target application area, look at the materials of the cone, and surround. The shape (straight/curvilinear) and construction (ribs etc) of the cone would give you a fair idea about its breakup tendencies and possibl HF limit. In the magnet and voice-coil area, look for ventilation mechanisms, flux rings, dual voice-coil / differential drive, aluminium coil etc. (for high-power drivers).

Using T/S parameters:

PA drivers usually have lower Mms, higher fs and higher midrange sensitivity for similar BL factors as in studio drivers. For studio, Mms is modest, fs is lower and HF extension is lower due to the straight rigid cone. Higher mechanical Q also suggests lower suspension losses. A low Xmax often indicates a midrange rather than a woofer. Xmax (and Xdamage) are usually high in both studio drivers (for bass) and PA (for power).

Using curves:

Monitor drivers usually have lower sensitivity, flatter frequency responses, greater bass and lesser HF extension when compared to PA drivers. However, it may not be a very good idea to use only a few curves to select drivers.

A few opinions would be appreciated or a link to an existing resource.

An old technote from JBL explaining differences with examples.

Regarding the title of your thread, I would just say that "good" being a relative term makes "goodness" an abstract quantity. My answer was therefore based on the concept "good for certain application", with the good becoming bad (and sometimes ugly) as the user shifts between applications.

From the subjective point of view, it's said that a high BL with low Re and Mms usually gives some kind of "punch" to the driver. However, it's all debatable.
 
Last edited:
That's quite a general question.

Here are some of the specs I look at for a Hifi driver:
A smooth response, no huge breakups or at least 2 octaves away from the xo point.
Very good waterfall plots (CSD)
Good sensitivity >= 90db
Impedance (4-8ohm) depending of usage (single, parallel or series)
Impedance graph without irregularities
Distortion should be low.

Now for TS params,
often it's a combination of TS params that tells us more:
Low induction Le
Low moving mass Mms + high Bl translates to low ratio Mms/Bl
Low mechanical damping Fs/Qms
For woofers EBP (Fs/Qes) < 100

Now, good luck with finding a driver that ticks all the boxes,
or try to find the one with the least compromises 🙂
 
Last edited:
A driver is good(for you) if it fits your needs without making life difficult for you.

Frame your needs in the best detail, and you will narrow your choices.

Like a google search, generalise and get 44,566,908,423 answers, be specific, and get eight.

I'm sorry, but the hard part is figuring out how to find what works easily.
😀.... there is a good one although ! The price 😛


I have my own definition btw about it : a driver priced lower than the sum of passive parts it needs to playback good enough !
 
Last edited:
As a very simple generalisation what is the difference between a 96 DB (1W@1M) and a 88DB (1W@1M) driver?
Does higher sensitivity mean greater detail retrieval (from the input signal) and greater definition of output or simply greater volume?
Thanks.
 
Last edited:
No it doesn't, it's just greater lectrical sensivity that may allow you to use less powerfull amp that may (sometimes) have expensive low watts and some because of that may have greater details behavior you may be able to hear or not.


Most of the time you should to look for here a driver that has low THD, a good motor in distorsion behavior and electrical damping but also a good cabinet enough to have something like a flat power response in his pass band it is used.




Edit : Danny firered before me ! I'm not totally in agreement though about the greater the sensibibility the greater the subjective sounding dynamic ! A driver should be used with the good amp. The driver being able to reproduce the dynamic scaling of the reording without thermal compression (it's an easier task for a PA driver but that has also (but not always) less details reproduction behavior for many reasons)
 
Last edited:
I think we joked a little ! It's of course how you use a driver in a design that makes sing it or not. But sometimes, excelent drivers may cost expensive for some reason. There are cheap drivers that measure fine and well used by good designers makes awesome loudspeakers...
 
Very frequently, alas. I could point to any number of expensive units that simply do not justify their price tag. On the other hand, I could also point to a few supposed giant-killers that aren't quite what some tout them as, so in fairness it does work both ways. But assuming you get what you pay for in audio is a very dangerous thing these days. Once upon a time, when titans like Harry Olson stood over the field, it might have been true, to a broader level than it is today.