Driver Construction

[TheWarrior]

I keep reading about companies that employ their own drivers, that are made in house, and others that employ Seas or Scan-Speak etc. to customize their proprietary design for a specific project. How does that work? Do you just show up in Norway with a couple million and say, "make me 'this'?"

 

[markw]

Pretty much. There's a lot of engineering going on before the machinery starts to grind.

 

[Steve81]

How does that work? Do you just show up in Norway with a couple million and say, "make me 'this'?"

From what I gather, things generally kick off by getting samples from a variety of vendors. Obviously the engineers at speaker company X have an idea of what they'd like to get, but it's not as simple as switching a few dials on the master control panel of the driver assembly line to deliver exactly what they're looking for. Then there's the matter of evaluating driver to driver consistency, reliability of the supplier & their ability to meet demand, etc.

 

[TheWarrior]

And so this is where us DIY folk really have to shine? Mastering the math behind it so that off the shelf components can be crossed over to create the most authentic sound possible? I guess thats a good thing considering the overwhelming choice of components available....

 

[Swerd]

And so this is where us DIY folk really have to shine? Mastering the math behind it so that off the shelf components can be crossed over to create the most authentic sound possible? I guess thats a good thing considering the overwhelming choice of components available....

There are software programs (LspCAD, Sound Easy, Speaker Workshop) that actually do the measurements and math. There are even packages with full-duplex sound cards that allow recording and playback at the same time. This lets a designer emulate various crossovers on the computer, see the predicted frequency response curves on screen, and actually hear the results through a test cabinet with the drivers of choice.

But these tools can't substitute for skill, judgement, and experience.

It helps a great deal to have a well developed sense of how a speaker should sound and the experience to know how to avoid common pitfalls.

That's why I follow existing DIY recipes made by other designers .

 

[Steve81]

And so this is where us DIY folk really have to shine? Mastering the math behind it so that off the shelf components can be crossed over to create the most authentic sound possible?

As Swerd notes, custom drivers don't really have anything to do with the knowledge required to create a high fidelity loudspeaker from a pile of parts. What custom drivers do is allow an engineer to mold a driver to fit specific design requirements, vs simply designing around off the shelf drivers.

 

[TheWarrior]

So moral of the story is, after I build a kit design, I may run the risk of developing a strong obsession that leaves me forsaking family and friends for long hours (years) spent tinkering away in my workshop? Eh, guess it could be worse...

 

[cpp]

To me, the moral of the story regardless of the driver you use, it all comes down to the design and build-out of the cabinet and its' crossovers to support the cabinet and drivers specs.

 

[TheWarrior]

Got the cabinet discussion already started, but here's my question; Which came first, the crossover or the driver?

To me it appears my 'battle plan' should be something like:

Identify speaker purpose/desired end specification > driver selection > crossover design/ computer aided testing > cabinetry > hope it sounds good > send to Gene for review haha

 

[Swerd]

Got the cabinet discussion already started, but here's my question; Which came first, the crossover or the driver?

To me it appears my 'battle plan' should be something like:

Identify speaker purpose/desired end specification > driver selection > crossover design/ computer aided testing > cabinetry > hope it sounds good > send to Gene for review haha

Pretty close. I'd make the list in a slightly different order. Assume this is a 2-way speaker.

Identify speaker purpose/desired end specification
Select woofer(s)
Design cabinet dimensions for optimum woofer performance
Select tweeter based on woofer's highest useable frequency
Build unfinished version of cabinet for testing & crossover work
Mount drivers with wires accessible from outside of cabinet
Test drivers & design crossover
Build & test a prototype crossover with real parts (not computer simulated)
Tweak crossover to optimize sound
Build and finish final cabinets

 

[TheWarrior]

Wait so thats what you mean by, "There are even packages with full-duplex sound cards that allow recording and playback at the same time." ?

I need to look these up! What do you use?

 

[Ampdog]

Not into driver expertise that much, my intuition tends toward frowning on specialised driver design, as in 'yet another new design'. The immediate picture coming to mind is that with so many makes and types of (good) drivers on the market already, is there really such a void here that the extra complexity etc. of 'specially designed' drivers is going to make a telling (selling?!) difference?

I cannot get the persistent 'mutton-dressed-as-lamb' picture out of the mind - with apology to those manufacturers offering an honest alternative. With the sense of hearing with its many limitations as the final arbiter, what's for actual real improvement?