do different drivers play louder at the same rate?

[falconsprint]

When desigining a mult-way speaker how do you know that the different drivers will play at the same relative SPL over for all input voltages. T/S sensitivity is measured at only one power level (1 watt) and often only at one frequency. How do you know that two drivers will play at the same relative SPL for 2,10,20,100 watts, etc...?

Of course I can measure the speaker at different levels once assembled. But what should I look at when evaluating drivers?

 

[fpitas]

Not sure I understand the question. If they're at all linear responding, the output will scale with the drive level.

 

[falconsprint]

That's basically my question. When you consider all of the different speaker designs (cone, dome, horn, ribbon, ...), materials used (paper, metal, fabric, sandwich,...) and physical size. Is it safe to assume that any driver will have SPL scale linear with input power?

 

[adason]

linear to certain level, yes...

 

[fpitas]

Exactly. If they're so non-linear the output is no longer more or less proportional to the input, they're probably unlistenable.

 

[CharlieLaub]

Good question!!!

To a first approximation, drivers are linear input-output devices. But every driver has its limits due to a finite voice coil length, so each will run out of excursion capability at some SPL, distortion will increase (a lot) and SPL will level off even if power input is increased. Then there is the issue of power input and heat generation. Remember that over 99% (typical for home audio drivers) is dissipated as heat. When a voice coil heats up, its resistance increases. This causes the sensitivity to a given voltage input to go down. This annoyance is very important for pro sound speakers! Thus, even if you are operating within the mechanical limits of the driver (e.g. within Xmax) if you deliver a lot of power over time to the driver the SPL will drop, and not necessarily in a fashion that is independent of frequency either. Each driver will do this differently, depending on its design.

So I guess the answer depends on the conditions you are operating under: short high power bursts or longer heat building continuous operation at moderate power levels.

-Charlie

 

[falconsprint]

Good question!!!

So I guess the answer depends on the conditions you are operating under: short high power bursts or longer heat building continuous operation at moderate power levels.

I understand that all drivers will behave differently at their limits.

I should have thought more before asking the question. I was thinking of the problem more mechanically: how can drivers of such different mechanical designs all have the same SPL/power response? But now that I think back to my physics education, I know that energy has to be conserved. So as the power increases, the SPL needs to increase proportionately until the point that the physical characteristics of the driver change (xmax, or coil overheating). At which point a larger percentage of the power is dissipated as heat.

 

[turbodawg]

Is it safe to assume that any driver will have SPL scale linear with input power?

Linear with respect to voltage........supposedly.

 

[chris661]

Ah, but..

If the THD of one driver rises faster than the other, would that driver sound louder than others given the same drive voltage?

 

[tinitus]

.. would that driver sound louder than others given the same drive voltage?

yes, but hope you wont be listening to it

very often its the crossover is causing a peak....been there a few times too
its tricky
it will sound like its the driver, but it really isnt

 

[Richard Ellis]

I think it does sound Louder....my book reads "averages slightly above 90 Db...." "Fast Fourier Transform system known as MLSSA."
My take on this brief notation is that as fundamentals and multiples start appearing the outright efficiency goes up.

______________________________________________________Rick.........

 

[Pano]

Quick answer: As you have seen, is "YES, they will play at the levels they are rated at and will change the same amount with the voltage change."

Not so Quick Answer: They will vary some, as discussed here. Most drivers are rated for 1 Watt, but is it really 1 watt? Or is it 2.83 volts? That would be 2 watts on a 4 ohm speaker. Some drivers are not linear, but most are within their limits.

The specs are cheated in other ways, too. But the ratings are a good starting point, you have to fine tune from there.

 

[weltersys]

So as the power increases, the SPL needs to increase proportionately until the point that the physical characteristics of the driver change (xmax, or coil overheating). At which point a larger percentage of the power is dissipated as heat.

Correct.
However, a relatively large, heavy woofer voice coil has a fair amount of thermal mass and takes a lot of average power over a relatively long time (minutes) to heat enough to cause thermal compression due to rising impedance, while a very lightweight tweeter voice coil has very little thermal mass and may change in impedance very rapidly (as little as fractions of a second), so upper frequencies may be reduced more than lower frequencies when subjected to peaks.

 

[CopperTop]

Then there is the issue of power input and heat generation. Remember that over 99% (typical for home audio drivers) is dissipated as heat. When a voice coil heats up, its resistance increases. This causes the sensitivity to a given voltage input to go down. This annoyance is very important for pro sound speakers! Thus, even if you are operating within the mechanical limits of the driver (e.g. within Xmax) if you deliver a lot of power over time to the driver the SPL will drop, and not necessarily in a fashion that is independent of frequency either. Each driver will do this differently, depending on its design.

So I guess the answer depends on the conditions you are operating under: short high power bursts or longer heat building continuous operation at moderate power levels.

There's an article here that says that this effect is virtually non-existent for home hi fi listening, which is good to know:

Hot Stuff: Loudspeaker Voice-Coil Temperatures | Stereophile.com

The results, shown in figs. 3 and 4, came as something of a surprise. Despite what rates as a high playback level for me and, I imagine, most Stereophile readers, I had anticipated there being only modest increases in voice-coil resistance. But the increases were even less than I'd expected...

...I strongly suspect that, for most hi-fi users—those who don't habitually wind the volume control to its highest position and indulge in PA listening levels—thermal compression is a paper tiger.

 

[POOH]

Many drivers compress sooner then others , the better drivers ahve very little compression and normally "hi fi" low sensitive drivers or long stroke subwoofers drivers tend to compress the worst. I have noticed ribbons seem to compress much sooner then a good cone in use but never measured it

 

[nigelwright7557]

Different driver have different efficiencies.
They also have varying output levels at different frequencies.

 

[CharlieLaub]

There's an article here that says that this effect is virtually non-existent for home hi fi listening, which is good to know:

Hot Stuff: Loudspeaker Voice-Coil Temperatures | Stereophile.com

I'm familiar with that article, in fact I want to repeat it someday to monitor the driver's Re of a feedback subwoofer system I want to build...

...but I wouldn't take that as the last word. It depends on the driver, program material, etc. In this case, yes, there was not much compression. But I suspect this is a robustly designed driver. What about a small full range cone? That might be completely different. The OP wasn't asking about "typical use" per se, so I answered more in terms of the limiting behavior.

Anyway, your link is a good piece of work and very relevant to this thread.

-Charlie

 

[chris661]

yes, but hope you wont be listening to it

very often its the crossover is causing a peak....been there a few times too
its tricky
it will sound like its the driver, but it really isnt

I've using active crossovers, so that's probably not the problem.

My system is a Fostex FE126 per side, crossed to a Tang Band W6-1139 per side.

When I turn it up so ~80dB or so (the "Signal" light comes on, indicating a couple of volts on the output of the amp driving the TBs), the Fostex has to be dialed down a couple of decibels to get a balanced sound. Turn the systems up further, and the Fostex has to be turned down further, relative to the woofer.

At a watt input, I really don't think there's going to be any thermal issues at work with the woofers, so all I can think of is increasing distortion levels on the Fostex. This led me to post my question.

Chris