• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Measuring speaker efficiency...roughly

dewdude

dewdude

High on resolution.
I'm sure some of you remember the DIY speakers I picked up sometime last year....slightly larger than a minifridge and about twice the weight. They're a basic cabinet design with the vintage JBL 030 system in it (d130 driver, bullet tweeter and 2400 crossover). They're extremely well built and aside from some chipped verneer they've stunningly gorgeous, double as pieces of furniture and sound just as amazing. I was never really that big on speakers from that era as the ones id heard had been so-so, but these totally changed my mind.

Anyway, my grandfather, who remembers the jbl system and how some guys he knew thought they were the best of the best then was admiting them and got to asking me if I knew what their efficiency rating was. Being built in an era where 20 to 30 watts was regarded as high power, I assume the rating is pretty high for a speaker that large (mostly since my 100wpc amp doesn't even break a sweat while being too loud in a room).

What id like to do is get a rough estimate of the efficiency rating by feeding a 1khz sine in to them at 1 watt of power, standing about 3ft away and seeing what my spl meter reads. The thing is, not sure how to narrow the amp output down to 1watt. So what id like to do is measure the voltage of the speaker outputs with my multimeter so I can get to that 1 watt mark.

My main question, if anyone knows, what's the voltage of a 1khz sine wave at 1 watt going in to 8 ohms? I'm not even sure how i'd calculate that.

Thanks.

Sent from my Virgincept using Tapatalk.
 
Register to hide this ad
1 watt at 8 ohms is 2.83V regardless of frequency, afaik.

Interesting question.

Considering that the speaker's impedance may or may not be 8 ohms at 1kHz, I wonder if the right approach would be to do what you propose using a dummy load for calibration, then connect to the speaker and measure?

Also, I'm wondering if the test should be done outside to eliminate any type of room reflections that may interfere with a true measurement?

Lastly, the SPL meter may or may not be accurate at that frequency so it too is a consideration in the net result.
 
Last edited:
White noise would be perfect. Get yourself a Radio Shack (or other) SPL meter, some FM hiss :-) and a meter stick (that's like a yard stick, but calibrated in them SI unit thingies). Put the R/S SPL meter on a tripod (or ask someone to hold it for you) on axis a meter away from the speaker.

A "nominal" 8 ohm speaker ought to be 8 ohms to a very good approximation in the midrange (from an octave or so above resonance on up into the low treble, for most 'normal' speakers)... I'd say set your white noise output to 2.83 VAC and measure.

Close enough for Government work :-)
 
dewdude said:
I'm sure some of you remember the DIY speakers I picked up sometime last year....slightly larger than a minifridge and about twice the weight. They're a basic cabinet design with the vintage JBL 030 system in it (d130 driver, bullet tweeter and 2400 crossover). They're extremely well built and aside from some chipped verneer they've stunningly gorgeous, double as pieces of furniture and sound just as amazing. I was never really that big on speakers from that era as the ones id heard had been so-so, but these totally changed my mind.

Anyway, my grandfather, who remembers the jbl system and how some guys he knew thought they were the best of the best then was admiting them and got to asking me if I knew what their efficiency rating was. Being built in an era where 20 to 30 watts was regarded as high power, I assume the rating is pretty high for a speaker that large (mostly since my 100wpc amp doesn't even break a sweat while being too loud in a room).

What id like to do is get a rough estimate of the efficiency rating by feeding a 1khz sine in to them at 1 watt of power, standing about 3ft away and seeing what my spl meter reads. The thing is, not sure how to narrow the amp output down to 1watt. So what id like to do is measure the voltage of the speaker outputs with my multimeter so I can get to that 1 watt mark.

My main question, if anyone knows, what's the voltage of a 1khz sine wave at 1 watt going in to 8 ohms? I'm not even sure how i'd calculate that.

Thanks.

Sent from my Virgincept using Tapatalk.
Click to expand...
If this were a contest you'd want ideal conditions. Since this is for curiosity, more than anything else, I'd simply do what Mhardy suggested. The number of db's you'll be off will be so inconsequential it won't matter much.

Have fun doing it!
 
I've heard the best way to avoid room reflections adding to the sound level is to put it on its back outdoors and suspend the mike above it.

If it's a JBL system, can't you assume it's about the same efficiency as the stock JBL speaker?
 
I don't know of this specific system, but AFAIK, the woofer is often the limiting factor, so, used as a direct radiator, sensitivity might be quite close to the woofer's spec.

Most amps are voltage sources, so the impedance at any frequency "should not matter", it will put out 2.83V accross the load. Current will vary and so will wattage though.

az
 
audiozaz said:
I don't know of this specific system, but AFAIK, the woofer is often the limiting factor, so, used as a direct radiator, sensitivity might be quite close to the woofer's spec.
Click to expand...

Well...since these things consist of just a 15" driver and tweeter, I'd say the woofer is the limiting factor as essentially, it's all it's got. It's not the first time I've seen a 15" used as a wideband driver...Seeburg jukeboxes from the 50's often only had a 15" driver in them....of course they weren't permanent magnet speakers and somehow got the job done. (Yes, room shaking bass and clear HF from a single speaker)

But, this is mostly a curiosity thing. It's like..I know my DCM's are 92db efficient, my Polks are 90db efficient (or maybe it's the other way around)...but I don't have a number to give people when it comes to these...so I've just want to get a rough idea other than telling people they're insanely loud. I've gone on that assumption thus far as my Denon is 100wpc...and they're blasting people out of the house before I even get the dial to the -50db's range. The thing has that problem I've encountered with age...and sealed pots...is that when you get the volume control almost all the way to the absolute min position...one channel will fade out completely before the other...and i've only ever really noticed it on headphones. But with these speakers it's almost too freaking loud by the time you get to the point it's even on both channels. It could just be that the room isn't really all that large either.

But thanks for the input. When I get the chance I'll get the SPL meter set up and get some rough numbers.
 
In-room measurements are a mess in the lower frequencies. If the room is not insanely tiny and you have a good combination of absortion and diffraction, carpet, bookshelf, plants, couch, a reading of 2.83V at 1kHz might give you a good idea IMHO.

I get this extremely-low volume channel disbalance thing too with my rather sensitive AD9710 in Karlsons. Good thing the power amp has input attenuation control.

az
 
The room, while small and having hardwood floor, has a rug covering 90% of the room....and stuff that asorbs the sound...plus its got a hallway on on the back corner and an entry way to the kitchen perpendicar on the same wall that comprises the hallway.

Usually, when the thing is in my rig, the only audio source I use is a dac, which has an output control, so I keep it down and main amp up. I don't actually get any major hiss from cranking the volume till I get to max, for a 30 year old amp it still sounds great....although its got some strange "servo-controlled" dc circuit....no real caps in the thing to speak of and I've done the calibration according to the service manual. It still gives me shivers up the spine when I play it because it's just so damn nice to listen to. If was well worth the back pain the first time I picked the thing up not realizing how heavy it was. 90% of the weight is in the toroid power transformer.

I'm looking forward to getting a rough value on efficiency...and my grandfather told me don't be surprised if its some isanely high number...."those things were built to be loud with 35 watt tube amps....you're basically strapping a rocket to a pinto with those"

Sent from my Virgincept using Tapatalk.
 
In the small points nit-picking department, what y'all are talking about is "sensitivity", not "efficiency". Different things entirely.

Sensitivity is measured as sound pressure level in dB at 1 watt output at 1 meter. Efficiency is how efficient a speaker is at turning amplifier output power into sound. Most speaker efficiency runs around 1%-2%.

Carry on.
 
Fisherdude said:
In the small points nit-picking department, what y'all are talking about is "sensitivity", not "efficiency". Different things entirely.

Sensitivity is measured as sound pressure level in dB at 1 watt output at 1 meter. Efficiency is how efficient a speaker is at turning amplifier output power into sound. Most speaker efficiency runs around 1%-2%.

Carry on.
Click to expand...
OK, this is interesting since many people (even I) have used both "words" interchangeably, with the idea that it was essentially understood "sensitivity" was implied.

I don't recall "efficiency" being used, but I haven't read any literature for speakers in a long time. Do you have any examples, in brochures or literature of some sort, where efficiency is referenced and referenced correctly?

Thanks!
 
There's lots of reference on efficiency in textbooks on loudspeaker design, typically where the low frequency driver anatomy is discussed.

0.5% efficiency is typical and 2%~5% is exceptional. IIRC, ~5% is what a horn driver delivers.

Now, as for the loudspeaker system in question, I'm guessing 93~94dBm/w based on loud volume with a 100W amp not breaking a sweat.
 
Fostex FE206E

Efficiency spec is given as Eff/n0, 1.73% in this case. It's a pretty sensitive driver at 96dB/W/m, but it's efficiency is not that high, though lots of driver are lower.

az
 
Sure is a lot of work to find out something that's long been documented..

I had the same speakers... I had looked up the info at one time.. and if I remember right, it was 96db/1w/1m

I might be off a db or so.. .it was a few years I looked it up.. but google is your firend.. the info was on the heritage site in one of the pamphlets on that JBL system if I remember right...
 
It would make things so much easier to standardize the test as throwing a standard voltage (such as 2.83Vrms) at the speaker, regardless of impedance. What matters to consumers, relative to a sensitivity spec, is how loud does it play when my volume control is at x. Trying to establish what impedance the speaker is with a band of frequencies is problematic. Sure, knowing the load impedance is important, but only as far as the amp's drive capability is. If an amp manufacturer decides they want a 96dB @ 1 meter speaker, but it requires them to make the speaker 1 ohm, so be it. I see sensitivity and load impedance as two parameters. Give them high marks on the high sensitivity, but ding'm on marketing a 1 ohm speaker.

Having some band limited noise, perhaps following a CCIR-2K or A-weighting curve would be nice too. These bandwidth reflect the ear's most sensitive ranges. Having a single frequency, such as 1kHz, let's the charletans get great specs with peaky speakers.
 
Copa1934 said:
OK, this is interesting since many people (even I) have used both "words" interchangeably, with the idea that it was essentially understood "sensitivity" was implied.

I don't recall "efficiency" being used, but I haven't read any literature for speakers in a long time. Do you have any examples, in brochures or literature of some sort, where efficiency is referenced and referenced correctly?

Thanks!
Click to expand...

When the words are used interchangeably, the word efficiency is being used incorrectly. There is a good, although necessarily technical, explanation on pages 36-37 of the Loudspeaker Design Cookbook. What's interesting and valuable about this particular discussion is that it mathematically explains the relationships between efficiency and sensitivity.
 
True - the concepts are pretty much interchangeable for us laypeople, but there are big (and not readily interconvertible) differences in the sound power output, SPL, and perceived loudness. It actually gets unbelievably tricky - as I (re)learned a couple of nights ago looking at specs for kitchen range vents. The noise level is rated in "sones". Now, you can convert "sones" to "phons", but you cannot directly convert either to decibels, not to mention, say, dBA (A-weighting).

As to efficiency - here's an example (FWIW), a symphony orchestra at full tilt has an acoustic power output of about 1 watt. If it takes 100 W input for ones loudspeaker's to reach "full concert volume", the (net) efficiency is about 1%. Note that "full concert volume" isn't the same as the acoustic output power of the loudepeakers, though - it is level, frequency, and distance-dependent (at the least).
 
mhardy6647 said:
As to efficiency - here's an example (FWIW), a symphony orchestra at full tilt has an acoustic power output of about 1 watt. If it takes 100 W input for ones loudspeaker's to reach "full concert volume", the (net) efficiency is about 1%. Note that "full concert volume" isn't the same as the acoustic output power of the loudepeakers, though - it is level, frequency, and distance-dependent (at the least).
Click to expand...

Really! That's amazing! I've never heard that example before. That's a keeper.
 
mhardy6647 said:
True - the concepts are pretty much interchangeable for us laypeople, but there are big (and not readily interconvertible) differences in the sound power output, SPL, and perceived loudness. It actually gets unbelievably tricky - as I (re)learned a couple of nights ago looking at specs for kitchen range vents. The noise level is rated in "sones". Now, you can convert "sones" to "phons", but you cannot directly convert either to decibels, not to mention, say, dBA (A-weighting).

As to efficiency - here's an example (FWIW), a symphony orchestra at full tilt has an acoustic power output of about 1 watt. If it takes 100 W input for ones loudspeaker's to reach "full concert volume", the (net) efficiency is about 1%. Note that "full concert volume" isn't the same as the acoustic output power of the loudepeakers, though - it is level, frequency, and distance-dependent (at the least).
Click to expand...

In an anechoic chamber, there is sound emission emanating directly to you which is where the mic is located for standard measurements. This measures just one vector of the sound emission and the value you get is 1 watt of power. If you consider the backside and take a measurement, you will also measure 1 watt.

Because these vectors are additive in Newtonian physics (our reference frame for our senses), this is 2 watts. If you add up all possible vectors from the origin (the orchestra), you will find much more than 1 watt if you can subtract the weaker cone field from each vector. A bit abstract but the point is that the current measuring technique seems to me to understate the true efficiency which is the conversion of an A/C signal in a copper wire to sound waves propagating in air.
 
You must log in or register to reply here.
Back
Top Bottom