Recessed midrange?

[Shock-Late]

As some of you may know, i'm using EV SM120A midrange horns in place of K400 in my LaScalas. it's been one year now, and i've always been very happy with the sound.

but recently I had to move the speakers into the corners of the room; the bass boost provided by such a placement makes the midrange sound too soft; sometimes I wonder if theres any midrange at all, between the somewhat boosted midbass and the rather harsh treble

i'm using AA x-overs, and until now i was delighted with the sound; but the LS were far from the corners...

I suspect the EV horns to be slightly less "loud" tan the K400, when used with the K55v drivers; but they also sound way smoother and have wider dispersion, and i don't wanna go back to the K400.

Is there any cheap way I could boost the midrange level (x-over mod?) until i have enough cash to buy ALK networks?

Or should i just fight with my flatmates until i can move the speakers back to their previous, out-of-corners place??

thanks for any advice...

[WMcD]

Right now the autotransformer is knocking down the mid by 3 dB. You could get that back by rewiring the crossover. But then again you will have to pay some attention to the crossover points.

Al K. recently posted the schematics for the AA and AL.

If you have the AA, you could rewrire so that the mid is directly fed by the 13 uF cap alone with out the autotransformer.

Now this does have the effect of raising the crossover point of the mid to about 800 Hz. This should be okay because the SM-120 is not going to load the driver much below this anyway, and could be beneficial.

Also, with the AA type (or A) the woofer is crossed over at 400 Hz with the inductor (no cap) and has a 6 dB per octave rolloff. Meaning it is probably going up to 800 Hz fairly well anyway.

Let me point out that 3 dB is not a heck of a lot of improvement.

Also, if you are fooling with the AL with 12 dB slopes (having both a cap and inductor on the bass and mid) there is going to be a lot more to consider electrically.

Overall, you might be better off living with what you've got. Doing crossover work requires some deep knowledge to start with and measuring equipment to get it right.

Best,

Gil

[D-MAN]

+3 db will be approximately twice as loud, if it is +3db.

3 is the magic number in decibels.

DM

[bdc]

I always thought +10db was twice as loud

[WMcD]

If you double the electrical power, that is 3 dB. It is perceptable, and may make a system sound better, or more balanced.

However, it takes 10 dB (10 times the power) to make it sound to our ears like it is twice as loud. (That is 10 1-10th of a Bell. Or, a deciBell is 1/10 of a Bell) This is probably why the works of Dr. Alexander Graham Bell established the unit.

You can investigate it with an RS SPL meter or a meter on a tape deck. It is one the many things in audio which is difficult to believe unless you do it for yourself.

- - - -

I believe this is why we see that using an SPL meter to balance an HT system is so important. Small errors in SPL are important to balancing things and the meter will show them up with much greater accuracy than our ears.

It is somewhat a contradition. Errors in level which are don't sound "twice as loud" are enought to ruin the effect.

I believe this is also true in simple two channel. If one speaker is just a bit farther away, using the balance control can help the image even though there is not a "twice as loud" perception in left versus right.

Gil

[Shock-Late]

Thanks guys!

Just moved the speakers about 6 inches forward into the room... and it solved the problem. Amazing room acoustics!

That said, i'm annoyed what you said about the EV horns. Was sure they were designed for a 400hz crossover, just as in the EV Sentry, and that's why i picked them up. I don't hear any "hole" in the midrange, tho.

[ben.]

A 10 dB increase in SPL is never twice as loud. A 3 dB increase in SPL is a doubling of acoustic energy. Decibels are not meant for subjective observations, they are representations of actual, physical facts.

In other words, if green "looks like" grey to me because I'm colorblind, does that make it so? A 3 dB increase sounds like a doubling, because that's what it is. Maybe some folks (a lot, actually) have trained themselves to think and feel that twice the SPL needs to sound a hell of a lot louder. It doesn't. It's 3 dB.

[WMcD]

Shock,

It may be that both of us are wrong. I checked the spec sheet for the SM120A. "Usable Low -Frequency Limit: 500 Hz"

"...provides precise pattern control over the full frequency range from *00 to 16 kHz."

" . . it shall provide useful acoustic loading at all frequencies above 500 Hz."

The spec sheet for the Sentry III, which uses the SM120 and ST350A says "Crossover Frequency: 600 Hz, 3500 Hz."

- - - -

I believe I picked 800 Hz because I tested mine some years ago and found they did not go down to 400 Hz, and perhaps not even 600 Hz. So I built some midranges based on the K-400.

But EV evidently thinks 600 Hz is a good lower limit.

Best,

Gil

[Arkytype]

"a 3 dB increase sounds like a doubling, because that's what it is."

Let's say you feed your audio power amplifier pink noise or a warble tone and the output power is measured at one watt continuous average power and your SPL meter (set for slow averaging) measures 100 dB (unweighted) at your listening position.

Now, if you increase the amplifier power output to two watts, the SPL reading will (theoretically) increase by 3 dB to 103 dB. You will hear an increase in perceived loudness, but not a doubling as you aver.

Gil is correct: we hear (and see) based on a logarithmic response to stimuli.

Lee

[ben.]

----------------

On 6/10/2004 5:08:52 PM Arkytype wrote:

"a 3 dB increase sounds like a doubling, because that's what it is."

Let's say you feed your audio power amplifier pink noise or a warble tone and the output power is measured at one watt continuous average power and your SPL meter (set for slow averaging) measures 100 dB (unweighted) at your listening position.

Now, if you increase the amplifier power output to two watts, the SPL reading will (theoretically) increase by 3 dB to 103 dB. You will hear an increase in perceived loudness, but not a doubling as you aver.

Gil is correct: we hear (and see) based on a logarithmic response to stimuli.

Lee----------------

I don't know about seeing logarithmically. Never thought about it.

However, you are correct that dB are a logarithmic scale. A 3 dB increase in power is an actual, physical, measurable, and irrefutable doubling of whatever power you happen to be measuring. How you perceive that actual, physical, measurable, and irrefutable doubling is your problem. Your assertion that it doesn't "sound like" a doubling just doesn't make sense. What sounds more like something than the actual thing itself?

[Klipschguy]

Gil is right about the dB scale. 3dB increase requires double the power but is only "perceptable" in how much louder it sounds. 10dB requires 10 times the power, but only doubles the perceived loudness.

Think about it. Take 100dB vs. 110dB - it should sound twice as loud since it is a 10dB jump. This is empirically about right: 100dB is loud, 110dB is very loud.

IF relative loudness doubled with every 3dB, then 110 would be TEN TIMES AS LOUD AS 100dB - which, obviously, it is NOT!!! (The math: 100dB to 103dB would be twice, 103 to 106 would be double again or four times 100dB, 106dB to 109dB would be double again or eight times 100dB, 109dB to 110dB would be ten times the original 100dB.)

It DOES take ten times the power to play 110dB vs. 100dB, but it will subjectively only sound twice as loud.

This is how Klipsch arrives at their theoretical acoustic output figure: 104dB 1M/1W, 10 watts = 114dB, 100 watts = 124dB (1000 watts for 134dB). Each step is 10dB and ten times the amplifier power to the rated output. (Note: These number do not consider losses to heat and mechanical driver limitations which increase with rising output.) BTW, 124dB is NOT, I repeat NOT, 100 times louder than 104dB!!

The decibel scale is a logarithmic scale, the above numbers are just basic application of the physics. The logarithmic scale makes the numbers confusing for some people.

Efficiency is very important for getting a speaker to crank with modest amplifier power.

Warm regards,

Andy

[ben.]

It will sound twice as loud as long as you tell yourself that.

1+1=2, even if 2 looks like 3 or 37 to you. Otherwise, let me know exactly how you managed to change the laws of physics. I believe there may be some practical application to that skill.

[WMcD]

Gee this is odd.

I noticed that a paragraph in my post about the spec sheet didn't show up. The missing paragraph explains the the * is a blot on the photocopy I have. It could be a 3 or 5 or 6 or 8. There is a little curve at the bottom of the blot.

But when I go to edit, the paragraph shows up.

Gil

[WMcD]

I looked up the perceived loudness issue this evening in Beranek's "Acoustics". It seem that Fletcher, the equal loudness guy, did a study. This is not an equal loudness issue per se.

There is a longish explanation and a graph of dB versus how much louder it sounds to the ear.

Above 40 dB SPL, it does take a 10 dB increase to be perceived as twice as loud. That covers most of our situation. My recall is that a quiet living room is about 35 dB.

However, this relation does not hold true at very low SPL's. For example, if you start at around 0 or 1 dB, which is the very lower limit of hearing, going up to 10 or 11 dB in SPL is perceived as a 20 times increase, instead of 2.

There is a corresponding decrease in rate perceived increase as you go to 20 and 30 and 40 dB. As mentioned above, it does go to "twice as loud" per 10 dB SPL above 40.

Best,

Gil

[GooseChaser]

bclarke421,

I think you are mistaking doubling of wattage (3dB increase) with doubling of sound pressure. Those two are not the same thing. Mathmatically, 3dB increase is twice the sound intensity, not to be confused with sound pressure. 6dB increase is technically doubling the sound pressure, but it is widely accepted that 10dB is "perceived" by human ears as twice as loud. 3dB may be twice the sound power per unit mathmatically but what we actually hear is something else.

[Klipschguy]

Believing that speakers will play 100 times louder with a 100 watt input verses a 1 watt input defys all kinds of physics. Talk about efficiency! Holy Moly.

[ben.]

----------------

On 6/10/2004 7:38:19 PM William F. Gil McDermott wrote:

I looked up the perceived loudness issue this evening in Beranek's "Acoustics". It seem that Fletcher, the equal loudness guy, did a study. This is not an equal loudness issue per se.

There is a longish explanation and a graph of dB versus how much louder it sounds to the ear.

Above 40 dB SPL, it does take a 10 dB increase to be perceived as twice as loud. That covers most of our situation. My recall is that a quiet living room is about 35 dB.

However, this relation does not hold true at very low SPL's. For example, if you start at around 0 or 1 dB, which is the very lower limit of hearing, going up to 10 or 11 dB in SPL is perceived as a 20 times increase, instead of 2.

There is a corresponding decrease in rate perceived increase as you go to 20 and 30 and 40 dB. As mentioned above, it does go to "twice as loud" per 10 dB SPL above 40.

Best,

Gil

----------------

That is an interesting study, and one of which I was not aware, though it certainly makes sense. My position is that those perceptions are in error, that's all. If we continually harbor comfort in false perception, then that will become our reality. I'm saying that twice as loud isn't as big a difference (at certain SPL, I now understand) as we would expect.

[ben.]

----------------

On 6/10/2004 7:59:36 PM GooseChaser wrote:

bclarke421,

I think you are mistaking doubling of wattage (3dB increase) with doubling of sound pressure. Those two are not the same thing. Mathmatically, 3dB increase is twice the sound intensity, not to be confused with sound pressure. 6dB increase is technically doubling the sound pressure, but it is widely accepted that 10dB is "perceived" by human ears as twice as loud. 3dB may be twice the sound power per unit mathmatically but what we actually hear is something else.

----------------

No, I understand the difference between power output measured in watts and SPL, commonly expressed in dB, measured relative to a nominal threshold of audibility. They are still both measurements of power, where 3 dB is a doubling of a linear measurement. If you're measuring a voltage, that's where your 6 dB is a doubling.

[jonwolfe]

this is a test...thanks

[D-MAN]

It can get confusing when you have to go back and forth between electrical and acoustical values, such as when discussing crossover networks, and such.

This gets especially confusing for me as I have a Mc300 amp which has power meters in Watts, deliniated by 3x increments from 0 to 300 watts. For example, 3W reference +10x wattage = 30 Watt = twice as loud from 10x the power or +3db, or one increment on the meter. So I tend to relate acoustical output with wattage, but that is technically wrong. Loudness is measured in decibels ONLY.

Electrical is represented in watts which is a measurement of heat disapated over time, or voltage and/or current.

+3db increase in SPL is twice as loud as the reference level(0db). This is acoustical level of signal only.

10 times the reference ELECTRICAL WATTAGE (current*voltage) will result in a +3db ACOUSTICAL signal level change, which is exactly TWICE as loud as the reference acoustical level represented by a 0db reference. Note that even if the acoustical reference is 100db SPL, a +/-3 db change is either (+) twice as loud or half as loud (-) as the reference (100db) level. The 3db reference is often used in speaker sensitivity ratings. Watch out for the above +/-3 db variance to be noted in sensitivity ratings, such as -5, +10db, etc.

The electrical power (current*voltage) is usually measured in watts which is a measurement of absorbed and radiated heat in a given period of time, typical of reactive components.

Some measurements of "sensitivity" are rated in watts by volts input with the measurement device placed at a specified distance (2.85V @ 1 meter) or a measurement of watts per distance to measuring device nndb SPL (1w @ 1 meter, etc). The voltage rating perhaps being the more precise, as the value of current can be calculated from it using P=iE where P=power in watts, i=current, E=voltage.

DM