Announcement

Collapse
No announcement yet.

Sensitivity question.....

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Sensitivity question.....

    Builds - C-Killa - Speedsters - LithMTM - Talking Sticks - Pocket Rockets - Khanspires - Dayton RS Center - RS225/28A - Kairos - Adelphos - SEOS TD12X - Dayton 8 - Needles - 871S - eD6c - Overnight Sensations - Tritrix (ported) - Lineup F4 - Stentorians - The Cheapies - Tub Thumpers - Barbells - Tuba HT - Numerous subwoofers - probably missing a few...... :p

  • #2
    First off, is your sensitivity in watts or volts? If that is the power sensitivity (1w/1m), then paralleling three of those will be an increase of 4.77 dB 1w/1m. If that is voltage sensitivity, then you'll see an increase of 9.5 dB, but that is also because it will be tripling the power (still 2.83v/1m).

    Comment


    • #3
      Builds - C-Killa - Speedsters - LithMTM - Talking Sticks - Pocket Rockets - Khanspires - Dayton RS Center - RS225/28A - Kairos - Adelphos - SEOS TD12X - Dayton 8 - Needles - 871S - eD6c - Overnight Sensations - Tritrix (ported) - Lineup F4 - Stentorians - The Cheapies - Tub Thumpers - Barbells - Tuba HT - Numerous subwoofers - probably missing a few...... :p

      Comment


      • #4
        Originally posted by joeybutts View Post
        So 6.5” has a sensitivity of [email protected] and the 3” has a sensitivity of 91db @1w....
        I'm assuming your ultimate limit of your amplifier will be voltage, so only compare voltage sensitivity. Your 6.5" driver is 96 dB @ 2.83V. If the 3" drivers are 91 dB @ 1w and are 16 ohm drivers (4V drive to get 1 w), then they are only 88 dB at 2.83V. Three of the 3" drivers in parallel will yield a 9.5 dB increase with the same voltage, so a total of 97.5 dB @ 2.83V.

        Comment


        • #5
          Ben, how did you know those figures to add?
          Builds - C-Killa - Speedsters - LithMTM - Talking Sticks - Pocket Rockets - Khanspires - Dayton RS Center - RS225/28A - Kairos - Adelphos - SEOS TD12X - Dayton 8 - Needles - 871S - eD6c - Overnight Sensations - Tritrix (ported) - Lineup F4 - Stentorians - The Cheapies - Tub Thumpers - Barbells - Tuba HT - Numerous subwoofers - probably missing a few...... :p

          Comment


          • #6
            halvedhalved the power by using 2.83V, you lost 3 dB, hence their equivalent rating of 88 dB @ 2.83V, each. So for 3 drivers all at 88 dB, you can "add" it all up by

            10log(10^(88/10) + 10^(88/10) +10^(88/10)) = 10log(3*10^(88/10)) = ~ 92.8 dB. Then you add in another ~4.5 dB for the mutual coupling due to surface area, that puts you at 92.8 + 4.5 = 97.3 dB. Call it 97.5 dB.

            Comment


            • #7
              Originally posted by joeybutts View Post
              Ben, how did you know those figures to add?
              There are a couple different ways to get there:

              First, a 16 ohm driver @ 1 watt is a 4V signal. Ohms law tells us that P = V^2 / R, so:

              1 = V^2 / 16
              V^2 = 16
              V = 4

              To calculate the output at 2.83V, it's a simple calculation of 20 x log (V2 / V1), so 20 x log (2.83/4.0) = -3.0 dB.

              Now, to calculate the increase in output by paralleling drivers, when increasing surface area, you have a 10 x log (n) increase. Adding a two more drivers will be a 10 * log(3) = ~4.5 dB increase due to surface area. However, since you area wiring the drivers in parallel and keeping the drive voltage the same, you also have a 10 x log (n) increase due to power. So, you'll get another ~4.5 dB increase due to tripling the power. You can put that into one equation so that the calculation becomes 20 x log (n) = 20 x log (3) = 9.5 dB.

              Comment


              • #8
                Not to add to or take anything away from the last 2 analytically great posts, but...
                IF you had drivers all rated @ 2.83v, then you could (can) compare them directly (AND, this the rating you have to use in a passive XO design program).
                When they're rated in dB/w, then you have to convert them to dB/2.83v to make comparisons (and for XO design).
                IF you had drivers, ALL rated at 90dB/w, and they were nominally:
                4n(ohms), 5n, 6n, 8n, 10n, 12n, and 16n; then when rated at 2.83v they'd be (and note the progression):
                93dB..., 92.., 91.., 90.., 89.., 88.., and 87dB.

                When you say each driver adds +6dB to sensitivity, that 's ONLY true of the 1st driver, and ONLY if said drivers are paralleled.
                HALF of the +6dB gain is from doubling the surface area, and half comes from halving the impedance (the same as above where the 4n driver is +3dB more sensitive than the 8n driver.
                To get +6dB MORE, you'd need to double the driver count again (so... adding 2 more for a total of 4), and they'd have to ALL be in parallel. Most likely you'd have to start w/16n drivers so you ended up w/a 4n load (since four 8n drivers would drop to 2ohms).

                Wiring drivers in series complicates matters.
                If you parallel two 8n (90dB/2.8v) drivers, you end up at 96dB/2.8v - BUT you're at a 4n load.
                Wiring two 8n (90dB) drivers in series gives you a 16ohm load, but the sensitivity stays at 90dB.
                Wiring one series pair in parallel w/another series pair gets you back to an 8n load, with a sens. of 96dB/2.8v, SAME sens. as a parallel pair.

                Comment


                • #9
                  Some people would argue that comb filtering isn't an audible thing. But it is measurable. Basically this has a lot to do with frequency. I forget the exact rule of thumb but i believe you want your center to center spacing no further apart than the smallest half wavelength you are crossing over to the tweeter. I built a center speaker with two 3 inchers with a center to center at around 7 inches apart crossed at 3000 12 db acoustic with no ill effects.

                  Comment


                  • #10
                    Originally posted by Derek84 ultimax18 View Post
                    Some people would argue that comb filtering isn't an audible thing....
                    not one of them

                    "Not a Speaker Designer - Not even on the Internet"
                    “Pride is your greatest enemy, humility is your greatest friend.”
                    "If the freedom of speech is taken away, then dumb and silent we may be led, like sheep to the slaughter."

                    Comment


                    • #11
                      Builds - C-Killa - Speedsters - LithMTM - Talking Sticks - Pocket Rockets - Khanspires - Dayton RS Center - RS225/28A - Kairos - Adelphos - SEOS TD12X - Dayton 8 - Needles - 871S - eD6c - Overnight Sensations - Tritrix (ported) - Lineup F4 - Stentorians - The Cheapies - Tub Thumpers - Barbells - Tuba HT - Numerous subwoofers - probably missing a few...... :p

                      Comment


                      • #12
                        9 inch wavelength is 1500 hz. Looking at half waves now were at 3000 hz. 3000 hz is where there will be noticeable comb filtering(roughly). Keep in mind though that people build line arrays in there homes with 10 to 20 tweeters that have been happy with there results. Why were you trying to avoid using a tweeter?

                        Comment


                        • #13
                          Builds - C-Killa - Speedsters - LithMTM - Talking Sticks - Pocket Rockets - Khanspires - Dayton RS Center - RS225/28A - Kairos - Adelphos - SEOS TD12X - Dayton 8 - Needles - 871S - eD6c - Overnight Sensations - Tritrix (ported) - Lineup F4 - Stentorians - The Cheapies - Tub Thumpers - Barbells - Tuba HT - Numerous subwoofers - probably missing a few...... :p

                          Comment


                          • #14
                            Originally posted by joeybutts View Post
                            ...Will I still run into issues?
                            yes - Though the1/2 wavelength ( 180 degree off ) is often mentioned; it is relationship ( which varies with wavelength ) that goes from Constructive Summation to Destructive at the120 degree off.
                            "Not a Speaker Designer - Not even on the Internet"
                            “Pride is your greatest enemy, humility is your greatest friend.”
                            "If the freedom of speech is taken away, then dumb and silent we may be led, like sheep to the slaughter."

                            Comment


                            • #15
                              YES, but . . .

                              They call it comb filtering because as you'd run along the line the drivers are on, you could measure (and plot) spots with positive summing, and spots (directly in-between those) where nulls are being generated. The resultant plot tends to look kinda like a comb's "teeth", that is: /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ .

                              If you laid your proposed speaker sideways (like many CCs), then if you played a sine wave and moved a mic across your couch (representing all the places people would normally sit), some folks will be sitting where the teeth are (good sound), and SOME will be in the nulls (between the teeth). If those nulls happen in the vocal range, dialogue would probably be missed as a result. The spacing of the nulls vary by freq. though - so it can be quite a complex issue.

                              If you arranged 3 FRs in a vertical stack (UNLESS you'd put them on a concave radius so all 3 are exactly the same distance from your ears, which means you'd have to have your head in a vise at that "sweet spot") you'll still end up with the nulls being generated, but now at various heights. If each FR driver HAD perfectly flat response, you'd now end up with all kinds of freq. variations, varying by frequency, bascally because you're playing the SAME note with more than one driver and the 2 drivers are different distances from your ears.

                              Comment

                              Working...
                              X