Announcement

Collapse
No announcement yet.

Peerless 830452 Subwoofer - Strange results

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

  • Peerless 830452 Subwoofer - Strange results

    I bought a Peerless 830452 for my car, but before buying it, I simulated it, in a sealed enclosure. There are two versions of the datasheet, but my subwoofer should be the latest one (Slightly higher Qts).

    Just for completeness, the simulated old version is here (Qts 0.18):

    Click image for larger version  Name:	uy4flRq.png Views:	6 Size:	319.7 KB ID:	1486465

    New version here (Qts 0.22)

    Click image for larger version  Name:	sZelQgT.png Views:	6 Size:	322.2 KB ID:	1486466


    Here's a simulation using the measurements I took with my DATS V3. The driver was blasted with a 20hz signal for a few hours, so nicely run-in.

    Click image for larger version  Name:	KJwSPGZ.png Views:	6 Size:	305.8 KB ID:	1486467

  • #2
    I built a cabinet with an internal volume of 11ltr. The datasheets don't specify the driver displacement, so I guessed around 2ltr for a net internal volume of about 9ltr.

    Here's the measured result:

    Click image for larger version

Name:	Y6ln6il.png
Views:	160
Size:	78.8 KB
ID:	1486469

    I've got very similar results to the new datasheet, but was expecting to see results similar to my simulation with the DATS V3 data. Strange

    The other strange thing is when I add an aperiodic vent, Qts goes up! In simulations, it should go down. Here's the same enclosure but with an aperiodic vent.

    Click image for larger version

Name:	AA8x2dZ.png
Views:	144
Size:	77.9 KB
ID:	1486470

    What's going on?

    Comment


    • #3
      The T/S parameters look difficult to read, so I'll show them again here:
      Measured in a 11ltr sealed enclosure:

      Click image for larger version

Name:	p6FGnar.png
Views:	146
Size:	124.5 KB
ID:	1486474

      And in the same enclosure with an Aperiodic vent:

      Click image for larger version

Name:	wp4a5lx.png
Views:	142
Size:	123.8 KB
ID:	1486475

      Why does R(e) change so much, too?



      Comment


      • #4
        Your RE changed, maybe a lead wasn't tight or dats needs the leads recalibrated.

        Comment


        • #5
          T/S parms are "free air" (so, no enclosure), although I've always believed that the Qts readout (if taken "in-box") SHOULD be equivalent to Qtc (the "in-box" Q).

          To ME, your 3 sims (and 2 DATS plots) don't look significantly different from one another.
          DATS should let you simultaneously display up to 20 curves. It would be easier to compare your 2 curves if they were plotted on the same background.

          Re (the DC resistance component of the voice coil) would (technically) be at zero hertz (so, DC, not AC).
          DATS is probably picking that value up BELOW your 20Hz plot limit. Set the low freq. to 5Hz (or lower, if it goes lower) and you'll SEE the DCR value plotted as impedance at the left edge.

          As far as your A-vented box being higher Q? To ME, it's "lower" Q (just looking at the plots) since the resonance peak is lower (considerably lower, in fact, it LOOKs like your Z-peak (at resonance) dropped nearly in half, from about 50ohms down to 25ohms). Every impedance device (or simulation - that I know of) measures qEs and qMs (the electrical and mechanical resonances), then determines Qts by calculation (like WinISD does). The formula is not complex. The fact that DATS SEEMs to calculate a (slightly) higher Qts value w/your A-vent is just because you're not dealing strictly with "T/S" values - since you're "in-box". As I've said, to ME your A-vented box has a lower "Q" because the Z-peak is lower (making the resonance less sharp).

          Comment


          • #6
            Originally posted by Chris Roemer View Post
            T/S parms are "free air" (so, no enclosure), although I've always believed that the Qts readout (if taken "in-box") SHOULD be equivalent to Qtc (the "in-box" Q).

            To ME, your 3 sims (and 2 DATS plots) don't look significantly different from one another.
            DATS should let you simultaneously display up to 20 curves. It would be easier to compare your 2 curves if they were plotted on the same background.

            Re (the DC resistance component of the voice coil) would (technically) be at zero hertz (so, DC, not AC).
            DATS is probably picking that value up BELOW your 20Hz plot limit. Set the low freq. to 5Hz (or lower, if it goes lower) and you'll SEE the DCR value plotted as impedance at the left edge.

            As far as your A-vented box being higher Q? To ME, it's "lower" Q (just looking at the plots) since the resonance peak is lower (considerably lower, in fact, it LOOKs like your Z-peak (at resonance) dropped nearly in half, from about 50ohms down to 25ohms). Every impedance device (or simulation - that I know of) measures qEs and qMs (the electrical and mechanical resonances), then determines Qts by calculation (like WinISD does). The formula is not complex. The fact that DATS SEEMs to calculate a (slightly) higher Qts value w/your A-vent is just because you're not dealing strictly with "T/S" values - since you're "in-box". As I've said, to ME your A-vented box has a lower "Q" because the Z-peak is lower (making the resonance less sharp).
            Thanks, Chris, that's very helpful. I was just surprised that Qts/Qtc in my simulations didn't match what I measured in the enclosure. Using the datasheet, they do, but using my free-air measurements with the DATS, they don't. I thought it would be the other way around.
            I was hoping that my simulation would match the measured result, mainly because I'd like to adjust the aperiodic vent to have just enough resistance that it doesn't cause the "Cone SPL" (Jeff Bagby - Woofer Box Model and Circuit Designer) to be too much above the "Total SPL". If I go extreme, and set QI to 1 (very leaky), rather than around 30/40, Jeff Bs WBMACD shows an 8db difference between Cone SPL and Total SPL at 20hz (you lose 8db of bass at 20hz).
            A QI of about 10 looks good (not much difference between Cone and Total SPL), but how do I match that in reality? I know that when people adjust resistance in an aperiodic vent they're looking at reducing the impedance peak, but surely it's the reduction of internal pressure that's the benefit of an aperiodic vent? I assume it's the back wave from the aperiodic vent, interfering with the front wave (phase cancellations) that cause the 'modelled' loss in output.
            Just had a thought - I could use the aperiodic vent to adjust the low frequencies. If there's too much output at 20hz in the car, I could reduce resistance in the vent to lower output, or vice versa. I'll take some measurements when it's in the car to see If I need to make any adjustments...

            Comment

            Working...
            X