Announcement

Collapse
No announcement yet.

Low impedance at high fq´s-What to do? (ref RFL-3way)

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

  • Low impedance at high fq´s-What to do? (ref RFL-3way)

    Hello,



    -The next iteration for the RFL-3way can be seen in the attached picture.

    -It is difficult to splice in the mid´s low-Fq-response from nearfield measurements so here IRL-measurements will have to tell….



    Pros:

    - Fairly good integration between mid and tweeter

    - Good phase-tracking in the overlap region (1-5 kHz)

    - Fairly few parts in the XO (Havent ronded off the optimizers values yet)



    Cons:

    - Large overlap region (2nd order soo….. )

    - Large bump in mid SPL 3-7 kHz, messing up the phase tracking

    - Very low impedance over 1 kHz



    I think that my main remaining issue in this before prototyping a XO and begin IRL measurements would be on how to take care of the impedance issue.



    Please chime in and comment.



    Regards//lasse
    Attached Files
    Perry Mason talking to his dentist:

    "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

  • #2
    Did the post go away?
    Perry Mason talking to his dentist:

    "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

    Comment


    • #3
      Hi again,

      Sorry, I just saw that I have made a misstake in the impedance plot for the mid.

      Will be back with new simulations.

      (But I Think that the trouble with low impedance at high fq´s will remain...)

      Regards//lasse
      Perry Mason talking to his dentist:

      "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

      Comment


      • #4
        27uf on tweeter?
        That seems awfully large.....

        Comment


        • #5
          Originally posted by kevintomb View Post
          27uf on tweeter?
          That seems awfully large.....
          Hi Kevin,

          yes I will make a complete remake of this.

          Will be back!

          Regards//lasse
          Perry Mason talking to his dentist:

          "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

          Comment


          • #6
            I agree on the 27uf being large for that Fo. I could see if it were a 3rd order topology and you needed that value to pull the phase in line, but it seems way too large for a 2nd order at 2K.

            I don't understand the 20ohm resistor across the inductor. It might not be wrong, but I just haven't seen that in a parallel crossover before. Could that be causing that dip and quick rise starting at about 600hz? As for the impedance in the treble, I think you're seeing the effects of no padding on the tweeter.

            I could be wrong about all of the above - I was wrong one other time....
            Last edited by tom_s; 10-05-2018, 01:32 PM. Reason: words and stuff
            Co-conspirator in the development of the "CR Gnarly Fidelity Reduction Unit" - Registered Trademark, Patent Pending.

            Comment


            • #7
              Hi,

              we all run the obvious risc of being wrong but if you never expose yourself, both by showing work-in-progress and commenting on it, you will never learn and eventually improve in a number of ways.

              The unusually high tweeter-cap-value (and it is getting higher!) is a result of the optimizer, trying to match my tweeter with the LR2 target at 2500 Hz, I will show you the complete design in a few days but I want to check and double-check a few things first. I´m using Speaker Workshop and when using that vintage software you end up with a bewildering number of files in no time at al so.....

              the 20 ohm resistor is a "trick", sometimes used for shaping the SPL in the lower region of the "woofer"-XO, I Think.

              And, by the way, I don´t Think that yoy are wrong.

              Regards//lasse
              Perry Mason talking to his dentist:

              "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

              Comment


              • #8
                FWIW - I kind of miss Speaker Workshop. Too bad they didn't continue to upgrade & support it. Yep - lots of files to sift through to find the chart you really wanted to view. I almost forgot about having to update/calculate every change you made. I'm surprised you're not using Xsim - it's so much quicker for crossover design.
                Co-conspirator in the development of the "CR Gnarly Fidelity Reduction Unit" - Registered Trademark, Patent Pending.

                Comment


                • #9
                  If I was designing a 3-way x-over, I'd start by deciding what sensitivity the speaker was going to be. That will depend on the woofer sensitivity, and on the amount of BSC desired. Then you can determine how much padding the mid range is going to need. If the woofer is 88dB, and the design is 3dB of BSC, then the mid range needs to be padded to produce ..... 85dB. If the mid range driver is 87 dB, then it needs to be padded 2dB. The padding can help increase the impedance, but you still have to be careful, or it will be low in spots. Low at high frequency might be caused by a low pass coil on the mid range that is too small. The nice thing is, making that coil larger will also ten to pad the mid range.

                  Comment


                  • #10
                    What drivers are being used?

                    Comment


                    • #11
                      Originally posted by kevintomb View Post
                      What drivers are being used?
                      Hi,

                      Sorry for "spreading" the build over a number of threads. Here are the previous ones and in there you can read what drivers are used and in which configuration.

                      http://techtalk.parts-express.com/forum/tech-talk-forum/1387333-design-topology-for-a-classic-3-way

                      http://techtalk.parts-express.com/forum/tech-talk-forum/1378295-to-select-a-10-woofer

                      http://techtalk.parts-express.com/forum/tech-talk-forum/1389631-first-trial-xo-for-rfl-3way-mid-tw-unit


                      Yes "rpb", you are of course absolutely right on that the first thing to do is to match the sensivities of the chosen drivers. I must admit that there is a number of years since I´ve been designing passive crossovers. The last two complete systems have been Active ones, both analog and digital. Then you only crank up the appopriate level to fix the difference in sensivity.

                      Best regards//lasse
                      Perry Mason talking to his dentist:

                      "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

                      Comment

                      Working...
                      X