Announcement

Collapse
No announcement yet.

Where to place the high-end components

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

  • Where to place the high-end components

    Hello,

    I have started a refurbishing project for a friend and now, after 40 years, all the caps and possibly some of the resistors will be replaced.



    This is a “phase-linear” design based on B&O M80 speakers from the 70-ies.



    My question for you is: due to obvious cost reasons but also space limitations, where in this design should I put high-end caps? And where, if any, high-end resistors?

    Regards//lasse


    Stockholm, Sweden

    Click image for larger version

Name:	image_82954.jpg
Views:	214
Size:	53.1 KB
ID:	1431123
    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
    That depends on what you mean by high end. Compared to what's probably in there now Dayton DMPC polys would be a major improvement. As for the resistors I can't imagine anything more expensive than Jantzen Superes being worth the price.
    www.billfitzmaurice.com
    www.billfitzmaurice.info/forum

    Comment


    • #3
      Originally posted by billfitzmaurice View Post
      That depends on what you mean by high end. Compared to what's probably in there now Dayton DMPC polys would be a major improvement. As for the resistors I can't imagine anything more expensive than Jantzen Superes being worth the price.
      Hi and thanks for that Bill,

      My thought is that the cap-values in this XO directs towards rather expensive components, (apart from a few 8 uF ones) and I wonder if you could advice me if there is any position in the XO where ordinary electrolytics are OK? Polys in the 50uF-rage are quite expensive.

      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
        NPE are fine where they're used as shunts to ground in low pass filters. Their higher ESR isn't in the signal path so it won't affect response within the driver pass band.
        www.billfitzmaurice.com
        www.billfitzmaurice.info/forum

        Comment


        • #5
          So, you think that the higher ESR is the "problem" with electrolytics?

          Does this mean that, practically speaking, when there is a resistor in series, a electrocap could be OK?

          Regards//L
          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
            If I'm readin' that right, that's 4 coils, 9 caps, and 10 resistors?
            That's a LOT of 4.7n(ohm) resistors. 6? Curious.

            Without having actual .zma files, the 12" woofer filter looks like it could be simplified. (The 47n "bypass" resistor SEEMs superfluous to me?)
            The pair of RC (shunt) legs could be VERY closely approximated by a single leg using 1.7n and 70uF.

            Comment


            • #7
              Must have had loads of old 4.7 ohm resistors laying around.

              Comment


              • #8
                Well, I don´t know if I have shown you all this design before but it is more or less a replica of the XO in B&O M75-system.
                In the 70ies, we had no measuring equipment or simulation SW, so we copied, listened and tweaked… and tweaked…..and tweaked.

                But back to the question. Do you think that 2 8uF and 1 37uF caps should be "highish-end"? The rest OK with cheaper electrocaps??

                Regards//lasse

                PS// I hesitate to post the M80 XO. Perhaps that would mean violating something danish… DS
                Last edited by lasse; 02-05-2020, 01:26 PM.
                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


                • #9
                  Originally posted by lasse View Post
                  Does this mean that, practically speaking, when there is a resistor in series, a electrocap could be OK?
                  Anytime you have a cap what you actually have is a cap and a resistor in series. For that matter all caps exhibit some resistance and inductance, all inductors exhibit some resistance and capacitance, and all wire wound resistors exhibit some capacitance and inductance. Where either a cap or inductor is in the signal path, as opposed to shunting to ground, you want the resistive component as low as possible. You get that with polys. It's easy to see how low the resistive component of an inductor is, it's shown by the DCR. That, BTW, should be kept at 5% of the driver impedance or lower. As for non-inductive wire wounds, there's no such thing. A more accurate description of wire wound resistors made for crossovers would be low inductance.
                  www.billfitzmaurice.com
                  www.billfitzmaurice.info/forum

                  Comment


                  • #10
                    And I guess that what I am really asking you, more knowledgeable ( spelling !?) in series XO design than I am. Which is/are the signal path (s) in the upper part of the OX, do you think?

                    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


                    • #11
                      I would say in the upper part that all caps are in series. Even the 12uF "bypassing" the "KM" driver (along w/a 4.7n resistor) I believe is feeding the "TW" (tweeter?).

                      Still...
                      it wouldn't bother me at all to use npes on all the big guys: 25, 30, 37, and 50s. They nearly all have some series resistance inline w/them anyway?

                      Comment


                      • #12
                        Originally posted by Chris Roemer View Post
                        I would say in the upper part that all caps are in series. Even the 12uF "bypassing" the "KM" driver (along w/a 4.7n resistor) I believe is feeding the "TW" (tweeter?).

                        Still...
                        it wouldn't bother me at all to use npes on all the big guys: 25, 30, 37, and 50s. They nearly all have some series resistance inline w/them anyway?
                        Hi Chris and thanks for the advice,

                        Your suggestions seem to be logical and gives me a "line of thought" to follow.

                        After all, this is a replicated XO from a design with completely different drivers. I know what you think of that but the resulting sound is amazing and the owner would like things to stay as close to that as possible.

                        Once again, Thanks.

                        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


                        • #13
                          Originally posted by billfitzmaurice View Post
                          NPE are fine where they're used as shunts to ground in low pass filters. Their higher ESR isn't in the signal path so it won't affect response within the driver pass band.
                          What about in parallel RLC notch filters where it's also shunted to ground?

                          Comment


                          • #14
                            Parallel notch filter components aren't grounded.
                            www.billfitzmaurice.com
                            www.billfitzmaurice.info/forum

                            Comment

                            Working...
                            X