My dipole RD-75 project is 70% done, but I have a question…

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • Victor
    Senior Member
    • Apr 2002
    • 338

    My dipole RD-75 project is 70% done, but I have a question…

    Hello everyone,

    I am getting very close to the 3rd (and final!) stage of my new all-dipole speaker project. My project is a 3-way design consisting of 2 towers. The 300 Hz and up range is taken care by a RD-75 driver in its own wing-shaped baffle. See pictures posted in ‘Dipole or Monopole?’ Thread on the 1st page, here is the link:



    The bass consist of 2 Phoenix W-woofers per side. I used the woofers as shown in Woofer3 in Linkwitz’s site (http://www.linkwitzlab.com/woofer3.htm on the very bottom of the page) utilizing the Peerless 830500 drivers. The two cabinets are setup on top of each other while facing in the opposite direction. The woofers will cover up to 70 Hz.

    So, at this time the RD-75 baffle and the woofers are done! I will post the pictures of this combination as soon as I get my hands on a digital camera.

    I intend to position the midrange baffle on top of the woofers. The midrange baffle will look a lot like the Phoenix main baffle. I intend to make it about 13-inches wide with 3.5-inch wings folded back and 50 inches tall. So the entire woofer-midrange tower will be a 3-unit structure of about 78 inches (14+14 inches for the woofer cabinets and 50 inches for the midrange) in a total height. This tower will eventually be positioned next to the RD-75 baffle, which is incidentally the same height.

    I chose the Scan Speak 25W8565, 10-inch driver for the midrange. The frequency range will be 70-300 Hz, utilizing 4th order L-R filters and appropriate EQ. My question is this, - how many 25W8565 drivers should I use? Initially I intended to use 4 units, simply because I can fit that many in a 50 inch height of the midrange baffle. Do I really need to use 4 units or perhaps I can get away with 3 or even 2? It seems that efficiency-wise even the 2 units may suffice; however I am not sure. Well, any thoughts…?

    Regards,
    Vic

    P.S. Does anyone need a pair of brand new unopened RD-75 units? I bought 4 initially at Bass List special price of $495 each, but I will only be using 2 for my project, so the other 2 units are now ‘collecting dust’ and I would like to sell them.
  • JonMarsh
    Mad Max Moderator
    • Aug 2000
    • 15261

    #2
    Glad to hear you're making so much progress, Victor!

    Regarding your midrange array, that's an interesting question you pose. While classical theory regarding line array's would indicate that a 50" array is sort of short to be a true line array in the 70 Hz to 300 Hz area (based on anechoic operation), the reality is that for a tall array relative to the ceiling height, you will get a "mirror" effect at the floor and potentially at the ceiling (if it's not too high) which may make your system's behavior as an array extend lower than you think.

    Aside from that, here's my take on it. The RD75's match up rather well with ThomasW's double pair of Acoustat 1+1, though using a higher crossover than you plan. In the range from 300 Hz down, I don't think it's reasonable to think that a pair of tens would match the output dynamics of an RD75 or a pair of Acoustat 1+1's. Particularly not in dipole form, where one tends to go with a 4:1 rule of thumb regarding swept area compared with a closed box.

    I've used an older Scanspeak 10" in some systems, but I don't have "hands on" experience with this one. Note that the quoted Xmax is only 6 mm; compared with the Titanic 10" (really an 11) which I'm using in a similar range, it's only got about 1/3 the Xmax. But it is a bit more sensitive, considering it's an 8 ohm driver, not four ohm.

    It just depends, I think, on your SPL and distortion goals. If you're willing to experiment, why not build a test setup for either two per side or four per side, and compare the performance? I suspect that over a wider range of listening distances, the four driver array will provide more consistent balance with the RD75's. But I could be wrong about that.

    Best regards,

    Jon




    Earth First!
    _______________________________
    We'll screw up the other planets later....
    the AudioWorx
    Natalie P
    M8ta
    Modula Neo DCC
    Modula MT XE
    Modula Xtreme
    Isiris
    Wavecor Ardent

    SMJ
    Minerva Monitor
    Calliope
    Ardent D

    In Development...
    Isiris Mk II updates- in final test stage!
    Obi-Wan
    Saint-Saëns Symphonique/AKA SMJ-40
    Modula PWB
    Calliope CC Supreme
    Natalie P Ultra
    Natalie P Supreme
    Janus BP1 Sub


    Resistance is not futile, it is Volts divided by Amperes...
    Just ask Mr. Ohm....

    Comment

    • Andrew Pratt
      Moderator Emeritus
      • Aug 2000
      • 16507

      #3
      Victor I made the URL links in your post clickable. If you'd like to do that in yor next post you simply have to click on the URL button above to make sure that you end up with [*url]www.yourlinkhere.htm[*/url] (only without the *




      Comment

      • Dennis H
        Ultra Senior Member
        • Aug 2002
        • 3791

        #4
        Here's the chart from Jim Griffin's white paper with some gridlines added. It shows the problem Jon's talking about. If you want to maintain line source behavior below 1 kHz, you're depending on floor and ceiling reflections to extend the apparent length of the line. Otherwise, the line's just too short.



        Seems like the ideal would be floor to ceiling drivers with hard floor and ceiling surfaces. Carpet would absorb the highs at the bottom of the array and could make the soundstage unbalanced with the highs seeming to come from a higher point than the mids.

        Comment

        • Victor
          Senior Member
          • Apr 2002
          • 338

          #5
          Jon,

          Yes, the project is going well. However, when I look at the size of the baffle housing the RD-75 standing beside the two stacked Phoenix woofers and then I picture in my mind the midrange baffle affixed on top of the stacked woofer assembly, - the site is truly frightening, as the resulting speaker system is rather huge; and I am talking about one channel here! I have an extremely understanding wife, - oh well…

          Your point about the array behavior is well taken, thank you. The midrange baffle is going to be about 13-inches wide, so considering the 3.5-inch wings; the Fequal is about 340 Hz. That would mean theoretically speaking that a single 10-inch driver in a sealed box will output the same SPL as 4 drivers in the open baffle would output at 85 Hz. This is close to my intended 70 Hz crossover to the woofers. I know that others have successfully used 10-inch driver (although not necessarily the SS 25W8565)
          in a sealed enclosure with RD-75, while extending the response lower then 70 Hz. The Soundline Audio products come to mind.

          So, considering this, the use of 4 drivers would seem prudent. The SS 25W8565 is a well-behaved driver, which is the reason why I chose it. When placed on a 13-inch baffle with 3-inch wings it exhibits a resonance peak at around 320 Hz, but only 2.5-3 dB. The response otherwise follows the theory as outlined by Linkwitz with a peak at 650 Hz and smooth 6 db/oct drop from there. The 10-inch Titanic you are using might be an interesting choice for the open baffle application. Too bad 'partsexpress' no longer carry it. Are you using it in the dipole arrangement?

          Well, I shall report once the midrange is built.

          Regards,
          Vic

          Comment

          • Victor
            Senior Member
            • Apr 2002
            • 338

            #6
            Andrew,

            Thank you for correcting the URL’s for me.

            Vic

            Comment

            • ThomasW
              Moderator Emeritus
              • Aug 2000
              • 10934

              #7
              Victor

              Jon is using the 10" Titanic II, it is a current item in the PE catalog




              theAudioWorx
              Klone-Audio

              IB subwoofer FAQ page


              "Complicated equipment and light reflectors and various other items of hardware are enough, to my mind, to prevent the birdie from coming out." ...... Henri Cartier-Bresson

              Comment

              • JonMarsh
                Mad Max Moderator
                • Aug 2000
                • 15261

                #8
                Yup, the 10" Titanic MkII- I should have been more explicit. It is being used in my baby dipole speaker, both because of the fairly high first cone mode, and because the VC inductance gives me most of my Dipole EQ for free. (I'm kinda chincy about throwing around a lot of money on fancy EQ ) Upper crossover is arround 200-250.

                The 25W8565 does appear to be very well behaved, and I think if you go with 4 per side, you'll be pretty happy with the results, acoustically.

                If you wife is understanding enough to tolerate the kind of system we're talking about here, she's a real jewel... so, if I were in your shoes, I'd think "jewelry" as a little reward for her patience!

                BTW, does she have a similarly understanding sister? Ladies like that are pretty rare... :LOL:


                Keep us posted, Victor!


                Best regards,

                Jon




                Earth First!
                _______________________________
                We'll screw up the other planets later....
                the AudioWorx
                Natalie P
                M8ta
                Modula Neo DCC
                Modula MT XE
                Modula Xtreme
                Isiris
                Wavecor Ardent

                SMJ
                Minerva Monitor
                Calliope
                Ardent D

                In Development...
                Isiris Mk II updates- in final test stage!
                Obi-Wan
                Saint-Saëns Symphonique/AKA SMJ-40
                Modula PWB
                Calliope CC Supreme
                Natalie P Ultra
                Natalie P Supreme
                Janus BP1 Sub


                Resistance is not futile, it is Volts divided by Amperes...
                Just ask Mr. Ohm....

                Comment

                • Hank
                  Super Senior Member
                  • Jul 2002
                  • 1345

                  #9
                  Hey, I was going to ask about a spare sister with that same attitude! Victor, does she have two sisters?

                  Comment

                  • Victor
                    Senior Member
                    • Apr 2002
                    • 338

                    #10
                    Well, guys…my wife does have a sister, but, and there is always ‘but’, you guys are too late! Yes, my ‘toys’ are tolerated to a degree, - they are never loved, but they are tolerated. It would suffice to say that I spend comparable amounts on other things as I spend on drivers. And the drivers along for this project cost a fair amount, to the tune of slightly more then $2K per channel. So, in the end, the speakers will cost me double. How very unfair! But it is the price of this progressive disease we call audio, in the context of a modern nuclear family.

                    The 10" Titanic Mk II looks like a very capable driver, certainly it is more cost effective then the Scan Speaks I intend to use. Perhaps I’ll look at it. It is great Jon that you manage to use it “...without the fancy EQ…” By contrast, the SS25W8656 will need a shallow (Q=1.5) 3 dB deep notch filter at 315 Hz and a shelving 6db/octave low-pass from 20 Hz to about 650 Hz. I will use 4th order L-R at 70 Hz (high-pass) and 300 Hz (low-pass) along with a 1-dB step attenuator that will correct the level from -4 db to +6 dB.

                    I also am now thinking that the comments that Jon made regarding the crossover frequency for the RD-75 is worth a further investigation. I intend to crossover my RD-75 at 300 Hz. Well, Jon feels that a higher crossover point may be beneficial from the power handling and distortion perspective. Hmm…, I guess I can try to go higher and see how I like it. The question is, - how high can I push the 25W85865s? I looked at the frequency response graph and it seems that the driver tops out at about 700 Hz. So crossing it over at 400 Hz may be possible without any detrimental effects. What do you guys think?

                    Regards,
                    Victor

                    Comment

                    • JonMarsh
                      Mad Max Moderator
                      • Aug 2000
                      • 15261

                      #11
                      The 8565 looks pretty smooth for a ways, but as I think you've spotted, the first significant glitch in the impedance curve is at about 700 Hz, even though it's only accompanied by a slight dip in the amplitude response. This is a pretty characteristic Scanspeak trait, in my experience.

                      So, depending on the crossover slopes, 400, maybe 550Hz tops for these drivers. Do try it out. You shouldn't see any comb filtering ripple or dispersion issues below 600 - 700Hz, assuming 4 ten inch drivers, and about 25-50 mm between drivers. Vertical directivity will be increasing above about 350 Hz, which shouldn't hurt anything.

                      If I remember right from one of your other posts, you're rolling your own active crossover, so depending on the topology you're using, trying out a few different frequencies shouldn't be hard.

                      I keep some Marchand XM-9 boards around just because it's so easy to play with them by changing out resistors in the dip freuqency module, and try out stuff "live". My X1 crossovers used a similar topology, though 3rd order, but I didn't lay them out for bundling the resistors in a single header. (silly me).

                      Marchand has a new pasive crossover module the XM-46.



                      I'm going to try those with my NLB modules- but only after I finish my DAC and preamp using them (and am sure I like what I hear under various conditions). This is just my curmudgeonly response to all the opamps that other wise wind up appearing in an electronic crossover! Sometimes I wonder, even when they're pretty nice ones, like the OPA627. It'll probably be like the solid state equivlant of a tube crossover- but direct coupled, and no glo bottles.... (p.s., I got my start with glo bottles as a kid, many years ago, and the first 10 years of electronics I did was about 90% glo bottles...)

                      What's your approach for your active crossover?


                      Best regards,




                      Earth First!
                      _______________________________
                      We'll screw up the other planets later....
                      the AudioWorx
                      Natalie P
                      M8ta
                      Modula Neo DCC
                      Modula MT XE
                      Modula Xtreme
                      Isiris
                      Wavecor Ardent

                      SMJ
                      Minerva Monitor
                      Calliope
                      Ardent D

                      In Development...
                      Isiris Mk II updates- in final test stage!
                      Obi-Wan
                      Saint-Saëns Symphonique/AKA SMJ-40
                      Modula PWB
                      Calliope CC Supreme
                      Natalie P Ultra
                      Natalie P Supreme
                      Janus BP1 Sub


                      Resistance is not futile, it is Volts divided by Amperes...
                      Just ask Mr. Ohm....

                      Comment

                      • Victor
                        Senior Member
                        • Apr 2002
                        • 338

                        #12
                        Jon,

                        My approach to crossover design is all about function. I use IC op-amps such as Burr-Brown 2604 and Analog Devises OP275 and AD744. I also, at least in the past used a discrete op-amp, all bipolar, with fully complementary topology. Over the years that I have been building low-power electronics, I noticed that the qualitative difference between the discrete and integrated circuits have been diminishing rather fast. In fact with modern audio op-amps available today, such as AD797, AD825 and a few offerings from Burr Brown, there is very little impetus for design with discrete components. The THD+N numbers are just as good if not better with IC’s.

                        Although I must admit that I still use 4 transistor ‘diamond’ buffers with filtering application. I also do not shy away from using passive components for my filters. J.W. Miller Company makes some very good inductors and I used them in my modifications to the Denon DCD 2650 CD player I use. Incidentally it is a 16x over-sampling player, so by modifying its 3rd order anti-imaging filter and fixing up its power supply I managed to produce a true 16-bit output with its AD1872 20-bit ladder D/As. Mind you I had to replace the original ‘N’ D/As with better grade ‘K’ D/As. Also this Player does use 2 D/A converters per channel in a push-pull configuration, so I ended up building 4 passive 3-rd orders Bessel filters with those J.W. Miller inductors. So, owing to its rare 16x over sampling architecture only the 3rd Order filter was necessary to recover the full 16-bits. You know, that normally very few CD Players are capably of true 16-bit output. But I digress…

                        My crossover for this project uses op-amps at this point. Circuit topology is quite simple as I use Sallen-Key type 2nd order section for most of the filtering, with the exception for notch filters where I use gyrators. I must say that since my speaker project is a kind of ‘one-off’ effort (providing it is successful!) I may design a circuit board that would incorporate all the crossover electronics along with a typical pre-amp switching capability and step-attenuator-type volume and balance control. This way I get to use a single transformer and a single high quality discrete voltage regulator. Right now I intend to use the LT337 and LT1086 adjustable regulators with massive decoupling. Also I may end up using some of those discrete buffers instead of all the op-amps and use op-amps only where I need gain.

                        The crossover section for RD-75 consists of a 4th order L-R high-pass filter (2 Sallen-Key sections) with a cut-off frequency of 300-450 Hz, with precise number to be determined later, followed by a FNDR-type notch filter. This notch filter is 8 dB deep centered at 5.5 kHz with about 2 kHz bandwidth. Next in the chain is a high frequency spectrum boost section. It is a gyrator-type filter that raises the response monotonically about 3-4 dB from 8 kHz to 13 kHz and then exponentially another 8 dB all the way to 20 kHz. Past 20 kHz the gain of this section drops to unity. This boost should give a nearly flat response to at least 17 kHz. My past experience with 60-inch Carver ribbon supports that assertion. Let me ask a question here, - do I need a baffle-step adjustment circuit?

                        The Phoenix woofers follow Linkwitz’s prescribed cross-over. The Midrange has 4th order L-R hi-pass at 70 Hz and 4th order L-R low-pass at 300-450 Hz, followed by the shelving low-pass (20-650 Hz) with a 200Hz wide 3-db notch at 315 Hz. Again Sallen-Key and gyrator realizations used throughout. The Phoenix woofer and midrange sections will also have a level adjustment capability. I will use a step-attenuator with 1 dB steps from -4 to +6 dB. Here I use an op-amp based circuit that uses liner taper potentiometer (a switch in my case) but approximates a logarithmic scale.

                        Do I miss anything?

                        Regards,
                        Victor

                        Comment

                        • JonMarsh
                          Mad Max Moderator
                          • Aug 2000
                          • 15261

                          #13
                          Very nice description- I don't think you missed anything!

                          Let me ask a question here, - do I need a baffle-step adjustment circuit?
                          The bottom end of the RD75's is variable, depending on the baffle size, but even with smaller baffles it doesn't seem to need any conventional "baffle step" compensation- sometimes they actually need a little shelving in the lower end if you use them with a large baffle, or in a monopole configuration. I suspect that BG is using some staggered diaphram resonance to extend the bottom end- I'll have more opinions on that after doing near field measurements on the RD50's which are now making their way to me. A cursory look at the nearfield response of the RD75 in a moderate size diopole test baffle shows some low end rise, about 6 dB.





                          My suspicion is that the correct choice of baffle size (not to large, not too small) will result in the smoothest lower midrange response, with some built in driver rise being compensated by the dipole cancellation. A larger baffle will tend to augment the low end response and dynamics, but would probably require some adjustment of the amplitude response in the low end.


                          I've ended up looking at gain stages and topologies other than opamps mostly because of my experiences with conventional loop feedback opamp structure power amps, vs. other solid state approaches. I think in the case of very wide band op amps, the problems which arise from loop feedback tend to be pushed upwards to a relatively high frequency, beyond most of the music range. But in also experimenting with output buffers for homemade DAC's, I eventually moved towards some similar line level circuits, based on the input stage concepts I'd been working on for the power amps. It's not super practical, maybe, having +/- 30V supplies for something that outputs 2 VRMS, but I've liked the sound, and it still measures pretty well on my HP8903. My last active crossover with opamps measured very well, and sounded good compared to most commercial units I've heard, and it would drive pretty much any kind of cable setup and load (it should- with output stage buffers consisting of OPA627 cascaded with AD815 video power amps), but I don't think it sounds quite as natural as the more recent things I've been fooling around with. Unfortunately demands at work have kept me from stuffing and testing the PCB's I designed and ordered last summer for the "full tilt" line level test board (which will get used in a new DAC and preamp if it pans out).

                          It sounds like you did a VERY nice job on your Deneon DCD2650- getting true 16 bit output accuracy up to 20 kHz is no mean achievement- few people buying inexpensive DVD players have any idea that most of the early ones could hardly manage 12 bit accuracy in the upper ranges...

                          Keep us informed, Victor! and thanks for the updates and info.

                          Though this isn't Canada's Thanksgiving, I hope you're having a good week while south of your border we take our holiday.

                          Best regards,

                          Jon




                          Earth First!
                          _______________________________
                          We'll screw up the other planets later....
                          the AudioWorx
                          Natalie P
                          M8ta
                          Modula Neo DCC
                          Modula MT XE
                          Modula Xtreme
                          Isiris
                          Wavecor Ardent

                          SMJ
                          Minerva Monitor
                          Calliope
                          Ardent D

                          In Development...
                          Isiris Mk II updates- in final test stage!
                          Obi-Wan
                          Saint-Saëns Symphonique/AKA SMJ-40
                          Modula PWB
                          Calliope CC Supreme
                          Natalie P Ultra
                          Natalie P Supreme
                          Janus BP1 Sub


                          Resistance is not futile, it is Volts divided by Amperes...
                          Just ask Mr. Ohm....

                          Comment

                          Working...
                          😀
                          😂
                          🥰
                          😘
                          🤢
                          😎
                          😞
                          😡
                          👍
                          👎
                          Searching...Please wait.
                          An unexpected error was returned: 'Your submission could not be processed because you have logged in since the previous page was loaded.

                          Please push the back button and reload the previous window.'
                          An unexpected error was returned: 'Your submission could not be processed because the token has expired.

                          Please push the back button and reload the previous window.'
                          An internal error has occurred and the module cannot be displayed.
                          There are no results that meet this criteria.
                          Search Result for "|||"