Announcement

Collapse
No announcement yet.

Help pick a midrange drive for vocal

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

  • Help pick a midrange drive for vocal

    Hello everybody, I am new here and recently I completed a little DIY 2 way speaker project with a TangBand W6-1721 woofer and a HiVi RT1.3WE Tweeter. After running in for a few days, the sound is getting better. This setup plays Jazz, classical guitar, orchestra music and in fact any instrumental music I throw at it very well but whenever it comes to vocal, it just seems to lack that bit of warm, forward sounding and full body sound. Now I am considering of adding a good midrange driver to help with the vocal and I have a few shortlist on my mind.

    1) Peerless NE123W-08 4"
    2) Dayton Audio RS100P-4 4"
    3) SB Accoustic SB12MNRX25-4 4"
    4) Scanspeak Discovery 10F/4424G 4"

    Which of these drivers would you recommend that is more capable to produce that realistic, warm and forwarding sounding vocal? Preferably not too bright or harsh sounding when listening at a higher volume. If you have a better recommendation, please feel free to do so. TIA.

  • #2
    The Dayton and the Scan probably would reach the tweeter better, but all those are good choices. I'd try tweaking the crossover first though.

    Comment


    • #3
      You might also consider the SB12CACS25. PWRRYD paired the 5.25" version with the same tweeter and it sounded fantastic. As with any speaker project, the crossover is the deciding factor. But from what I've read, the W6-1721 is really not a slouch in the midrange department. I would start by doing some tweaking to the current crossover and see where that might get you.
      Co-conspirator in the development of the "CR Gnarly Fidelity Reduction Unit" - Registered Trademark, Patent Pending.

      Comment


      • #4
        Originally posted by chrisn View Post
        The Dayton and the Scan probably would reach the tweeter better, but all those are good choices. I'd try tweaking the crossover first though.
        I agree, seems phase alignment could be adjusted so the drivers integrate better. This can make or break a speaker.

        Comment


        • #5
          Originally posted by tom_s View Post
          You might also consider the SB12CACS25. PWRRYD paired the 5.25" version with the same tweeter and it sounded fantastic. As with any speaker project, the crossover is the deciding factor. But from what I've read, the W6-1721 is really not a slouch in the midrange department. I would start by doing some tweaking to the current crossover and see where that might get you.
          Hi Tom, thanks for the advice. I definitely agree with you that the W6-1721 is not a slouch in the midrange department. This TB and HiVi combo is sounding so much better than my Monitor Audio MA700 and B&W DM601 when playing classical, chinese orchestra music, Jazz, strings musical piece and so on. All the musical instruments sounded so real & detail, the bass is tight and punchy. Even the sound from double bass and bass guitar you can feel their presence. When everything else sounded so good, why I find the vocal not to my liking? Somehow I just feel that the TB and HiVi combo does not have that vocal sound signature I am looking for, which is more warm and forward sounding.

          At first I am also thinking could it be the XO problem, but shouldn't everything sound bad if it is a faulty XO design? Or could it be more of a personal taste problem? Btw the XO frequency is set around 2.3kHz which is slightly higher than the recommended value at 2kHz.

          Comment


          • #6
            The speaker should not have a sound of it's own. It should reproduce what it is given. I have used several of the RS series and find them very good for their price.
            The SB I also like. I find Scan Speak to need more work to tame and quite expensive. A three way is a lot more complicated to do than a 2-way. Not a good start for a relative beginner.

            I would have to know a lot more about your design, but I suspect your problem is not lacking a mid-range driver, but a problem in the crossover. The TB should easily be able to cross over @ 2500 or so if you deal with the first breakup modes. Both drivers look on paper to be pretty well behaved. Do you have a calibrated mic? What simulation did you do for the crossover? What electrical order and what effective acoustic order? Did you compensate for the AC offset, baffle step? Cabinet baffle edge radius? What about the room? Is the speaker more than a couple feet from any reflective surface? You need to know what you think is lacking in measured frequency terms. For that mater, measure your others and see what the differences are. Just looking on paper, either a dip at crossover or a hump from the breakup would be my first suspects for an out-of balance system.

            Monitor and the B&W are both highly respected speakers. Not a lot on the bottom end and in need of a sub. B&W always a bit laid back, but both very good products.

            Comment


            • #7
              Hi tvrgeek, I agree with most of your comments and observation but somehow I tend to disagree on the part where you say speaker should not have a sound of it's own. Different material itself already produce quite different sound especially when used in the tweeter. Also a compress driver tweeter will sound pretty different when compared to a soft dome tweeter.

              As you mentioned, I am a beginner going in blind with no measuring equipment. All my ZMA and FRD files are obtained from manufacturer spec sheet. I used Xsim as the simulator when designing my XO. As a beginner, many of the terms you mentioned sound very new to me and I fully understand many factors need to be taken care to build a good speaker. This DIY project started out as a challenge and also a bit of fun to see how far I can achieve with the limited resources I have. So far I am pretty happy with the result.

              I also like to have a discussion about comment saying the lack of warm vocal and not forward sounding is due to a XO design fault. Just to be clear I am not defending my XO design but a serious discussion on this topic. I have been into HiFi for the past 30 years and I also used to be a member of a Chinese Orchestra group playing flute. So when I say my current setup sounds realistic, clear and detail, I know what I am saying. Now the question is, if the current XO design is giving me all the accurate and detail sound regardless in the low, mid or high region for instrumental tracks, than how is it possible the XO design is flaw just because the vocal is not warm enough and not forward sounding enough? Please don't take it the wrong way, the vocal from my current setup is not unbearable or lousy in anyway but rather not to my taste. I seriously doubt by changing the XO design, one can made a speaker sound warmer or more transparent. Of course, this is open for discussion. Cheers. Btw, attached is the FR from Xsim simulation for my TB and HiVi combo.
              Click image for larger version  Name:	2wayTBHiViFR.JPG Views:	0 Size:	108.5 KB ID:	1451562

              Comment


              • AEIOU
                AEIOU commented
                Editing a comment
                A loudspeaker should be a reproducer of sound, not a producer of sound. Individual components should be as neutral as possible. True, different drivers have their own unique sonic signature.

              • JamesTan
                JamesTan commented
                Editing a comment
                Hi AEIOU, I totally agree with your comment.

            • #8
              The crossover design and phase matching round Fc will definitely play a large role in transparency and imaging. Unfortunately, you can't do this by ear.
              If you have a 2nd order LP on the woofer and a 3rd order on the tweeter (very common), you can reverse the polarity of the tweeter and look for a deep reverse null at Fc and
              that will show you good phase matching.
              You can get a Dayton UMM for about $80 and run ARTA or HolmImpulse for free. I would not advise adding a midrange at this time.
              The complexity becomes about 10x harder to get everything working well.(in my experience)
              Hope this helps!

              I think I hear a difference - wow, it's amazing!" Ethan Winer: audio myths
              "As God is my witness I'll never be without a good pair of speakers!" Scarlett O'Hara

              High value, high quality RS150/TB28-537SH bookshelf - TARGAS NLA!
              SB13/Vifa BC25SC06 MTM DCR Galeons-SB13-MTM
              My Voxel min sub Yet-another-Voxel-build

              Tangband W6-sub

              Comment


              • JamesTan
                JamesTan commented
                Editing a comment
                Hi donradick, I used 3rd order on both the woofer and the tweeter with a notch filter to tame a bump around 2.6kHz region. I have already tried reversing the polarity of the tweeter but it only make the speaker sounds worse with lost of details. Your advice is appreciated.

            • #9
              AFAIK, the only way to get +6dB of summing @ Fc is to have nearly perfect phase alignment - and (if your sim is accurate) you already have that.

              Comment


              • #10
                JamesTan - echoing what many have said, your tweeter and woofer have been used in well-received designs so it is likely the crossover.

                For measurement if you really want to save money you might be able to get away with the inexpensive iMM-6 microphone but you would need several additional cables to get it to work with a laptop so the UMM microphone, which you could turn around and sell on eBay or the Tech Talk Classifieds if you weren't going to use it again. Measuring in the actual cabinet with the actual baffle dimensions and driver placement on the baffle matter, as well as the fact that the manufacturer specs may not be that accurate. Also, where do you live, because there might be someone nearby who has the measuring equipment to help you?

                Comment


                • JamesTan
                  JamesTan commented
                  Editing a comment
                  Hi a4eaudio, as mentioned I am totally loving this TB and HiVi combo so no doubts on how well these drivers can perform. I live in Singapore, the audio DIY group here is extremely small and not to mention it is extremely difficult to get components here, ready made components like the air core inductor is almost non existence. The next best option is to get from online stores like Part Express or Madisound but the shipping cost is a killer. Thank you for the tips.

              • #11
                If your 2way's woofer and tweeter can meet somewhere in the middle without noticeably distorting or beaming, then switching to a 3way doesn't really have much/any advantage that I can think of.

                Your current 2way might have less-than-ideal summing at the crossover because of phase/Zoffset that might not be fully accounted for in that SIM, or there might be a dip in the mids from some reflection/cancellation, or some of the lower-mids might be lost from BaffleStep and not fully made up for in the XO right now, or the entire FrequencyResponse could be really flat/smooth like the SIM and you simply prefer a slightly exaggerated mid....any of these can be changed by tweaking the crossover.

                From your description it doesn't sound like you're suffering from bothersome distortion or beaming, so an XO tweak seems like the better fit.

                Some music can be more forgiving OR more likely to bother you in certain frequencies than others. I've personally noticed more bothersome flaws listening to spoken voice recorded in specific room conditions from speakers that I otherwise really like with all kinds of music...so don't be too surprised if you only really notice problems with some singers while many other kinds of music seem to sound fine.
                Adding a little more lower-mids/mids for extra warmth and forwardness within reason might not even be noticeable for a lot of the other things you listen to, and that should be possible at the crossover (maybe by switching the LowPass to use a slightly smaller inductor and re-tweak the rest of the XO to work around the change).


                EDIT: I might've missed it, but if you haven't already posted what parts/values are used for the current crossover, sharing that might help along with the front baffle's heightXwidth (and roundover radius amount if possible).
                For anyone super familiar with those drivers it might also help if you can share how they're mounted into the baffle (inset/flush or simply top-mounted, angled or flat) and how the speakers are positioned in height and tilt compared to your ear height when listening....I'm assuming someone might know the driver phase/Z-axis offset parameters if they've already measured these specific drivers before. This can be pretty important for XO summing.
                My first 2way build

                Comment


                • JamesTan
                  JamesTan commented
                  Editing a comment
                  Hi LOUT, thank you for your very detailed reply. Really glad to know I am not the only one have "problem" with vocal while many other kinds of music sounded really good on their speakers. I agree adding a little more to lower-mids/mids will make the vocal sound more pronounce but I kind of worry will it make the speaker sound too "shouty" or too bright. What frequency range you think I should try to boost for that stronger vocal without making it sound too harsh? 200Hz-2000Hz?

              • #12
                Just an analytical observation here:

                At this point if the OP has a multiband EQUALIZER - would now shaping the response of the speakers to match his desired mid FR not indicate where the XO/speakers are falling short of his desires such that he could take the boost/cut settings as a guide to further refine the XO?

                Comment


                • LOUT
                  LOUT commented
                  Editing a comment
                  Seems like a good way to try a bunch of different small boost positions to see where and how much tends to work and when it definitely gets too far or not enough.
                  Might even be able to use a free software EQ (plug the computer into the soundsystem and tweak around) if you don't have easy/cheap access to a physical one.

              • #13
                Good start. But the manufacture files are only good for purchase decision, not design. You need to measure your drivers in your baffle. I suggest @ 1M. You don't listen nearfield where the factory curves are measured. If you measure real, you may find some large surprises. Not that the manufactures are not being truthful, but drivers vary, your baffle design matters, driver placement, acoustic offset, baffle step, on or off axis listening etc. I was surprised how big a difference routing a radius on the BACK edges of my cabinets made. Everything matters so measure!

                Yes, drivers have their own signature. Another way of saying each has their own defects different from other driver defects. If you look at a pulse to a tweeter in a 3 dimensional waterfall, they will show their unique defects. None are perfect. The amount of HD will also greatly change the character.

                Changing the crossover absolutely changes the "presence" A couple dB here and there can make a huge difference. Maybe you prefer the old "West Coast Sound" or JBL presence. ( What I call Just Bloody Loud) I prefer the much flatter sound. If the artist and producer want a raised mid, then I rely on them giving it to me. I suspect a defect in the response though. Is the balance of your B&W and Monitors to your liking? Just prefer the top end of your Hi-Vi? If so, that pretty much confirms a crossover defect.

                I have a set of musical tracks I use to do listening tests. Music I know very well. I found Joni Mitchel can make defects stand out. A Julian Bream guitar can sound metalic if wrong. One of my Harry James pieces will sound more aggressive, kind of like a Buddy Rich band sound. Some Amanda McBroom will just plain melt a man if it is right. Excitement? Well how about Night on Bald Mountian. My wife is super sensitive to tweeter breakup as they cause IM down in the audible range. Violins and horns. The better the balance, the better it produces everything.

                Comment


                • #14
                  Originally posted by Steve Lee View Post
                  Just an analytical observation here:

                  At this point if the OP has a multiband EQUALIZER - would now shaping the response of the speakers to match his desired mid FR not indicate where the XO/speakers are falling short of his desires such that he could take the boost/cut settings as a guide to further refine the XO?
                  A valid method if close, but measurement can get you further faster. EQ won't fix phase interference.

                  Comment


                  • Steve Lee
                    Steve Lee commented
                    Editing a comment
                    EQ'ing AND then MEASURING would tell him what he needs to know where measuring ALONE gives him nothing definitive to work with - right?

                • #15
                  Originally posted by JamesTan View Post
                  Hi tvrgeek, I agree with most of your comments and observation but somehow I tend to disagree on the part where you say speaker should not have a sound of it's own. Different material itself already produce quite different sound especially when used in the tweeter. Also a compress driver tweeter will sound pretty different when compared to a soft dome tweeter.

                  As you mentioned, I am a beginner going in blind with no measuring equipment. All my ZMA and FRD files are obtained from manufacturer spec sheet. I used Xsim as the simulator when designing my XO. As a beginner, many of the terms you mentioned sound very new to me and I fully understand many factors need to be taken care to build a good speaker. This DIY project started out as a challenge and also a bit of fun to see how far I can achieve with the limited resources I have. So far I am pretty happy with the result.

                  I also like to have a discussion about comment saying the lack of warm vocal and not forward sounding is due to a XO design fault. Just to be clear I am not defending my XO design but a serious discussion on this topic. I have been into HiFi for the past 30 years and I also used to be a member of a Chinese Orchestra group playing flute. So when I say my current setup sounds realistic, clear and detail, I know what I am saying. Now the question is, if the current XO design is giving me all the accurate and detail sound regardless in the low, mid or high region for instrumental tracks, than how is it possible the XO design is flaw just because the vocal is not warm enough and not forward sounding enough? Please don't take it the wrong way, the vocal from my current setup is not unbearable or lousy in anyway but rather not to my taste. I seriously doubt by changing the XO design, one can made a speaker sound warmer or more transparent. Of course, this is open for discussion. Cheers. Btw, attached is the FR from Xsim simulation for my TB and HiVi combo.
                  Click image for larger version Name:	2wayTBHiViFR.JPG Views:	0 Size:	108.5 KB ID:	1451562
                  I think a sim is only a rough starting point. Did you sim the effect of the baffle?

                  Comment

                  Working...
                  X