Announcement

Collapse

Midwest Audio Fest

It’s that time audio enthusiasts! Registration for the 2019 Speaker Design Competition is now open! Visit midwestaudiofest.com for details and to list your speaker project. We are excited to see all returning participants, and look forward to meeting some new designers this year, as well! Be sure your plans include a visit to the Parts Express Tent Sale for the lowest prices of the year, and the Audio Swap Meet where you can buy and trade with other audio fans. We hope to see you this summer! Vivian and Jill
See more
See less

New Bookshelf MT for 2019 - Peerless meets TB

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

  • New Bookshelf MT for 2019 - Peerless meets TB

    I've been tossing around some ideas for a new 2-way bookshelf speaker and wanted to get some feedback from the forum. In all honesty I definitely don't need a new pair of speakers, but it's the start of a new year and I'm itching for a project to work on. I'd like to do something a little different, or in other words I'm not trying to build a tried-and-true design, I know there's plenty out there, I'm more interested in the challenge of a new design and at least for me, picking some new drivers I haven't had a chance to play with. So I'll just walk through my initial design/thought process and let me know what you guys think!

    First off, driver selection, we all have our reasons for picking driver A or tweeter B and our reasons may or may not be grounded in any kind of reality. Maybe it's price, maybe it's value, maybe it's cosmetic, perhaps you're a purist and it's completely based on performance, or you simply go to PE's website and sort by "most reviews" and pick the driver that has the best track record. I don't think there is a right or wrong way to go about driver selection, everybody has a reason. For me this design started with, I'll admit it, cosmetics. The visual. Looks. Probably the least important actual thing that matters for sound reproduction, but I really want these speakers to just look awesome. It's completely subjective, I know, so I started my search with the tweeter and initially I was drawn to the new Peerless Corundum dome tweeters. Ever since they came out and I took one look at that wide faceplate and metalic silver dome and grill, I knew I wanted to use them in a project. I jut thought they looked super slick! And the little grill probably doesn't mean much to most people but I have kids that love to poke in tweeter domes so I always love finding a good tweeter with a grill. (I know I can build speaker grills too, but that kind of defeats the purpose of picking a good looking driver in the first place!) So the Peerless Corundum dome tweeter it is, based solely on its looks, plus just the fact that it's new and hasn't seen its way into a lot of designs. So I thought it would be fun to work with.

    So the second criteria for this design after cosmetics would be, surprise, performance. Even if it's just datasheet stuff, even the best-looking tweeter I would pass on if it didn't have something real to back it up. The manufactured specs reveal some great characteristics, low fs, flat/smooth response, decent power handling, nominal sensitivity, should pair nicely with many mids, and an easy 8 ohm impedance. Looks good on paper, though there is an unusually high self resonance at like ~28 kHz. I haven't seen that on many published specs. Metal dome tweeters tend to have a fairly high-frequency resonance, sometimes in-band (<20 kHz) but they don't normally peak that high, but also tend to give metal domes their signature "sparkle" which can be for good or for bad depending on your preference. This peak has 15 dB gain but is well above the audible range. Does this concern anyone? I can't imagine why this would even matter. I'm chalking it up to a "don't care" at the moment and let's move onto the woofer selection. Oh and of the two diameter versions available, I'm leaning towards the smaller 25mm dome (DA25TX00-08) at the moment. Fs is only marginally higher and it's a few bucks cheaper.

    So the woofer selection was harder, a lot harder. Initially one would need to decide on the driver size. Did I want a little bookshelf speaker so like a 4" or 5" driver? Or was I looking for a full-size bookshelf speaker with the potential to have a little more low-end and go with a 6"-7" driver? Or should I just go all out and do an 8" driver? Oooh, an 8" with the 32mm Corundum looks tempting, probably one of the only tweeters that could cross low enough to keep cone break-up modes of a huge 8" woofer in check, but no, not for this build. Crossover complexity might be a bit much and it's hard to fit an 8" driver into a "bookshelf" sized speaker and have it model properly. So I settled on a 6-1/2" driver size and started with selecting every 6-1/2" driver PE sells (which is 64, more than any other driver they carry) and sorting by price - high to low. Holy cow, Morel make some expensive drivers, pass. While I didn't mention budget, obviously every speaker designer has to consider budget as a factor and budget always matters. For this build I was leaning towards the more expensive side of things opting to forgo the less expensive drives such as the ever-so-popular 6-1/2" poly cone woofer that costs all of $6. I'm sure it's a great driver and definitely has its place but ultimately didn't quite meet my criteria - which was once again, give me just a really great-looking driver!

    I considered several of the Peerless drivers, with their many different fancy cone materials and thought a Peerless/Peerless combo is always nice. Not a strict requirement though to keep woofer/tweeter of the same manufacturer. I know I tend to want to do that though, as do many others. For now I wanted to leave the option open for any manufacturer just to widen the playing field. Dayton RS series was considered but I swear almost every speaker in my house is Dayton Audio and while I do love them, I just wanted something different this time. And that's when I saw it, the Tang Band W6-1721. Wow that looks like such a beefy driver and I'm not just talking about the magnet. The wide surround and small dust cap with a traditional paper cone just look great. Does it have the specs to back up the looks? I think it does! Low fs for a 6 incher, some serious xmax at 8 mm, not typical for a driver this size, and where does this thing role off? TB clearly exaggerates the response as there is very little peaking (where are the cone breakup modes) and the response barely starts to roll off just above 10 kHz. Either way, I'll be measuring the drivers anyway so I don't need to get too hung up on published FR specs. But that's when it hit me, the motor is an underhung design, just like the tweeter! Wow, cool. But aren't most tweeters not already underhung? Peerless at least specs them that way so now you can say that both woofer and tweeter incorporate an underhung motor design and then whatever benefit is reaped (lower distortion, higher power, etc.) at least can be said about the entire speaker as a whole. So hey, that's kinda cool, right?

    So with the TB woofer picked I ran some sims in Unibox and came up with a 25 liter enclosure tuned to 36 Hz provides a small bump with an f3 of 30 Hz. It's a slightly compromised design, the driver really wants a full 28-29 liters (basically 1 cu.ft.) but in order to still meet my original design criteria that this be a bookshelf-sized speaker, it's already getting to be on the larger side of things. This requires an enclosure size of 15x10x16 (HWD) with a double-front baffle made from 3/4" MDF and reasonably braced. Also the port, oh my gosh, the port. I hate ports. Balancing port air speed with port length (ie, consumed internal volume) just drives me nuts. Who came up with the 5% minimum recommended port air speed anyway? What if I like shuffing? I must like shuffing, because I don't think I've build a ported box yet that actually meets the maximum port air speed of 17 m/s. The "ideal" port diameter for this driver in this enclosure is 3.25". Tuning a 25L enclosure to a low 36 Hz requires a whopping 17" long port. I could barely fold a 3.25" port to fit inside this tiny cabinet to get it to be that long let alone the fact that it would consume ~2.5L of internal volume to do so. So I compromised and went with a 2.5" port (which is a standard size) which ends up needing to be only about 10" long which can easily fit into the cabinet without elbow/bends and takes up only 0.8L. Of course the crux of it all is port air speed increases to 31 m/s at fb, so yeah I guess I like chuffing. For that all I have to say is, the port location is definitely going in the back! And by the way I did model about 8 other drivers that met my initial "looks good" criteria but then couldn't muster the "looks good modeled" criteria. Like what's with the Aurum Cantus 6" driver? You can't make the enclosure small enough to make that driver happy. Not to mention xmax is exceeded on many drivers with anything more than 10 watts input power (Dynavox drivers were the worst offenders in that category).

    Anyway, this post is way too long for a normal person to read, so I'll end with some drawings and definitely looking for some feedback or thoughts on my driver selection process. How do you pick drivers for your projects? What motivates you do pick a particular driver? What would anyone like to see as far as going deeper into this design? I have plenty of measurement gear and plan to design the crossover based entirely on measurements taken with the drivers in the cabinets, so the crossover design will have to come later. But initial box modeling and enclosure design is at least taking shape. Thanks for reading!

  • #2
    Okay- first off, this is the Gallery, meant for final photos of a build. If you have a query, please post it in the TechTalk sector.

    I actually have a plan to build with drivers on hand for the W8-2096 and the DA32TX, just need to get the time and finish the previous project or 2 first.

    TB plots are wider vertical dB scale than normally seen, which makes them look smoother in response. However, I have traced and adapted plots before for simulation and came out pretty close to actual using TB spec plots.

    As to port chuffing, and the model exceeding the level of 17m/s and being reasonable in reality upon completion; how are you doing your models? Are you modeling for power input that the drivers are rated? I see your output level is about 104dB in your Unibox simulation. Let's be reasonable here, that does not look like a level that is unattainable. However, my model does not match yours. I entered the specs directly from the pdf for the TB W6-1721, with the exception of the Le, as it as obviously a decimal point misplacement.

    As I put in the data, and no, I did not include the supposed coil DCR from the xover:
    Click image for larger version

Name:	ParameterInput.jpg
Views:	1
Size:	167.1 KB
ID:	1399056

    Here is what I get for the box you show in the first graphic:
    Click image for larger version  Name:	BoxAs Suggested.jpg Views:	1 Size:	180.4 KB ID:	1399053
    Click image for larger version  Name:	XmaxAsSuggested.jpg Views:	1 Size:	122.4 KB ID:	1399054
    Click image for larger version  Name:	MachAsSuggested.jpg Views:	1 Size:	119.2 KB ID:	1399055

    This is clearly not what you have posted as modeled. Next post, I'll post my simulation and thoughts...
    Wolf
    "Wolf, you shall now be known as "King of the Zip ties." -Pete00t
    "Wolf and speakers equivalent to Picasso and 'Blue'" -dantheman
    "He is a true ambassador for this forum and speaker DIY in general." -Ed Froste
    "We're all in this together, so keep your stick on the ice!" - Red Green aka Steve Smith

    *InDIYana event website*

    Photobucket pages:
    http://photobucket.com/Wolf-Speakers_and_more

    My blog/writeups/thoughts here at PE:
    http://techtalk.parts-express.com/blog.php?u=4102

    Comment


    • #3
      Now, with knowing a bit about how the underhung TB midbasses behave, I did a simulation as well. Rather than tuning to a number for maximally flat response as you did, I tuned to the Fs of the driver. These magnet monsters have a very well damped suspension, as well as Fs magnitude. Being an underhung motor, they do not unload gracefully when Xmax is breached. To keep this from happening, tuning to Fs is a good idea. It damps the driver the best it can be across the board. Also of note- I lowered the power input to reflect that of about 101-102dB. Your 104dB will approximately be twice the power input. Most rooms and people really don't want to exceed 100dB as a worst-case scenario, and even in larger rooms, this level being acceptable keeps the moderate level at longer distance in the safety realm of the driver. I really see no reason to model at higher than 102dB in a typical listening/living room except in the case of subwoofers. Subwoofers really need a higher threshold of safety margin, and I shoot for the best I can get there.

      I also chose a 2.5" diameter port, but I did go up to a 28.3ltr box, which is a full cubic foot. I also modeled for max-mach and coinciding Xmax breach. The max power the driver can take before exceeding 17m/s is 35W, and at the point, the Xmax breaches spec + 15% (linear overdrive) at 32.8Hz. This is a peak output of 103dB before calamity starts to ensue. It should be pointed out that the Xmax breach is handled differently by different drivers, and the unloading may not be as severe as pictured. Due to the low Qms of this driver, it may not care one bit.

      So- here is my model....

      Click image for larger version

Name:	MyModel25WTunedToFs.jpg
Views:	1
Size:	179.9 KB
ID:	1399058

      Click image for larger version

Name:	MyModelXmax.jpg
Views:	1
Size:	106.8 KB
ID:	1399059

      Click image for larger version

Name:	MyModelMach.jpg
Views:	1
Size:	121.8 KB
ID:	1399060

      I hope you find this useful,
      Wolf
      "Wolf, you shall now be known as "King of the Zip ties." -Pete00t
      "Wolf and speakers equivalent to Picasso and 'Blue'" -dantheman
      "He is a true ambassador for this forum and speaker DIY in general." -Ed Froste
      "We're all in this together, so keep your stick on the ice!" - Red Green aka Steve Smith

      *InDIYana event website*

      Photobucket pages:
      http://photobucket.com/Wolf-Speakers_and_more

      My blog/writeups/thoughts here at PE:
      http://techtalk.parts-express.com/blog.php?u=4102

      Comment


      • #4
        Hey Wolf, thanks for the great reply! Sorry I thought this forum was appropriate as once I start building the speakers I figured I would just keep this thread going with build pics. But if a moderator wants to move it to Tech Talk I don't mind. So you are correct, the T/S params from the datasheet do not reflect what I posted since I used the T/S params from Zaph's website (http://www.zaphaudio.com/6.5test/compare.html). Well I modeled both, but Zaph's numbers gave me a better response with the smaller cabinet, as you noted, so obviously I chose to use those values instead . The biggest difference between the two data sets is the Vas - which he states is about half what the datasheet reports with a higher fs as well. Perhaps he didn't allow the driver to break in properly before running the test? His numbers model better in a smaller enclosure because of this. I suppose the best thing to do is measure the actual drivers and not rely on datasheet values or even measured values from other sources. It was just something to start with to get an idea of possible performance in an enclosure. I'll post a pic of the parameters from Zaph and if you want you can run those and see if it changes your results.

        I usually model the power for peak xmax up to the max spec'd power. As you can see in this case, they line right up at 50W input power (at or above fb, excluding driver unloading below fb). And with the other T/S params, peak output power is calculated at a more reasonable 101.7 dB. You make a very interesting point about breaching xmax with underhung coils and therefore shooting to design for tuning, fb = fs. I have no problems adjusting the design to account for this. Using Zaphs numbers though a 44 Hz tune peaks quite a bit. Either way, a larger cabinet does model better, I did notice that. But this box is getting big! Looks like if I add a 1/2" to the width and 1" to the height I can buy myself another 3.5L. Every little bit helps but consider when I first sketched out what I thought was a "reasonable size" speaker box just based on practicality, the total enclosure volume was only 18L, so I grew the box quite a bit just to get to 25L. But at this point I suppose another inch isn't going to matter. These are going to be huge, relatively speaking, for a bookshelf speaker.

        Anyway thanks for your thoughts, always appreciated!

        Comment


        • #5
          Fair enough, I did not recall Zaph measuring this one. Either way, I don't always get what he does(did) when I measure T/S either. That also said- the W8 measured quite different than the spec sheet.

          Later,
          Wolf
          "Wolf, you shall now be known as "King of the Zip ties." -Pete00t
          "Wolf and speakers equivalent to Picasso and 'Blue'" -dantheman
          "He is a true ambassador for this forum and speaker DIY in general." -Ed Froste
          "We're all in this together, so keep your stick on the ice!" - Red Green aka Steve Smith

          *InDIYana event website*

          Photobucket pages:
          http://photobucket.com/Wolf-Speakers_and_more

          My blog/writeups/thoughts here at PE:
          http://techtalk.parts-express.com/blog.php?u=4102

          Comment


          • #6
            why the dual binding posts?

            Comment


            • #7
              Leaves the option to bi-amp or run an active crossover down the road. Or you can temporarily use an active crossover during the "tuning stages" of the crossover before committing to a passive one. I've never actually done it that way, but seems like it would be practical. At least in terms of adjusting the amount of baffle step compensation, or picking crossover points and slopes. Other than that, there is no real reason. It looks cool? The only update so far is that I've been toying around with the Peerless 830883 6-1/2" Nomex Cone HDS driver instead of the TB driver. It models really well in a smaller cabinet, with just a small hit to f3 but it's half the cost. Plus it's a good looking driver as well. And it would be a nice Peerless matched combo set.

              Click image for larger version  Name:	peerless woofer.PNG Views:	2 Size:	150.1 KB ID:	1400336Click image for larger version  Name:	VB Response Zaph Peerless by Tymphany 830883.gif Views:	2 Size:	32.8 KB ID:	1400337 Click image for larger version

Name:	peerlessside.PNG
Views:	1
Size:	16.8 KB
ID:	1400537

              *edited to add pics
              Last edited by danmarx; 01-15-2019, 10:17 PM.

              Comment

              Working...
              X