Simple DIYable Waveguide testing (dome tweeters)

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • dlneubec
    Super Senior Member
    • Jan 2006
    • 1454

    Simple DIYable Waveguide testing (dome tweeters)

    Hi All,

    I have been doing some testing with shallow waveguides for my BaSSlines project that I thought you all might be interested in. I posted this in my BaSSlines thread, but realized many of you who might be interested could miss it there. I created an insert into the baffle that would allow me to make multiple WG's and insert them into the same baffle (see photo below), with same tweeter, same mic setup, position, etc. The tweeter is the Peerless HDS with the face plate removed. The tweeter is mounted by pressing it to the back of he WG cutout, held by a woof fixture. The measurements were all taken at 1 meter, with the mic 80" high, in the center of a room with vaulted ceilings probably 15' high at that point and with a 4ms gated window.

    All the WG's were created from mdf by boring a hole with a forstner bit, then rounding over the appropraite amount and then routing from the back for the palnned depth of either 1/2" deep or 3/4" deep in these cases. These are something just about any DIYer can build petty easily, which was the idea. There are a few tricks to doing this, most notably, when using a 3/4" rounvoer bit, you have to have a backer piece to you mdf baffle that you also bore a hole through. This backer piece is needed for the bearing of the deeper 3/4" roundover bit to ride on. All the bore through holes on these were adjusted so the final throat sizes are about 1-1/4" +- 1/32" at the final size.

    Most of the plots show the waveguide perfomance at 0-15-30-45 and 60º of axis horizontally, so you can see the effect on axis and the directivity off axis. The first plot is the HDS with a standard flush mount. My goal with the waveguide is to get some boost at the low end so that when qualized out in the crossover you get the possible benefits of lowered distortion, slighly higher sensitivity, better off axis directivity and a counter to the typical rising responce of the HDS tweeter on axis, which might be fatiguing to some over time. Also included are some comparisons of the favorite WG results on axis with each other.

    I'll include some photos of the waveguide testing insert setup in this post and in the next one I'll post the results.

    Click image for larger version

Name:	DSCF0169.webp
Views:	124
Size:	19.1 KB
ID:	941892

    Click image for larger version

Name:	DSCF0171.webp
Views:	122
Size:	22.2 KB
ID:	941893

    Click image for larger version

Name:	DSCF0175.webp
Views:	121
Size:	25.7 KB
ID:	941894

    Click image for larger version

Name:	DSCF0174.webp
Views:	119
Size:	28.0 KB
ID:	941895

    Click image for larger version

Name:	DSCF0144b.webp
Views:	120
Size:	27.6 KB
ID:	941896
    Last edited by theSven; 23 June 2023, 22:51 Friday. Reason: Update image location
    Dan N.
  • dlneubec
    Super Senior Member
    • Jan 2006
    • 1454

    #2
    Here are some results:

    HDS mounted Flush: 0-15-30-45-60º

    Click image for larger version

Name:	hds_flushMt_1m_4ms-0-15-30-45-60.gif
Views:	115
Size:	22.5 KB
ID:	941897

    1/2" deep WG, 1/2" roundover: 0-15-30-45-60º

    Click image for larger version

Name:	hds_halfInch-halfRO-WG_1m_4ms-0-15-.gif
Views:	120
Size:	22.0 KB
ID:	941898

    1/2" deep WG, 3/4" roundover: 0-15-30-45-60º

    Click image for larger version

Name:	hds_halfInchWG_1m_4ms-0-15-30-45-60.gif
Views:	123
Size:	21.7 KB
ID:	941899

    3/4" deep WG, 3/4" roundover: 0-15-30-45-60º

    Click image for larger version

Name:	hds_three-fourthsWG_1m_4ms-0-15-30-.gif
Views:	121
Size:	21.6 KB
ID:	941900

    3/4" deep WG, 3/4" roundover, with 1/2" deep 45º chamfer after, sharp edges sanded round by hand: 0-15-30-45-60º

    Click image for larger version

Name:	hds_three-fourths-45dWG_1m_4ms-0-15.gif
Views:	124
Size:	22.3 KB
ID:	941901

    3/4" deep WG, 3/4" deep 45º chamfer, sharp edges sanded round by hand: 0-15-30-45-60º

    Click image for larger version

Name:	hds_three-fourths-45dWG-only_1m_4ms.gif
Views:	118
Size:	22.8 KB
ID:	941902


    My conclusion is: it depends on what you are looking for. The deeper wavegides have more boost, but all peak at about 4kHz on axis. The 1/2" deep waveguide, with 1/2" roundover has the flattest FR on axis from 4kHz and up on axis. The 1/2" deep/3/4" roundover WG and the 3/4" deep/3/4" roundover WG have the best off axis responce when you get out to 30º and beyond and the most directivity. I'm now leaning toward either the 1/2" deep/3/4" roundover of the 1/2" deep/1/2" roundover WG's from a raw response point of view. I will see how they look in the crossover design.

    I hope these results will encourage some of you to experiment with this concept. I think it has a lot of merit and is not at all hard to do with just a router and standard bits. It helps if you have a drill press and a router table as well.
    Last edited by theSven; 23 June 2023, 22:53 Friday. Reason: Update image location
    Dan N.

    Comment

    • fjhuerta
      Super Senior Member
      • Jun 2006
      • 1140

      #3
      That looks quite promising... did you do any off-axis measurements? Does this also help flatten off-axis response?
      Javier Huerta

      Comment

      • dlneubec
        Super Senior Member
        • Jan 2006
        • 1454

        #4
        fjhuerta,

        The first series of graphs for each different waveguide configurtion and include horizontal off axis plots at 0-15-30-45 and 60º off axis. You can see which is which in the titles above the photos. Only the last graph are all on axis and that is for several of the different waveguide configurations I tried, so that it is easier to see how they compare to one another on axis.
        Dan N.

        Comment

        • fjhuerta
          Super Senior Member
          • Jun 2006
          • 1140

          #5
          Sorry Dan, you are right!
          Javier Huerta

          Comment

          • augerpro
            Super Senior Member
            • Aug 2006
            • 1866

            #6
            Awesome work Dan, thanks!

            Now that my eyes are totally crossed from staring at these graphs I think I'm mostly liking the 3/4" waveguides. They are not quite as smooth as the 1/2" but they are still very good where it counts and can be eq'ed just as flat I'm guessing. Given that the main thing I like about them though is the considerable gain around 2-3khz. Although it isn't obvious from the off axis responses, that gain has to come as a loss off axis. Power output is the same, it's just "aimed" a little differently. This IMO is a real plus with this particular tweeter. From my experience using this tweeter I'm betting this will be audibly better even if it isn't obvious from the plots. Also being able eg down that 3dB gain will help. Just my .02, what is pulling you towards the 1/2" option?
            ~Brandon 8O
            Please donate to my Waveguides for CNC and 3D Printing Project!!
            Please donate to my Monster Box Construction Methods Project!!
            DriverVault
            Soma Sonus

            Comment

            • dlneubec
              Super Senior Member
              • Jan 2006
              • 1454

              #7
              One reason is that I'm working on some preliminary crossovers and I'm findiing it harder to get a good phase match with the 6md38 when using the 3/4" depths, once I eq out the additional lower end boost. It is early, however, and I will try to create plausible versions with both depths so I can compare them not with my ears as well.
              Dan N.

              Comment

              • JonW
                Super Senior Member
                • Jan 2006
                • 1582

                #8
                Hi Dan,

                Cool experiments. Thanks for posting. :T A few things I was wondering about…

                -The 1/2” deep WG, 1/2” roundover looks excellent to me. Why not just go with that? Is it to try and get more low end extension like, say, with the 3/4" deep WG, 3/4" roundover case? Although not as flat as at 1/2”, the peak (if you can call it a peak) will come down with the crossover, anyways.

                -Do you think these results are specific to the HDS tweeter or might they be more generalizable to others? My guess is they’re specific.

                -How do these results to compare to the regular tweeter without any G treatment? I thought the HDS looked OK on it's own. Not great low end extension, but generally OK. Zaph's measurements show a dip around 7k which you seem to be avoiding.

                Comment

                • dlneubec
                  Super Senior Member
                  • Jan 2006
                  • 1454

                  #9
                  Originally posted by JonW
                  Hi Dan,

                  Cool experiments. Thanks for posting. :T A few things I was wondering about…

                  -The 1/2” deep WG, 1/2” roundover looks excellent to me. Why not just go with that? Is it to try and get more low end extension like, say, with the 3/4" deep WG, 3/4" roundover case? Although not as flat as at 1/2”, the peak (if you can call it a peak) will come down with the crossover, anyways.

                  -Do you think these results are specific to the HDS tweeter or might they be more generalizable to others? My guess is they’re specific.

                  -How do these results to compare to the regular tweeter without any G treatment? I thought the HDS looked OK on it's own. Not great low end extension, but generally OK. Zaph's measurements show a dip around 7k which you seem to be avoiding.
                  There are certainly benefits of the 1/2", which is smoother, flatter response but there is more peaking in the off axis. However, the 3/4" with 3/4" roundover has more boost and hence the ability to play louder at the low end and with less distortion or strain and better off axis directivity. Both also offer different physical offsets which can make getting a good phase alignment easier with a given combination of drivers.

                  At some point, I'll pull together some other tweeters to test with, but my guess is that the results will be pretty similar, as long as the dome is the same size and general shape. For example, a 3/4" dome would likely need a smaller throat size to replicate these results, but I bet would be close with a reduced throat. A ring radiater might look quite different due to its different geometry. Unfortunately, I don't have any on hand to test, though I've got some other domes, the same and smaller sizes.

                  The results of the standard flush mount HDS tweeter is the first graph in the post above and its on axis is compared to the WG's on axis in the last graph in the post. May people complain and/or have had trouble with the rising response of the HDS. One of the reasons for exploring the WG options was to physically counteract that, rather than doing so electrically, plus the benefits of a potentially stronger, lower distortion low end performance and increased off axis directivity.
                  Dan N.

                  Comment

                  • JonW
                    Super Senior Member
                    • Jan 2006
                    • 1582

                    #10
                    Thanks Dan. OK, I see that the HDS rising response could be problematic without a WG. I’d just pick another tweeter without the problem.

                    Right, I forgot about the acoustic offset benefit.

                    And I see the standard in the first graph now. (d’oh!).

                    Makes sense about the boost and more low end.

                    I can see that it’s a lot of work to find the right combination for a given tweeter. Kind of like when I made a bunch of different test cabinets to see which one measured best, on my recent Scanspeak MT design. It’s kind of worth it, though, in the long run.

                    Comment

                    • dlneubec
                      Super Senior Member
                      • Jan 2006
                      • 1454

                      #11
                      Originally posted by JonW
                      Thanks Dan. OK, I see that the HDS rising response could be problematic without a WG. I’d just pick another tweeter without the problem.
                      Jon,

                      When it comes to dome tweeters, with very low distortion, very strong low end performance, sensitivity at 93db or above and a reasonable price (let's say under $100), there is not a ton of options to choose from. :B The HDS seems to be very well thought of, despite the rising response.

                      Also, don't forget the potential benefits from increased off axis directivity, which even the small waveguide options provide from about 4kHz up. I have a feeling the combination of the waveguide and HDS will be a delicious one in the end.
                      Dan N.

                      Comment

                      • CraigJ
                        Senior Member
                        • Feb 2006
                        • 518

                        #12
                        Hi Dan,

                        I've got an extra pair of the MCM waveguides pre-cut to fit your HDS tweeter if you'd like to try them as well. I'll ship. I didn't have testing equipment then, but they sounded good on a Isiris prototype.

                        Craig

                        Comment

                        • dlneubec
                          Super Senior Member
                          • Jan 2006
                          • 1454

                          #13
                          Craig,

                          Thanks for the offer. I do have at least one or two of the MCM's around here myself, so no need to send them. One day I'll pull them out and give them a test for comparison.

                          What I'd also like to test, eventually, are these small waveguides with some ribbons or planars. Since they are typically low end challenged, it could be worth exploring.
                          Dan N.

                          Comment

                          • Saurav
                            Super Senior Member
                            • Dec 2004
                            • 1166

                            #14
                            Is there any way you could create an elliptical waveguide? That may remove some of the on-axis dip that you're seeing in all your measurements (between 2k and 3k, almost all the waveguides have higher output off-axis than on-axis). *If* it is what I'm thinking it is... some kind of diffraction effect that's maximized on-axis because all points on the WG are equidistant from the dome as well as the mic.

                            Comment

                            • augerpro
                              Super Senior Member
                              • Aug 2006
                              • 1866

                              #15
                              Saurav I thought the same thing, but you see the dip in the flush mounted plots as well-it's just baffle diffraction.
                              ~Brandon 8O
                              Please donate to my Waveguides for CNC and 3D Printing Project!!
                              Please donate to my Monster Box Construction Methods Project!!
                              DriverVault
                              Soma Sonus

                              Comment

                              • Dennis H
                                Ultra Senior Member
                                • Aug 2002
                                • 3791

                                #16
                                What I'd also like to test, eventually, are these small waveguides with some ribbons or planars. Since they are typically low end challenged, it could be worth exploring.
                                The B&G Neo3 mounting plates have a bit of a waveguide shape to them.

                                Click image for larger version

Name:	BG-Neo3-Faceplate-Flushmount.jpg?strip=all&lossy=1&ssl=1.jpg
Views:	101
Size:	28.2 KB
ID:	941916
                                Last edited by theSven; 23 June 2023, 23:11 Friday. Reason: Update image location

                                Comment

                                • Saurav
                                  Super Senior Member
                                  • Dec 2004
                                  • 1166

                                  #17
                                  Originally posted by augerpro
                                  Saurav I thought the same thing, but you see the dip in the flush mounted plots as well-it's just baffle diffraction.
                                  Heh. I did hesitate and think about that when I was posting. I'd have thought that the WG would reduce the extent to which the baffle edges were 'lit', but maybe not so much with a shallow WG.

                                  Comment

                                  • JonW
                                    Super Senior Member
                                    • Jan 2006
                                    • 1582

                                    #18
                                    Originally posted by dlneubec
                                    Jon,

                                    When it comes to dome tweeters, with very low distortion, very strong low end performance, sensitivity at 93db or above and a reasonable price (let's say under $100), there is not a ton of options to choose from. :B The HDS seems to be very well thought of, despite the rising response.

                                    Also, don't forget the potential benefits from increased off axis directivity, which even the small waveguide options provide from about 4kHz up. I have a feeling the combination of the waveguide and HDS will be a delicious one in the end.
                                    OK, I see. Low distortion, low end extension, sensitive, and cheap is a pretty tough list to find in one tweeter. For the Scanspeak 6600 I am using, it's low distortion and plays low frequencies well. Sensitivity is ~91 db which is low for your use. And it's certainly not cheap. But the reason I mention it is that someone here said that it has really good off-axis capabilities. I'm not sure but it seems to play well off-axis to my ears and my mic. I could dig out some 0 and 30 degree measurements I took, if anyone wants to see them.

                                    Comment

                                    • norcad
                                      Member
                                      • Jun 2008
                                      • 84

                                      #19
                                      What about Seas new H1499-06 27TBCD/GB-DXT?

                                      Comment

                                      • cotdt
                                        Senior Member
                                        • Oct 2005
                                        • 393

                                        #20
                                        The stock Peerless HDS tweeter faceplates already have a shallow waveguide. I'm thinking a larger but still shallow waveguide is needed in order to cross this tweeter really low.

                                        Comment

                                        • dlneubec
                                          Super Senior Member
                                          • Jan 2006
                                          • 1454

                                          #21
                                          cotdt,
                                          I would agree with that, however crossing lower was not my goal in this particular case. Improving the low end performance by equalising out the boost provided, countering the rise in the natural reponse of the HDS tweeter and increasing off axis directivity were my goals. The other goal was that I wanted this to be somehing the typical DIYer could do with tools they already own at little or no cost.
                                          Dan N.

                                          Comment

                                          • cotdt
                                            Senior Member
                                            • Oct 2005
                                            • 393

                                            #22
                                            Originally posted by dlneubec
                                            cotdt,
                                            I would agree with that, however crossing lower was not my goal in this particular case. Improving the low end performance by equalising out the boost provided, countering the rise in the natural reponse of the HDS tweeter and increasing off axis directivity were my goals.
                                            Why??? Low end performance is already good, I had mine crossed at 1.8kHz LR2 (acoustical) and it did not strain at all even at high volumes. This tweeter pretty much "crosses itself over" and might not need a high-pass at all, but I don't know I've never tried. With a larger waveguide this tweeter can probably do 1kHz.

                                            What is wrong with the natural rising response? It adds more detail and airiness, and becomes flat a little off axis. That said, I will note that this tweeter is naturally clean and airy sounding even when the crossover flattens the response.

                                            I'm glad to see that yours doesn't have that 6.5kHz dip that Zaph's sample had. Mine doesn't have that dip either.

                                            Comment

                                            • dlneubec
                                              Super Senior Member
                                              • Jan 2006
                                              • 1454

                                              #23
                                              Well, the biggest single complaint I've seen about the Peerless is the rising response of the peerless HDS and the extra effort and components it takes to get it to flatten out. So, the waveguide mount offers the potential for a simpler crossover, with less parts and a more natural sound, yet a flat FR and good thermal and over-excursion protection. Many designers and listeners prefer a little roll off at the high end rather than a rise. Obviously you prefer the rise.

                                              The reason for the boost at the low end is simple. In addition to flattening out the rising response, it acoustically increases the output at the low end. Let's say that it increases by 5db. So now when you equalize that to flat in the crossover, the distortion levels are dropped by that amount as well, so you have a cleaner low end. Arguably, the tweeter can also play to a higher spl with less distortion, less thermal compression and less xmax limitation. I'm using this in a design that is intended to be high sensitivity, which is typically not the domain of dome tweeters and the ribbons that are typically used and can handle the low end duties are very expensive.

                                              Then there is the benefit of increased directivity off axis. IMO, there is little reason not to do it this way. YMMV.
                                              Dan N.

                                              Comment

                                              • JohnInKY
                                                Junior Member
                                                • Sep 2008
                                                • 13

                                                #24
                                                Those graphs look interesting, Dan. I wonder about your statement that:

                                                "So now when you equalize that to flat in the crossover, the distortion levels are dropped by that amount as well, so you have a cleaner low end"

                                                have you actually measured lower distortion? While amplitude induced distortion would be expected to decrease, I'm not so sure about non-linear artifacts, such as other dispersion modes induced by the waveguide itself as it re-directs the output, or from effects induced by the xover.

                                                John L.

                                                Comment

                                                • dlneubec
                                                  Super Senior Member
                                                  • Jan 2006
                                                  • 1454

                                                  #25
                                                  Hi John,

                                                  No I have not measured it to confirm and should have qualified that statement, to say that I believe it will drop them. I do think I recall reading Zaph making this statement in one of the forums in regards to his suggested us if a 1/2" waveguide mount for the Vifa D26. At some point I will do some distortion measurements to augment this study and will certainly post the results, whatever they are.

                                                  Hey, you will at least be able have a listen to my very prelimary prototype pair where this is implemented in a couple weeks in Lexington. Not that this will provide any answers to the question posed, however. See you soon.
                                                  Dan N.

                                                  Comment

                                                  • JohnInKY
                                                    Junior Member
                                                    • Sep 2008
                                                    • 13

                                                    #26
                                                    That'l be cool.. looking forward to it.

                                                    I've got 2 more of these ATC neoplanar 25"ers as rears in my 7.1 HT setup, with a routed 1/2" roundover on a simple rectangular 1" cherry baffle, so I hope what you say is true. I can bring one if you'd be interested in it. They claim to be good from ~ 150 Hz to >20Khz, with a gradually rising FR... <mfg. data


                                                    Images not available
                                                    Last edited by theSven; 23 June 2023, 23:06 Friday. Reason: Remove broken image links

                                                    Comment

                                                    • dlneubec
                                                      Super Senior Member
                                                      • Jan 2006
                                                      • 1454

                                                      #27
                                                      Sure, bring it, that is if you have room! It sounds like you are going to have a pretty full load as it is, with the mains, subs and all. I'm really looking forward to hearing those big mains.
                                                      Dan N.

                                                      Comment

                                                      • JonW
                                                        Super Senior Member
                                                        • Jan 2006
                                                        • 1582

                                                        #28
                                                        Hi Dan,

                                                        I’m thinking up a very small speaker project, for which I need a small tweeter. Now that you’ve been brave enough to take apart an HDS tweeter, without the faceplate, it might be small enough to use for my purposes. (Well, it’s still bigger than I want but it’s getting closer.)

                                                        How might one mount this tweeter, without the faceplate, on a baffle? I didn’t quite follow how you did everything from the photos. Unscrew the faceplate. Make a hole with waveguide in the baffle. Then shave the baffle down to, say, 1/2” thick (or keep it 3/4”). And have a bracket in the rear to push/hold the tweeter in place? Or was it something more secure, maybe using screws? It looks like you might have some doughnut or something in there as well but I’m not sure where that went.

                                                        Thanks. I think it’s a cool collection of experiments you’re doing.

                                                        Comment

                                                        • dlneubec
                                                          Super Senior Member
                                                          • Jan 2006
                                                          • 1454

                                                          #29
                                                          Originally posted by JonW
                                                          Hi Dan,

                                                          I’m thinking up a very small speaker project, for which I need a small tweeter. Now that you’ve been brave enough to take apart an HDS tweeter, without the faceplate, it might be small enough to use for my purposes. (Well, it’s still bigger than I want but it’s getting closer.)

                                                          How might one mount this tweeter, without the faceplate, on a baffle? I didn’t quite follow how you did everything from the photos. Unscrew the faceplate. Make a hole with waveguide in the baffle. Then shave the baffle down to, say, 1/2” thick (or keep it 3/4”). And have a bracket in the rear to push/hold the tweeter in place? Or was it something more secure, maybe using screws? It looks like you might have some doughnut or something in there as well but I’m not sure where that went.

                                                          Thanks. I think it’s a cool collection of experiments you’re doing.
                                                          Hi Jon,

                                                          It is not hard to do, but does take some specific steps and a little time. to make the hole and rear rebate for the rear mounted tweeter in a shalwo waveguide, here are the steps I used, lets say for a 1/2" deep WG in a 3/4" deep baffle, with a 1/2" roundover:

                                                          1) mark the center of the hole and drill for using a circle guide with a router. In my case that is the Jasper Jig.

                                                          2) from the rear side, route out a 1/4" deep hole in the baffle at the outer diameter of the tweeter body (face plate removed and room left for the terminals). Lets say that is a 3-1/4" diameter hole. You will only do one or two bit widths of the rebate at this time, since you will need the entire 3/4" depth when routing the roundover for the center hole for the WG. So you end up with a slot that is about 1/2" wide, 1/4" deep and 3-1/4" outside diameter. You have to do this before you bore the center hole, becuase you will lose the ability to use the circle jig, once you have bored the larger hole for the waveguide opening in the next step.

                                                          3) Drill a 1-1/4" hole or whatever size is just larger than the tweeter dome and it's small surround. Take care to make it as straight a bore as possible. I use a drill press and a forstner bit, but it could be done with a spade bit and standard drill, with care. (Note: if you are doing a 3/4" deep waveguide with a 3/4" roundover, you don't need to do step two, but you will need to temporarily install a backer board where the waveguide hole is bored through, since a 3/4" roundover bit is about 1-1/8" deep to the edge of the bit's bearing.)

                                                          4) Roundover the front side of the 1-1/4" hole with a 1/2" roundover bit. A router table is easiest, but it can also be done with a stand alone router.

                                                          5) Reinstall the router bit you use to cut holes and set it to a 1/4" depth. From the rear side, using the router by hand, remove the additional material inside the rebate you already started routing in step 2. You are now left with a 1-1/4" hole with a 1/2" roundover,1/2" depth and the it steps to a larger hole at the back side of the baffle that is 3-1/4" in diameter.

                                                          From this point on it is up to you how you mount the tweeter. I cut a U-shaped piece as shown in the photos and screwed it into the back of the baffle to hold the tweeter by pressure. The legs of the U have to be just a little less (maybe 1/8" or so)than the depth of th tweeter body sticking out. An easier way might be to glue some spacers just short of the depth the tweeter body sticking out and then use a flat piece of stock across the back of the tweeter, screwed into the spacers. A little gap between the spacers and the cross piece allows for the pressure mount.

                                                          Yes you do have to remove the tweeter faceplate, because it has a shallow waveguide like feature built into it and it would be quite hard to mate your hole with the tweeter faceplate wihout a big gap. Also, as you are suggesting, it makes it possible to get the tweeter closer to the adjacent driver, another plus to this concept.

                                                          I hope this is clear. It is hard to describe in words.
                                                          Dan N.

                                                          Comment

                                                          • JonW
                                                            Super Senior Member
                                                            • Jan 2006
                                                            • 1582

                                                            #30
                                                            Hi Dan,

                                                            Thanks- that’s wonderfully clear and explained perfectly. Not easy to do without pics.

                                                            It just so happens that I have a drill press and a set of Forstner bits. And a Jasper Jig and router. The only thing you mentioned that I don’t have is a router table.

                                                            So 3.25” is about the size that the tweeter takes up on the baffle. For a small speaker, that’s better than the typical ~4.25” of most tweeters. I’m still looking into smaller-flanged tweeter options. But the HDS can also be a contender. It’s got better-than-average distortion and, as you showed, the frequency response can be altered with mini waveguides. And quite importantly for me, you’ve done the hard work of taking one apart. So I won’t have to buy some tweeter and destroy it if it doesn’t come apart easily. The HDS looks like it can just have the faceplate screwed off.

                                                            OK, you’re holding the tweeter in place by pressure of that U bracket. You must have some very long screws in there. Do you find that, because you don’t have the tweeter screwed in place, it will move a bit, relative to the baffle hole?

                                                            I like the sound of the slightly different method wherein you glue in some spaces and put flat stock across the back of the tweeter. But I’m wondering if there’s a better way to hold in the tweeter, so that it doesn’t move even a mm from the dome being centered in the hole. My speaker will be a sealed box, not an open baffle. So I won’t be able to nudge the tweeter any, once I close up the box. Screws would be nice. But it looks like the faceplate screw holes would interfere with the waveguide shape? Press fitting doesn’t look likely because of the leads sticking out. Hmmm…

                                                            -Jon

                                                            Comment

                                                            • dlneubec
                                                              Super Senior Member
                                                              • Jan 2006
                                                              • 1454

                                                              #31
                                                              Jon,

                                                              This was the easiest and best method I could think ofor my particular open back speaker. I will probably to use the spacer method this time with the spacers made of hardwood and glued so that the tweeter has little oppotunity to shift (a tight fit of the tweeter body between the spacers). I may also use threaded inserts and cap head screws to hold it in place instead of screws.

                                                              I suppose one could installed spacers on 4 sides in order to press tighly against the tweeter body or perhaps a spacer ring(s) could be made with a tight body fit and notches cut out for the terminals, which when glued in place and the tweeter then press fit with a cross piece, would have little opportunity to move. That would probably work for you.

                                                              Due to my open baffle design, I want to keep as many reflective objects off the back of the baffle as possible and I have to be concerned about what it all looks like, since it will be visible.
                                                              Dan N.

                                                              Comment

                                                              • JonW
                                                                Super Senior Member
                                                                • Jan 2006
                                                                • 1582

                                                                #32
                                                                Got it. If the small tweeter options I'm looking at don't work out, and I turn to the HDS, I could get one in my hands and see what might make sense for how to mount it. Thanks again, Dan. :T

                                                                Comment

                                                                • cotdt
                                                                  Senior Member
                                                                  • Oct 2005
                                                                  • 393

                                                                  #33
                                                                  Dan, do you have some spare Excel W22's? You can rip out the cone and use those as the waveguide. They have the perfect shape for a shallow waveguide.

                                                                  Comment

                                                                  • Dennis H
                                                                    Ultra Senior Member
                                                                    • Aug 2002
                                                                    • 3791

                                                                    #34
                                                                    Originally posted by cotdt
                                                                    Dan, do you have some spare Excel W22's? You can rip out the cone and use those as the waveguide. They have the perfect shape for a shallow waveguide.
                                                                    No, no, rip out the phase plug and install a neo tweeter for a nice coax.

                                                                    Comment

                                                                    • JonW
                                                                      Super Senior Member
                                                                      • Jan 2006
                                                                      • 1582

                                                                      #35
                                                                      Easy does it there, fellas. He’s already taking apart moderately expensive tweeters for testing. And now you want him to rip apart a $200+ woofer just to satisfy your idle curiosity? :P

                                                                      But if you want to try using woofer cones for waveguides, I would think that there are many options out there. Madisound offers reconing services for the Scanspeak drivers, at least. So it might be possible to just buy the cone parts. That would make life pretty easy- just buy some cones from some woofers. Maybe even easier would be to get the whole driver and just take out the voice coil, magnet, etc. But keep the frame, rubber surround, and cone. That way you’d have an easy way to mount it to a baffle. Maybe metal cones would be a little stronger and a little easier to deal with.

                                                                      I didn’t see the Seas W22 reconing offered there, but it might be possible to get the part. Or look at other woofers of similar size. An RS180 or RS225, etc. could be an option. Not sure if the shape is perfect, but it's metal, pretty cheap, and there's already a perfect hole cut into it for the phase plug.

                                                                      Kind of a cool idea to think about…

                                                                      Comment

                                                                      • dlneubec
                                                                        Super Senior Member
                                                                        • Jan 2006
                                                                        • 1454

                                                                        #36
                                                                        Guys,

                                                                        I don't discount that using an old driver for a waveguide might work from a shape standpoint, but I would think it has to be a non resonant surface, which an woofer cone would not be, unless you used it as a form and filled the back in with Great Stuff, fiberglass resin, or something.

                                                                        FWIW, here is a graph that shows the extremes of my experiment. The bottom series is the HDS flush mounted, no waveguide. It shows FR at 0-15-30-45 and 60 degrees off axis. Pasted just above that, only 15db higher, is the 3/4" deep waveguide with a 3/4" roundover. The red lines you see on both are actually at the same spl level, 80db. You can see the boost the waveguide provides bout 5db at peak) and also the increased off axis directivity. If you want more directivity or boost or you want to lower the peak of the boost, you probably need to go deeper and wider with the waveguide mouth, which no longer becomes simple to construct.

                                                                        Click image for larger version

Name:	hds_flushMt_wg3q-0-15-30-45-60-adj.gif
Views:	100
Size:	28.3 KB
ID:	941915
                                                                        Last edited by theSven; 23 June 2023, 23:06 Friday. Reason: Update image location
                                                                        Dan N.

                                                                        Comment

                                                                        • JonW
                                                                          Super Senior Member
                                                                          • Jan 2006
                                                                          • 1582

                                                                          #37
                                                                          Hi Dan,

                                                                          Yeah, that makes sense, that you want a waveguide to not move around any. So forget the mention of using driver cones.

                                                                          So your experiments have made me curious and I’ve been reading a little bit about waveguides. Interesting topic. Some good links:

                                                                          Hi all (and specifically Jon) Would love to hear your thoughts about utilizing tweeter waveguides. I believe in low XO high order designs such Jon's, but I see precious little on DIY speaker forums about this subject. I believe the benifits are obvious - large increases in low end output that (when corrected for flat response)






                                                                          Have you thought about trying out the MCM “horn lens?”



                                                                          It’s cheap, easy to shave down to size, and you’d get more of a waveguide effect than your small ones. Your 3/4” deep, 3/4” roundover mini looks quite good. It’s just a suggestion if you want to get even more boost, etc.

                                                                          Just curious: In the final speaker, what will you do about finishing the waveguide. You can’t veneer it. Might you remake it out of a hardwood or something? You know, you’ve set the finishing bar pretty high for yourself.
                                                                          Last edited by theSven; 23 June 2023, 23:12 Friday. Reason: Update image location

                                                                          Comment

                                                                          • dlneubec
                                                                            Super Senior Member
                                                                            • Jan 2006
                                                                            • 1454

                                                                            #38
                                                                            Hi Jon,

                                                                            I'll take a look at those links, thanks.

                                                                            My waveguides will be done in a Lexan front baffle with a hardwood backing. See the illustration below (the BaSSlines thread).

                                                                            BTW, I have come up with a way to perhaps mount the HDS tweeter without removing the faceplate. It uses an O-ring. I have ordered afew at 1-5/16" ID and 1-3/8" ID, 1/8" diameter to experiment with. See drawing below. Shown in the plans are with a 1/2" roundover and 3/4" roundover. My guess is that the FR may not be as smooth at the top end as when the flange is removed, but it may be good enough not to matter. If this works, it will make it possible to use more tweeters without having to worry about removing the face plate, ruining a warranty, or mounting from the rear, having a tweeter slip or move, etc. I'll let you all know how it works out and post some measurements of the results.

                                                                            Click image for larger version

Name:	final_option-3-base_maple-padouk_F-.webp
Views:	114
Size:	40.6 KB
ID:	941913

                                                                            Click image for larger version

Name:	FullTweeterMnt_O-ring.webp
Views:	112
Size:	53.1 KB
ID:	941914
                                                                            Last edited by theSven; 23 June 2023, 23:05 Friday. Reason: Update image location
                                                                            Dan N.

                                                                            Comment

                                                                            • dlneubec
                                                                              Super Senior Member
                                                                              • Jan 2006
                                                                              • 1454

                                                                              #39
                                                                              Originally posted by JonW
                                                                              Have you thought about trying out the MCM “horn lens?”

                                                                              It’s cheap, easy to shave down to size, and you’d get more of a waveguide effect than your small ones. Your 3/4” deep, 3/4” roundover mini looks quite good. It’s just a suggestion if you want to get even more boost, etc.

                                                                              I actually use the MCM 6.5" waveguide (and still have an unused one somewhere) in one of my prototype omni designs, which became known as the waveguide omni (see the Mentor II omnidirectional thread). It would certainly work if you wanted more boost, at a lower frequency and if the 6.5" dia size worked for you in your particular design, solved your design objectives and was visually acceptable.

                                                                              It definitely needs to be treated on the back to keep it from resonating. I used a tub of plumbers putty on mine and that did a pretty good job. It makes more sense where the back end of the WG will not be visible, i.e. inside a box. In my case, it was not the way I wanted to go. The shallow waveguide seems to solve the issues from a design perspective and an aesthetic one as well.
                                                                              Dan N.

                                                                              Comment

                                                                              • Dennis H
                                                                                Ultra Senior Member
                                                                                • Aug 2002
                                                                                • 3791

                                                                                #40
                                                                                BTW, I have come up with a way to perhaps mount the HDS tweeter without removing the faceplate. It uses an O-ring...... My guess is that the FR may not be as smooth at the top end as when the flange is removed, but it may be good enough not to matter.
                                                                                I wonder if a felt ring might cause less diffraction than a rubber ring?

                                                                                Comment

                                                                                • cotdt
                                                                                  Senior Member
                                                                                  • Oct 2005
                                                                                  • 393

                                                                                  #41
                                                                                  Originally posted by dlneubec
                                                                                  I don't discount that using an old driver for a waveguide might work from a shape standpoint, but I would think it has to be a non resonant surface, which an woofer cone would not be, unless you used it as a form and filled the back in with Great Stuff, fiberglass resin, or something.
                                                                                  In any speaker that uses the Excel W22, the tweeter can set off the resonances of the W22 cone, yet I don't hear Orion owners or Nomad Audio owners complaining, so the issue shouldn't be too big, especially if the woofer cone is mounted rigidly to the baffle.

                                                                                  Comment

                                                                                  • toolsresearch
                                                                                    Junior Member
                                                                                    • May 2006
                                                                                    • 18

                                                                                    #42
                                                                                    Dan,

                                                                                    I had no idea that Bloomington is turning into the DIY capital that you seem to be making it!

                                                                                    Thanks for the posts...the waveguide idea is fantastic in principle...thorny in implementation. Yours is far and away the most clear and usable conception I've seen.

                                                                                    Comment

                                                                                    • JonW
                                                                                      Super Senior Member
                                                                                      • Jan 2006
                                                                                      • 1582

                                                                                      #43
                                                                                      Dan,

                                                                                      Lexan and hard wood. Those look very nice. Up to your usual standards.

                                                                                      Interesting idea of trying to keep on the faceplate. That might get more people to experiment. You don’t have to rip apart a tweeter. Best of all, you can mount the tweeter with screws in the faceplate so it won’t move. I like Dennis’ idea of a felt ring, rather than a rubber O-ring. If the O-ring sticks out even a little it might get in the way a bit. Or not.

                                                                                      OK, the MCCM still needs some support on the back. Good to know. But I agree that it ain’t pretty.




                                                                                      As someone from a more northern, more refined university town, I feel compelled to respond...
                                                                                      Originally posted by toolsresearch
                                                                                      I had no idea that Bloomington is turning into the DIY capital .
                                                                                      (cough) backwater (cough) backwater (cough, cough)
                                                                                      :W :

                                                                                      Comment

                                                                                      • dlneubec
                                                                                        Super Senior Member
                                                                                        • Jan 2006
                                                                                        • 1454

                                                                                        #44
                                                                                        Originally posted by cotdt
                                                                                        In any speaker that uses the Excel W22, the tweeter can set off the resonances of the W22 cone, yet I don't hear Orion owners or Nomad Audio owners complaining, so the issue shouldn't be too big, especially if the woofer cone is mounted rigidly to the baffle.
                                                                                        Well, we'll have to agree to disagree here. I an't recall reading any research on waveguides, where the waveguides advocated were anything but inert/non-resonant. In most cases they go to great pains to make them non resonant. You don't want the waveguide adding any sound of its own, under any circumstances that I can think of. That would be like making your baffles or box out of 1/8" hardboard, IMO.
                                                                                        Dan N.

                                                                                        Comment

                                                                                        • dlneubec
                                                                                          Super Senior Member
                                                                                          • Jan 2006
                                                                                          • 1454

                                                                                          #45
                                                                                          I considered felt rings. The concern I have with felt rings is getting them the exact size you need for the inside diameter, the correct thickness and the correct outside dia. I'm not sure you could cut them by hand accurate enough. The rings can't be much wider than 1/8"-1/4" and 1/8" thick or they will not compress enough to allow the faceplate to settle evenly. The 1/8" dia. O-ring should not have that problem. It may have others, however. I suspect that the 1-3/8" O-ring, slightly compressed, will fill the space adequately. Measurements will surely answer that question.

                                                                                          Another approach would be to do this. Remove the tweeter face plate. Add something to the surface so that nothing will stick to it. Mount the face plate only to the waveguide. Fill the space in by handwith silicone, caulk, wood filler, spackling compound, or something like that. Once dry, you would have to be able to remove the face plate so that the material could be finished and the tweeter body added before re-mounting it. If the O-ring does not work, I may try something like this.
                                                                                          Dan N.

                                                                                          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 "|||"