Announcement

Collapse
No announcement yet.

New first build, not very exciting

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

  • New first build, not very exciting

    I have made a lot of prototypes, but nothing that has stayed in the living room. I am now building 2 surrounds that will.

    Before I realized the amount of work and money a nice cabinet took, I planned on building fronts with Dayton RS100's and a small neo-tweater to learn. I was already planning on replacing them with something better once I knew what I was doing, and decided to skip the first project. I've been playing with the Esoterics, but that is expensive and time consuming. I wanted something simple I can actually use in the near future.

    For surrounds, I currently have a 7.1 system hardwired to my receiver with very cheap speakers. I have a wireless Dayton Audio XRA25 that will work with the sub and power 2 speakers, supposedly at 25 watts into 4 ohms. The plan is to gain WAF by eliminating wires, and improve sound by upgrading 4 crappy speakers to 2 better DIY ones.

    I need something:
    high efficiency into 4 ohms.
    don't need to be full-range, but base extension is nice to have.
    not necessarily ruler-flat, but decent audio quality without too much distortion
    a high WAF
    Sealed box for placement flexibility

    The RS100-4 were a target on my early prototypes because they can play low for such a small woofer, and has low distortion and reasonable quality. I already have them, and they are perfect for this project. To gain the efficiency I need, I plan on 4 drivers in each box. Each box will be 5x5x24 inches high - close to the recommended size, high enough not to need a huge stand, but thin and small enough for WAF. To cut costs, I am eliminating the neo-tweeters (cost of crossover parts) which leaves me crossover-less (probably need a notch filter), which makes this a very boring build as far as electrical engineering and acoustical adjustments go. Still an adventure for me, as it will be the first time I try veneering something. I am hoping comb-filter isn't much of an issue, but it would be a possible issue even with a neo-tweeter. Surrounds make noise (bullets, explosions, etc) during movies, so hopefully the top-end will be acceptable.

    Right now, I am trying to decide if I need bracing, and how much to use.

  • #2
    I'm not a fan of single driver systems, but it might work out for surround duty. Do you have measurement capability? That's important to getting reasonable response no matter what you're designing.

    If you haven't already discovered joewoodworker.com, it's a wealth of veneering knowledge. His companion site veneersupplies.com is a good source of quality veneer. I've purchased raw veneer there and paperbacked veneer at tapeease.com.

    A first project is always exciting - veneering learning curve is steep but the results can be amazing. Enjoy and keep us posted with progress pictures.

    Comment


    • #3
      I have a calibrated microphone, and plan to use it, but the goal for this build is to get it right on the first try. There isn't much I can adjust once I measure, and I don't have an anechoic chamber.

      Comment


      • #4
        Not quite sure I understand your "get it right the first time comment. The normal procedure is build the box, put the driver(s) in, measure, simulate filter, build, measure to verify and adjust as required. The measurement box can be an unfinished test article or the finished piece. You'll want to design it so you can install whatever filter you need after the fact. With such a small driver that probably means a removable back or bottom.

        If you don't already have a CAD program consider http://kimmosaunisto.net/Software/Software.html To get the most out of it you'll want ARTA measurement software as well.

        Comment


        • #5
          Originally posted by BobEllis View Post
          Not quite sure I understand your "get it right the first time comment. The normal procedure is build the box, put the driver(s) in, measure, simulate filter, build, measure to verify and adjust as required. The measurement box can be an unfinished test article or the finished piece. You'll want to design it so you can install whatever filter you need after the fact. With such a small driver that probably means a removable back or bottom.

          If you don't already have a CAD program consider http://kimmosaunisto.net/Software/Software.html To get the most out of it you'll want ARTA measurement software as well.
          I probably will include a notch filter which will get updated as I measure. I don't plan to change cabinet size or drivers unless things are way off, and I can't adjust the crossover because I don't have one.

          Comment


          • #6
            Originally posted by bvbellomo View Post
            I probably will include a notch filter which will get updated as I measure. I don't plan to change cabinet size or drivers unless things are way off, and I can't adjust the crossover because I don't have one.
            Ah, but that is not necessarily true. Many full range diver systems incorprate a filter to implement baffle step correction. If you flush mount them in the wall, then that isn't necessary, but otherwise I would listen to Bob's advice.
            the AudioWorx
            Natalie P
            M8ta
            Modula Neo DCC
            Modula MT XE
            Modula Xtreme
            Isiris
            Wavecor Ardent

            SMJ
            Minerva Monitor
            Calliope
            Ardent D

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


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

            Comment


            • #7
              Originally posted by bvbellomo View Post
              I have made a lot of prototypes, but nothing that has stayed in the living room. I am now building 2 surrounds that will.

              Before I realized the amount of work and money a nice cabinet took, I planned on building fronts with Dayton RS100's and a small neo-tweater to learn. I was already planning on replacing them with something better once I knew what I was doing, and decided to skip the first project. I've been playing with the Esoterics, but that is expensive and time consuming. I wanted something simple I can actually use in the near future.

              For surrounds, I currently have a 7.1 system hardwired to my receiver with very cheap speakers. I have a wireless Dayton Audio XRA25 that will work with the sub and power 2 speakers, supposedly at 25 watts into 4 ohms. The plan is to gain WAF by eliminating wires, and improve sound by upgrading 4 crappy speakers to 2 better DIY ones.

              I need something:
              high efficiency into 4 ohms.
              don't need to be full-range, but base extension is nice to have.
              not necessarily ruler-flat, but decent audio quality without too much distortion
              a high WAF
              Sealed box for placement flexibility

              The RS100-4 were a target on my early prototypes because they can play low for such a small woofer, and has low distortion and reasonable quality. I already have them, and they are perfect for this project. To gain the efficiency I need, I plan on 4 drivers in each box. Each box will be 5x5x24 inches high - close to the recommended size, high enough not to need a huge stand, but thin and small enough for WAF. To cut costs, I am eliminating the neo-tweeters (cost of crossover parts) which leaves me crossover-less (probably need a notch filter), which makes this a very boring build as far as electrical engineering and acoustical adjustments go. Still an adventure for me, as it will be the first time I try veneering something. I am hoping comb-filter isn't much of an issue, but it would be a possible issue even with a neo-tweeter. Surrounds make noise (bullets, explosions, etc) during movies, so hopefully the top-end will be acceptable.

              Right now, I am trying to decide if I need bracing, and how much to use.
              I would recommend you look at this:
              http://www.divine-audio.com/pecorino/ maybe use for surrounds.
              I could try to simulate 4 RS100's and 1 nd20 in a mmtmm design but i would definitely say add the tweeter and crossover parts. i wouldn't do it without the summing of the 3" drivers above ~8khz will create lulls because of the center to center distance (line array's run into the same issue) which is why the CBT arrays use a TON of 3/4 in tweeters to reduce the center to center distance http://projectgallery.parts-express....y-loudspeaker/

              Comment


              • #8
                Originally posted by JonMarsh View Post
                Ah, but that is not necessarily true. Many full range diver systems incorprate a filter to implement baffle step correction. If you flush mount them in the wall, then that isn't necessary, but otherwise I would listen to Bob's advice.
                This is just semantics. To me, unless you have drivers operating differently at different frequencies, it isn't a cross over, since nothing gets crossed over. But I agree more likely than not, I need a filter of some sort.

                Comment


                • #9
                  Originally posted by hyperducky View Post
                  I would recommend you look at this:
                  http://www.divine-audio.com/pecorino/ maybe use for surrounds.
                  I could try to simulate 4 RS100's and 1 nd20 in a mmtmm design but i would definitely say add the tweeter and crossover parts. i wouldn't do it without the summing of the 3" drivers above ~8khz will create lulls because of the center to center distance (line array's run into the same issue) which is why the CBT arrays use a TON of 3/4 in tweeters to reduce the center to center distance http://projectgallery.parts-express....y-loudspeaker/
                  I mentioned comb-filtering earlier. It is still a problem (at least in theory) with the ND20 at 3kHz, which is pushing that tweeter too low. That pecorino is similar to what I planned on my 'temporary mains' when I bought the RS100s, but different goals from what I want now. I agree crossing this to a good tweeter around 2.5Hz would sound better, but that doubles the budget. I don't know much about designing CBT arrays, but it looks way more complicated than I intended for surrounds.

                  I am not sure where you get the 8k lulls. I see the problem starting much lower - around 2kHz (when drivers are 1/2 wavelength apart). I assume if comb-filtering is a problem, it is a problem that can be measured - right? If it is bad enough, I will add a tweeter (and know where I have to cross to avoid it). But as I said, ruler flat, especially above telephone band, is not a goal here.

                  Comment


                  • #10
                    Originally posted by bvbellomo View Post
                    This is just semantics. To me, unless you have drivers operating differently at different frequencies, it isn't a cross over, since nothing gets crossed over. But I agree more likely than not, I need a filter of some sort.
                    Semantics or not the word 'crossover' is used generically to describe any filter components added. I get where you are coming from and if we all wanted to remain perfectly technically accurate we'd probably adopt a different word to describe single driver equalisation. This is just one of those situations where everyone knows what you are talking about if you say crossover, even if it's not technically crossing two drivers over to one another.

                    From another point of view single driver systems tend to be quite popular among first timers because they have the potential to sound fairly decent without any of the usual difficulties encountered with multi-way systems. Most do tend to require some form of equalisation, baffle step compensation would be needed, as Jon has said, unless the speakers are mounted into a wall. Lots of full range drivers also tend to have a rising and/or peaky upper end response. Sometimes the rising response can be corrected, at least for critical listening, by altering how far off axis you are listening to them via the amount of toe-in. Of course this is impossible with an in-wall speaker.

                    Severe peaking will need a notch or two but really the attractive nature of the single driver system is that you can just hook it up and play music. If you can then measure, all the better. You don't need an anechoic chamber, gated measurements get around this limitation, at least for everything besides lower frequencies. But it's relatively easy to apply some basic EQ via a PC as the signal source until you arrive at something you like, then you have a good starting point for what your passive filter might potentially do.
                    What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
                    5th Element, otherwise known as Matt.
                    Now with website. www.5een.co.uk Still under construction.

                    Comment


                    • #11
                      Originally posted by bvbellomo View Post
                      I mentioned comb-filtering earlier. It is still a problem (at least in theory) with the ND20 at 3kHz, which is pushing that tweeter too low. That pecorino is similar to what I planned on my 'temporary mains' when I bought the RS100s, but different goals from what I want now. I agree crossing this to a good tweeter around 2.5Hz would sound better, but that doubles the budget. I don't know much about designing CBT arrays, but it looks way more complicated than I intended for surrounds.

                      I am not sure where you get the 8k lulls. I see the problem starting much lower - around 2kHz (when drivers are 1/2 wavelength apart). I assume if comb-filtering is a problem, it is a problem that can be measured - right? If it is bad enough, I will add a tweeter (and know where I have to cross to avoid it). But as I said, ruler flat, especially above telephone band, is not a goal here.
                      I missed the part of your post about comb filtering that's what I was trying to point out. I guess if you want 4 per side you would just have to live with the comb filtering issue. the solutions to getting 4 to work with even a cheap tweeter (nd20) would make the parts count on the crossover and blow the budget you have said. I was giving the CBT example because they discuss the solutions to comb filtering. So I guess I have a hard time understanding your goals because to me the issues with the higher octave that you would run into due to comb filtering would be a nonstarter for me.

                      If you haven't purchased the drivers yet I would suggest you try the overnight sensations MTM (i haven't heard it but many reviews are quite good)

                      Comment


                      • #12
                        As I said, the goal of this build was to use drivers I already purchased, develop sealed 4 ohm high efficiency speakers, and work on carpentry skills. Overnight sensations are 8 ohm, ported, extremely low efficiency and come with premade cabinets. They are also too wide and too deep.

                        I've read a lot about comb filtering, and done some modeling on the computer. I've reached the conclusions:
                        1) I can't be certain of the exact effect without physically building the system.
                        2) The problem is location dependent, and therefore might not be measurable.
                        3) If the problem is theoretical, but not measurable, and not audible, I am happy to live with it for surrounds.

                        If anyone has anything I haven't read regarding 1) and 2), I would be interested reading it.

                        Comment


                        • #13
                          1) Comb filtering can be simulated with a high level of accuracy and precision, you just need the right software, but generally rules of thumb get you where you need to go.
                          2) Everything with loudspeakers is location dependent and they can all be measured.
                          3) It is not just theoretical, it is not unmeasurable but the audibility of such a problem varies significantly upon the design (ie what frequencies this would start at).

                          A true line array needs drivers of a certain diameter (small) for it to work well up to certain frequencies. Comb filtering gets less audible the tighter the 'peaks and nulls' are together but the floor to ceiling approach for a line array relies on the drivers being below a certain C2C distance re frequency so that a cylindrical wave front is accurately created. Normally you see a line of tiny tweeters next to a line of mid/bass drivers with the two crossed over at an optimum frequency. Naturally you are going to get severe comb filtering on the upper end of the tweeters response but due to the nature of the filtering it tends to go unnoticed, or at least the benefits of the array outweigh the cons.

                          Some people choose to do arrays with a line of larger full range drivers. This in and of itself worsens the comb filtering issue but to those people brings about other gains that they think are worth it.

                          I don't have time to read the entire thread but if the gist is using 4 RS100s per side in a surround loudspeaker then I'd say just go for it. Running them fullrange in a series/parallel configuration with, or without any kind of EQ. Surround duties are more about creating a spread of sound to add ambience, rather than pin point localisation cues. Some designs go so far as to have drivers mounted on opposing surfaces with the options to wire them in or out of phase to create the most spacious presentation.

                          If you were wanting to go with a front loudspeaker using the RS100s then for sure you'd want to either add in a tweeter or use the other drivers as fillers for a 1.5 way system, if such a thing exists.

                          A 2.5 way MMTMM or MMMTM 2.5 way would definitely be two worthwhile options you could consider. The MTM section would form your basic MTM with the Ms wired in series. The outer Ms would also be wired in series but be configured as .5 way drivers and would be filtered and wired in parallel to the main Ms and compensate for baffle step losses.
                          What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
                          5th Element, otherwise known as Matt.
                          Now with website. www.5een.co.uk Still under construction.

                          Comment


                          • #14
                            Comb filtering with a spherical radiation without room reflections is acceptable at listening distance. To simulate with a high degree of accuracy, I'd need to model every surface in the room, including how much sound it reflects and absorbs. The shape of the sound wave is also not spherical, and it depends on the frequency. So this almost seems possible until I realize I have 2 ears, which aren't point sources! Trying to model human hearing would be ridiculous.

                            I plan to finish the build, and measure from the listening position. If I can measure comb filtering, I will decide if it is bad enough to add a tweeter and know how low I have to cross my tweeter. If comb filtering doesn't show, I will assume it is there but I can't measure it, and use my ears to judge if the result is acceptable.

                            I assume a measurement less than 1 foot away would show horrible comb filtering - I don't care about that, as I don't plan to wear these as headphones.

                            Comment


                            • #15
                              So far, I am done rough cutting:
                              https://drive.google.com/open?id=0B7...kFEbWJ4MGc2c00


                              What I've learned so far:

                              I am getting better with the router and jig, but that part was still almost all of the time.

                              The flange doesn't quite fit. My hole is 3 7/8" and the spec says the flange is 3.86". I was actually worried my hole would be too big, and it is too small. I am not sure what is off, but is is small enough I can fix it with the chisel. When I do the 2nd speaker, is it is worth going out to 3 15/16ths or not?

                              Making holes that almost touch is very hard. Imperfections even 1/16 of an inch shows from far away. I am hoping once I enlarge the holes and veneer, I won't have as much of a problem. If I do, the wife wants grills, so it won't matter. I could have added an extra 1/8th inch between drivers, but this would make comb filtering worse (although probably not by much).

                              The small piece with the hole was meant to be a brace. Do I need bracing? How can I tell? With 4 drivers, I am not sure I can fit bracing.

                              Getting the flange depth correct seems impossible. Even if I had calipers and measured the flange, I have no idea how I'd set the router perfectly. I am too shallow, and will have to fix this with the chisel as well. Even if it is perfect, how do I know how much to leave for the veneer? If it is 1/40th of an inch, is it still that wide once it is glued and covered with a finish? This is probably being too OCD.

                              Comment


                              • #16
                                I glued and screwed everything together today. I am not a fan of screws into MDF - I had 2 minor splits on the end, both fixed with glue and running a screw through the other way. I would think the glue is strong enough to hold without the screws, and the screws won't be very strong (since the mdf isn't very strong). I may try to hold the next speaker together just with glue. Now I need to let it dry, sand and then try to veneer.

                                Comment


                                • #17
                                  pre-drill for screws and even Brad Nails.
                                  Consider buying a Biscuit setup.
                                  Use a Lock Miter router bit for less glue up movement
                                  Consider adjustable 90 degree positioning clamps used
                                  for framing.

                                  making plumb square & level boxes, cabinets (trim carpentry)
                                  or _______ is not as easy as it looks.

                                  Comment


                                  • #18
                                    Originally posted by Carl V View Post
                                    pre-drill for screws and even Brad Nails.
                                    Consider buying a Biscuit setup.
                                    Use a Lock Miter router bit for less glue up movement
                                    Consider adjustable 90 degree positioning clamps used
                                    for framing.

                                    making plumb square & level boxes, cabinets (trim carpentry)
                                    or _______ is not as easy as it looks.
                                    Square, plumb and level isn't that important, at least to me. If the pieces are cut precisely (I have a small table saw), they easily go together more square than the walls of my house.

                                    I am predrilling, and agree a biscuit setup would be the way to go if I were making dozens of these. I still think conventional clamps and glue would work better than what I did yesterday. Try hard to drill a 2.5 inch deep hole straight through the middle of. 75 inch board, drill a larger hole to counter sink, screw in the screw and pray it doesn't split, repeat 20 times. I don't believe screws are stronger in mdf.

                                    Comment


                                    • #19
                                      Originally posted by bvbellomo View Post
                                      I glued and screwed everything together today. I am not a fan of screws into MDF - I had 2 minor splits on the end, both fixed with glue and running a screw through the other way. I would think the glue is strong enough to hold without the screws, and the screws won't be very strong (since the mdf isn't very strong). I may try to hold the next speaker together just with glue. Now I need to let it dry, sand and then try to veneer.
                                      Running screws into mdf, end on, is typically a loosing preposition. If you have to, use a coarse thread for better holding. +1 on the suggestions that Carl made, if you can find a gently used biscuit jointer, they can be used on all kinds of stuff besides speakers ...... you would be surprised how much they help with alignment and overall strength on a joint. Miters, rababbet, etc. all make for a far stronger joint due to the increased surface area for glue.
                                      Hold on to your butts - It's about to get Musical!



                                      WEBSITE: http://www.smjaudio.com/

                                      Comment


                                      • #20
                                        When can I sand and veneer? The white glue I used should be full strength after 24 hours, but I've read some of the moisture may still be present in the wood and sanding too early will create dips later.

                                        Comment


                                        • #21
                                          Veneer attached much easier than I expected, just with white glue. Trimming it was much harder than I expected, and I ruined some around the edges. First speaker will have some filled pieces in the corners, I need to find a better way, or just get better with a razor.

                                          Comment


                                          • #22
                                            I put everything together, sealed the drivers, but not the wire-plate, so I can add electronics later. I still have some finishing work to do as well. Maybe black nail polish on the screws?

                                            Image not available

                                            All and all, not too bad for an inexperienced carpenter with a router, dremel, small table saw and tube of glue. Lots I'd do better if I could do over, but that was the whole point.
                                            Last edited by svenarajala; 30 April 2023, 18:34 Sunday. Reason: Update image location

                                            Comment


                                            • #23
                                              Without anything other than drivers and cabinet, here is frequency response and distortion:

                                              Image not available

                                              I suspect the problem at 580Hz is room related, since I measure similar problems with other speakers. This is 48 inches, twice the length of my cabinet, so I haven't ruled out cabinet resonance. I see no reason for baffle step compensation, which I'd expect around 2700Hz. This could be from near-wall placement, or the driver's response.

                                              Everyone warned me about comb filtering, so here is an unsmoothed graph above 1k:

                                              Image not available

                                              And again above 10k:

                                              Image not available

                                              Everyone should expect everything ok until 3500Hz (drivers 1 wavelength apart), with a null at 7000Hz and worse at 14000Hz. As I expected, given 4 woofers, room reflection and my less than perfect spacing of drivers, this doesn't show up far away.

                                              45 degrees off axis, everything falls off above 4500Hz, which is in general agreement with Dayton's charts.

                                              Image not available

                                              If I do any kind of notch filter, I will take that into consideration. I might leave these as is - as surrounds I will never listen to them without the receiver doing fancy stuff, so I can equalize them.
                                              Last edited by svenarajala; 30 April 2023, 18:35 Sunday. Reason: Update image location

                                              Comment


                                              • #24
                                                Critical listening is almost impossible with a surround, at least with the source material I tried. I did replace my center channel for a while, and found the result mostly acceptable but nothing special. The lack of upper few octaves is noticeable, especially off axis. Probably not a great build for main speakers, but they should work as surrounds. They have the excellent sensitivity I was looking for.

                                                Comment

                                                Working...
                                                X
                                                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 "|||"