Announcement

Collapse
No announcement yet.

Updating my Blastoramas...

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

  • Updating my Blastoramas...

    Blastorama build here - http://www.diysubwoofers.org/projects/home/blastoramas/

    I've been mulling over a minor upgrade to the Blastoramas - basically replacing the piezos with other tweeters. The piezos have been doing the job - I just want to try something different to see if the difference is noticeable and better, and worth the price.

    The x-over point for the Blastoramas is pretty high, around 4.5kHz, to take advantage of the natural roll-off of the Beta 8 midbass drivers used in them. So any reasonably cheap tweeter that can extend to 4.kHz or below with a 12dB/octave x-over that's around or above the sensitivity of the Beta 8 drivers should work. Response should also be reasonably flat (or can be made to be reasonably flat by simple passive filtering), and of course it should be able to withstand the rigors being used for PA purposes. Oh, and if horn or wave-guide loaded, the maximum dimensions should not exceed 4" x 8".

    Any suggestions?

    I've looked at a few options on the PE site, but there's always something missing, e.g. no FR curve available, horn or wave-guide is too large, or the driver's just too expensive for consideration.
    Brian Steele
    www.diysubwoofers.org

  • #2
    Re: Updating my Blastoramas...

    Originally posted by Brian Steele View Post
    Blastorama build here - http://www.diysubwoofers.org/projects/home/blastoramas/

    I've been mulling over a minor upgrade to the Blastoramas - basically replacing the piezos with other tweeters. The piezos have been doing the job - I just want to try something different to see if the difference is noticeable and better, and worth the price.

    Yes.. 1 on 1 a dynamic tweeter of some type will make a very noticable difference, once you get a whole pile of them not so much but that's not the case here. I built some 15 3way home audio speakers when I first began this hobby and I used Motorolla piezos because they were brand new at the time, and was suitably impressed with the sound at typical home audio volume levels, my amp at the time was a 65w Sansui AU-D5. The second speakers I built were PA boxes and again my go to tweeter was a piezo, but in this application I wasn't as impessed, at higher drive levels the piezos detail dissolved into a smeared frying bacon sound, but I used them like that for a year or so. It wasn't until I had someone else drive the system and blow all the piezos that my hand was forced and I decided to try a dynamic super tweeter instead. What an eye opening that was, these drivers were more efficient so crossover changes were needed but right away I noticed much more clarity in the upper mids and much greater dynamic range.. these drivers retained their sound quality to much higher drive levels, powering these boxes I had a Yamaha M85 or Bryston 4B. That is my experience with 4 different original Motorola piezo models one of which is similar to what you are using so I think you'll find upgrading to a traditional compression driver of some type to be worth the cost.

    Suggestions... Eminence APT-150 or even the EM200 that is on clearance, both of these have a sensitivity up around 105dB so you're gonna have to add an l-pad but they'll fit and perform just fine crossed at 4.5khz.
    Paul O

    Comment


    • #3
      Re: Updating my Blastoramas...

      Originally posted by Paul O View Post
      The second speakers I built were PA boxes and again my go to tweeter was a piezo, but in this application I wasn't as impessed, at higher drive levels the piezos detail dissolved into a smeared frying bacon sound
      I haven't run into this issue with the piezos in my Blastorama, but that could be due to the characteristics of the filter I used, which peaks at around 8KHz and cuts off pretty rapidly below that, to compensate for the peak-dip-peak response of the GW 2x5 piezos (see image). I've been tempted to replace them with my frankenpiezos (which feature a much flatter raw response) and a different filter, but I'm thinking that if I'm going to do that, I may as well consider the option of using any driver for reproducing the high frequencies.

      Click image for larger version

Name:	20130106-blastorama-PCD.gif
Views:	2
Size:	44.5 KB
ID:	1170971


      I've considered the APT-150, but that ~5dB dip just above 6kHz, I'm a little concerned about that, as well as the 5kHz suggestions I'm seeing for filtering, which is a little higher than I'd like. The dip also appears in the FR curve for the EM200, so it may be a feature of the driver rather than the horns. Then there's the cost, which is about 14x that of the GW piezos they'd be replacing .
      Brian Steele
      www.diysubwoofers.org

      Comment


      • #4
        Re: Updating my Blastoramas...

        Originally posted by Brian Steele View Post
        I've considered the APT-150, but that ~5dB dip just above 6kHz, I'm a little concerned about that, as well as the 5kHz suggestions I'm seeing for filtering, which is a little higher than I'd like. The dip also appears in the FR curve for the EM200, so it may be a feature of the driver rather than the horns.
        How much power are you exposing these boxes to? It can't be anywhere near the 250w or so I regularly fed to my old PA speakers, if so.. after you add a 10dB or so pad you could use just about any crossover frequency you like(within reason) as you'll never get even remotely close to maxing these drivers out.. which is another thing in their favor. I'm not sure where you see the 5k crossover requirement either, I'm seeing a minimum of 3.5k at 12dB/oct for a 35w RMS power handling which should be plenty in your case.

        Originally posted by Brian Steele View Post
        Then there's the cost, which is about 14x that of the GW piezos they'd be replacing .
        True.. but then why are you even thinking of changing them, what is it about the sound you think you can improve on? IMO piezos prove that it doesn't cost much to make some noise at higher frequencies, and in this case "higher" is 10k and above, but down below that where all the details of the vocal range live(4k -8k) it takes some more engineering to do it well. You don't have to spend $30 a driver to hear the difference here, just about anything PE has listed as a bullet or supertweeter will outperform a piezo in this area.
        Paul O

        Comment


        • #5
          Re: Updating my Blastoramas...

          Originally posted by Paul O View Post
          How much power are you exposing these boxes to? It can't be anywhere near the 250w or so I regularly fed to my old PA speakers
          Piezos are rated up to about 30V. My Blastoramas are typically used with an iNuke 3000DSP that's limited via the built-in DSP to 33V per Blastorama, which works out to around 72W or so. For a 250W design, I'd likely opt to use two such piezos wired in series, rather than just one.


          Originally posted by Paul O View Post
          I'm not sure where you see the 5k crossover requirement either, I'm seeing a minimum of 3.5k at 12dB/oct for a 35w RMS power handling which should be plenty in your case.
          I've been reading a few comments on the use of these drivers. They seem to suggest that they're a bit fragile when crossed over lower.


          Originally posted by Paul O View Post
          just about anything PE has listed as a bullet or supertweeter will outperform a piezo in this area.
          I'm not sure about that. I've been reading some not so nice things about those cheaper bullets .

          Here's the on-axis (halfway between woofer and tweeter) 1M Fr and distortion of one of my Blastoramas down to 1kHz, with a bit of EQ provided by the iNuke's DSP. The response is actually flatter without the DSP, but I've found that a slight depression around 3~6 kHz seems to sound better in-room. I'm curious as if this can be bettered for a reasonable price using a different tweeter, and if this "better" response shows up in the measurements.

          Click image for larger version

Name:	20160104-blastorama-fr.png
Views:	1
Size:	12.0 KB
ID:	1170972
          Brian Steele
          www.diysubwoofers.org

          Comment


          • #6
            Re: Updating my Blastoramas...

            Originally posted by Brian Steele View Post
            I've been reading a few comments on the use of these drivers. They seem to suggest that they're a bit fragile when crossed over lower.
            I have experimented with the bullets I still have from those old PA speakers(http://www.mcmelectronics.com/produc...S-BT-2-/53-375), tried to make them work on a much larger horn with a 2.5khz crossover and no... they got really harsh with any appreciable power applied but that is well below their rated crossover so not surprising. In no way have these drivers been fragile, I had 8 at one time and none blew up. The APT-150 is a much more capable driver.

            Originally posted by Brian Steele View Post
            but I've found that a slight depression around 3~6 kHz seems to sound better in-room.
            Well that is the most sensitive part of human hearing so a depression in speaker response here sounds more natural, I find the same thing with the speakers I have built to be perfectly flat.. I end up adding a scoop through the mids with DSP.

            Originally posted by Brian Steele View Post
            I'm curious as if this can be bettered for a reasonable price using a different tweeter, and if this "better" response shows up in the measurements.
            I think you can improve the sound but I don't think you will see it in the FR since the mic doesn't differentiate between clean or distorted output, and IMO you're just not getting clean output from a single piezo at the lower end of it's range in this type of application.
            Paul O

            Comment


            • #7
              Re: Updating my Blastoramas...

              Originally posted by Paul O
              I think you can improve the sound but I don't think you will see it in the FR since the mic doesn't differentiate between clean or distorted output
              That's why I included the THD distortion plot as well, as any such distortion will be reflected there. From the distortion plot, it looks like THD remains fairly flat and sloping downwards throughout the FR curve from 1kHz up and is around -47dB (0.45%) at the piezo's x-over point (I think this is one advantage of the high x-over point I'm using on the piezo - it reduces the distortion at the lower end of its passband). I'm going to run a few other tests tomorrow to see if THD rises significantly at higher power levels. Of course it's possible and perhaps very likely that the distortion from a CD is even lower in its passband, but 0.45% seems pretty low already.
              Brian Steele
              www.diysubwoofers.org

              Comment


              • #8
                Re: Updating my Blastoramas...

                96dB/W/m 1,350Hz Fs, I've used it low, like 2,500Hz in my Leviathan setup. Plus they're pretty cheap, $15

                http://www.parts-express.com/tymphan...eter--264-1028

                रेतुर्न तो थे स्रोत
                return to the source
                leviathan system thread
                deadhorse thread
                shockwave build thread

                instagram :: greywarden_13

                in war, victory . . . in peace, vigilance . . . in death, sacrifice.

                Comment


                • #9
                  Re: Updating my Blastoramas...

                  Originally posted by Brian Steele View Post
                  Piezos are rated up to about 30V. My Blastoramas are typically used with an iNuke 3000DSP that's limited via the built-in DSP to 33V per Blastorama, which works out to around 72W or so.
                  As you may have guessed I have explored the power handling capacity of piezos.. who else do you know that has actually blown any? IMO you will start to hear the piezos limits well below that drive level, but you're probably nowhere near that right now. What is the energy content above 5khz of a 72w fullrange signal.. probably no more than about 10% or 7w right? I'd say you could safely double input power to 150w on those cabs without any issue but after that you're really gonna start to hear the limitations of the piezo design.

                  Originally posted by Brian Steele View Post
                  For a 250W design, I'd likely opt to use two such piezos wired in series, rather than just one.
                  Sure but now you're introducing comb filtering problems while a single APT-150 would be cruising along with only having to dissipate abut 2.5w. 10% of 250w is 25w but the driver is sitting behind a 10db pad so it only sees about 2.5w. The piezos would have to dissipate the full 25w since their sensitivity is so much lower so their distortion is gonna be much higher.
                  Paul O

                  Comment


                  • #10
                    Re: Updating my Blastoramas...

                    Originally posted by greywarden View Post
                    96dB/W/m 1,350Hz Fs, I've used it low, like 2,500Hz in my Leviathan setup. Plus they're pretty cheap, $15

                    http://www.parts-express.com/tymphan...eter--264-1028
                    Thanks. I'd looked at those. The specs seem to be along the lines of what I'm looking for. My concern about them however was how they would stand up to PA duty.
                    Brian Steele
                    www.diysubwoofers.org

                    Comment


                    • #11
                      Re: Updating my Blastoramas...

                      Originally posted by Paul O View Post
                      Sure but now you're introducing comb filtering problems while a single APT-150 would be cruising along with only having to dissipate abut 2.5w. 10% of 250w is 25w but the driver is sitting behind a 10db pad so it only sees about 2.5w. The piezos would have to dissipate the full 25w since their sensitivity is so much lower so their distortion is gonna be much higher.
                      The impedance of the GW 2x5 is about 300 Ohms @ 4.5kHz and around 115 Ohms @ 10 kHz. They will not be "seeing" (i.e. dissipating) anywhere near 25W at those frequencies. More like [email protected] and [email protected], assuming that they're driven by the same voltage that would result in 25W being dissipated over an 8 ohm impedance. Maybe the voltage drive at that level is still sufficient to cause distortion even with proper filtering. I'll find out later today when I do my tests .

                      I agree re the "comb-filtering" concern.
                      Brian Steele
                      www.diysubwoofers.org

                      Comment


                      • #12
                        Re: Updating my Blastoramas...

                        Originally posted by Paul O View Post
                        Well that is the most sensitive part of human hearing so a depression in speaker response here sounds more natural, I find the same thing with the speakers I have built to be perfectly flat.. I end up adding a scoop through the mids with DSP.
                        Funnily enough I find that it only sounds better that way in-doors. Outside, flat, at least from about 300 Hz up, sounds better to me.

                        Originally posted by Paul O View Post
                        I think you can improve the sound but I don't think you will see it in the FR since the mic doesn't differentiate between clean or distorted output, and IMO you're just not getting clean output from a single piezo at the lower end of it's range in this type of application.
                        I've attached some FR and distortion measurements I took of one of the Blastoramas this morning. Green represents a reasonable playback level. Red represents 10dB about that, and blue represents another 6dB on top of that. I didn't try any higher - it was getting way too loud at that point. Normal playback level for PA purposes is somewhere between the green and red levels I believe, though I know my daughter likely cranks it up higher when she's at the controls . The graph suggests peaks in THD at around 2.5% at 7kHz and again at 9.3kHz. THD actually *reduces* at 3kHz and again at 6kHz, which I wasn't expecting. I'm curious as to the reason for that...

                        Click image for larger version

Name:	20160105-blastorama-fr.png
Views:	1
Size:	16.5 KB
ID:	1170974
                        Brian Steele
                        www.diysubwoofers.org

                        Comment


                        • #13
                          Re: Updating my Blastoramas...

                          Originally posted by Brian Steele View Post
                          I've attached some FR and distortion measurements I took of one of the Blastoramas this morning. Green represents a reasonable playback level. Red represents 10dB about that, and blue represents another 6dB on top of that. I didn't try any higher - it was getting way too loud at that point. Normal playback level for PA purposes is somewhere between the green and red levels I believe, though I know my daughter likely cranks it up higher when she's at the controls . The graph suggests peaks in THD at around 2.5% at 7kHz and again at 9.3kHz. THD actually *reduces* at 3kHz and again at 6kHz, which I wasn't expecting. I'm curious as to the reason for that...
                          That is quite interesting, it's curious that distortion increases significantly in the 5k-10k region but remains almost unchanged above 10khz. To me this is evidence of the classic piezo sound that I keep referring to and why I feel they are better restricted to supertweeter duty above 10khz only. As to why the distortion peaks at some frequencies and reduces at others that sounds like classic constructive/destructive interference, only question is where is it happening... in the driver chamber or in the horn or a bit of both.
                          Paul O

                          Comment


                          • #14
                            Re: Updating my Blastoramas...

                            Originally posted by Paul O View Post
                            That is quite interesting, it's curious that distortion increases significantly in the 5k-10k region but remains almost unchanged above 10khz. To me this is evidence of the classic piezo sound that I keep referring to and why I feel they are better restricted to supertweeter duty above 10khz only. As to why the distortion peaks at some frequencies and reduces at others that sounds like classic constructive/destructive interference, only question is where is it happening... in the driver chamber or in the horn or a bit of both.
                            I think the rapid reduction in distortion above 10kHz is more a measurement artifact than anything else (as is the rapid drop in FR above 20kHz).

                            Any distortion artifacts above 10kHz will be beyond our hearing anyway.

                            The distortion peaks at 7kHz and again at 9.3kHz. For a 7kHz fundamental, the first distortion component will appear at 14kHz, and the next above that. I'm not sure how audible 2.5% distortion would be at that frequency. I've subsequently tested the Blastoramas up to the voltage limit with recorded music and I don't hear any changes at the top end.

                            BTW, the huge suckout between about 14kHz and 20kHz suggests that this particular piezo would be miserable as a supertweeter, LOL.

                            FWIW, I think the "piezo" sound has a lot more to do with the wildly varying FR through its passband for the cheaper ones, i.e. this particular piezo from GW has a significant rise around 4kHz, which would probably sound even more horrible once the piezo starts to get overloaded, as the overtones will still be in the audible passband. The 3x7 GW is even worse. Some of this can be treated though with some easy physical modifications, e.g. damping to the rear chamber, and damping the piezo element itself. I've addressed these modifications in other posts.

                            The particular piezos I've deployed in the Blastoramas have very little treatment done to them, apart from some damping on the back of the horn and a bit of silicone around where the cone is joined to the back, to eliminate any air leaks (the build was done before I got really interested in modifying piezos). I noticed that there are two narrow-band peaks above 10kHz in the FR - I notched them out with the iNuke's DSP and live recorded instruments like cymbals seem to sound a lot more realistic now. Hmm...

                            I think I've convinced myself at this point - I'm going to do two mods to the Blastoramas - one replacing the existing piezo with a frankenpiezo (easy to do seeing that I have two on hand), and the other replacing the piezo with an APT-150 or something similar. Then compare the two to see what the difference is really like.
                            Brian Steele
                            www.diysubwoofers.org

                            Comment


                            • #15
                              Re: Updating my Blastoramas...

                              Originally posted by Brian Steele View Post
                              I think I've convinced myself at this point - I'm going to do two mods to the Blastoramas - one replacing the existing piezo with a frankenpiezo (easy to do seeing that I have two on hand), and the other replacing the piezo with an APT-150 or something similar. Then compare the two to see what the difference is really like.

                              That would be informative for sure. It would be interesting if I was proved wrong and it is possible to make a piezo sound as good as conventionial compression driver. I have never seen a speaker that doesn't benefit from some EQ either including the various tweeters and compression drivers I have used over the years, and I must say It sure helps to have a measurement system and some DSP to correct anomalies. All this talk makes me want to pull out the boxes I built a couple years ago and stick them in front of the test mic again, they include a Beta8 and a compression driver on a Dayton 8" round waveguide, The current CD in them is a complete POS that I pulled out of some white van speakers which are horrible sounding, but i do have adapters to mount the BT2 drivers on the waveguides. I already know these drivers won't handle the current crossover freq but I wonder what they would do with the crossover pushed back up around the 5k area. I thought I had a measurement of this box but I can't seem to find it at the moment..
                              Paul O

                              Comment

                              Working...
                              X