Announcement

Collapse
No announcement yet.

PCD7 help? My project has stalled. SEOS12

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

  • PCD7 help? My project has stalled. SEOS12

    I have been trying to hold off as long as possible before asking for help on my project. I wanted to learn as much as I could before I started asking for serious help in my project.
    I am trying really hard not to get lazy and ask for people just to design stuff for me as that is not my style at all. This is to learn, but I can't do much more on my own very efficiently.


    Here is my goal.

    I have a SEOS12 with de250 copy and I would like to build the best crossover I can for it and the Cerwin Vega 12" I had re-coned (D7).



    WHY use the CV 12" you smart people may ask?

    I can buy new woofers with pre-built crossovers and custom machined baffles to make the building easier. But I really want to learn PCD7 and other means to design crossovers and buying everything new would take out a lot of the learning. I have lots of speakers and drivers laying around that I can learn from as well. This is just a good starting point for me. After the research I have done, I know I am not going to be good at passive crossovers like most on here. But I am going to try to learn as much as I can. I may go active in the distant future, and may buy new woofers. Always an option. Re-purposing the woofers or selling speakers as original complete D7 are also viable options. But for now, the CV's are my learning toys.




    Here is where I have gotten so far.



    I am currently using a passive generic econowave crossover that was designed for a different horn and different compression driver that can be used as a starter for any 4 ohm under 95db woofer. I knew it was not optimal when I built it. I just wanted to have some noise until I could decide what to do with the waveguide and CD. It actually sounds pretty good. A little less linear that the original CV D7 crossover and mids/highs BUT the sound is less irritating due to waveguide and CD quality. I was actually surprised it worked as well as it did. I did sense something was off before measuring, and one I did measured it all came to light. The blips at 2k, and the fact I have the Lpad cranked all the way up, but it's still not equal output between the woofer and tweet. I need to adjust that so I can turn up the CD a bit.



    The horn is sitting on top of cabinet, so I can easily move it to check/measure/alter time alignment. It is not on the baffle.

    I can measure speakers with my LIMP wiring jig. I have Holmimpulse on my computer now and have played with it a bit after purchasing an EMM6. I am using an ART mic preamp with phantom power. So far, it has worked out very well. I have measured lots of speakers I have laying around, just for the practice. I have learned a bit just from playing around.

    I can't tell you everything I do and don't know. I ask you be patient if you are willing to help me, and I won't take it personal if you think I don't know something, even if I do.


    I am ok with getting measurements from Holmimpulse and exporting the FRD files. I also am ok with converting the SPL output on Excel. I have not gotten comfortable with Response Modeler, yet. I got comfortable enough with altering the SPL in Excel that I just do that quickly. I have no way of checking true sensitivity of my speakers. I don't have the money for a Radio Shack SPL meter, yet.

    Here is where I am stuck and need help/insight.

    I understand the basics of what compression driver compensation is.
    I don't understand how it is implemented in a passive or active crossover.
    I don't even know what it looks like on a schematic.
    I DO know what 12db vs 6db etc. looks like on in a passive x-over.
    I need to learn notch filters, especially for the 12" peak just under 2k on the CV.


    As far as specific PCD7 issues, this is where it is going to be a long learning process I think. Mainly because it is my first time EVER getting this deep into x-over design. Completely uncharted territory for me.

    I had no issues importing the FRD files. I have not imported a ZMA file, yet.
    Because I don't know about the CD compensation and notch filters, I don't know how to make one on PCD. I am completely clueless. I have tried using search engines and also using instructional guides. I have started re-reading the ECONOWAVE thread zilch had looking for specific info on making a CD compensation, and having trouble find straight forward help in the older info. I know Wayne Parham is really good with the CD compensation, but may be too technical for me to read his white papers right now.

    I can re-measure everything if need be. I will post the data I have now. Both raw driver responses, with x-over on individual components, and both playing together.

    I will not be able to afford crossover parts right now. I am trying to learn the design tools a bit better before I start attempting to invest money any farther. I was hoping the simulations would help my knowledge a bit while I try to save up.

    Thanks to any who have taken the time to read this. Ask me anything if you need info that can give me insight. Help me help you help myself, so to speak.

    I am calling them "CERWAVES" :D

    Now for the data I have so far.
    Attached Files
    Last edited by tmhutson; 01-02-2014, 12:11 AM. Reason: Added graph

  • #2
    Re: PCD7 help? My project has stalled. SEOS12

    hi,

    here's the following steps that i usually do with xo design under pcd7
    1. determined Acoustic Offsets, put the mic at Tw axis @80cm, measured each Tw and M individually on the same level, then measured altogether. Put the T and M FRD in PCD, then set the "Listening / Measuring Distance" = 0.8, then go to "Import Overlay" just under the "System Response" graph. Match the BLACK lines with the GREY lines.
    2. Measure the Driver's impedance using DATS, or using a breakout cable with REW on your laptop.
    3. On a fresh PCD, load all FRD and ZMA, put the necessary offsets, such as vertical (distance between your tweeter axis to the Midwoofer axis), Driver Offset is the Acoustic Offset (Z-offset)

    You will need to measure with phase on to aligned the phase between drivers in your crossover region.

    i hope that helps.

    cheers
    henry

    Comment


    • #3
      Re: PCD7 help? My project has stalled. SEOS12

      I am no expert on compression drivers but I am pretty sure you would treat them ad a normal driver. I have never heard of compression driver compensation unless you are talking about a l pad to bring the level down.

      If you can use holm and measure impedance with limp I don't think you really need response modeler ( unless this is a program for tuning a enclosure ). If I remember correctly response modeler is for predicting the affects of the exteral cab dimesions and room affects. Just model your box for the tuning you want first, measure speakers in said box, then put that in pcd.

      If you can build it and measure it we can help guide you pretty easily ;)

      Comment


      • #4
        Re: PCD7 help? My project has stalled. SEOS12

        Originally posted by henry218 View Post
        hi,

        here's the following steps that i usually do with xo design under pcd7
        1. determined Acoustic Offsets, put the mic at Tw axis @80cm, measured each Tw and M individually on the same level, then measured altogether. Put the T and M FRD in PCD, then set the "Listening / Measuring Distance" = 0.8, then go to "Import Overlay" just under the "System Response" graph. Match the BLACK lines with the GREY lines.
        2. Measure the Driver's impedance using DATS, or using a breakout cable with REW on your laptop.
        3. On a fresh PCD, load all FRD and ZMA, put the necessary offsets, such as vertical (distance between your tweeter axis to the Midwoofer axis), Driver Offset is the Acoustic Offset (Z-offset)

        You will need to measure with phase on to aligned the phase between drivers in your crossover region.

        i hope that helps.

        cheers
        henry
        +1 Jeff B. has a good tutorial on finding the acoustic offset. You can also use the time lock feature in holm but it will keep you from being able to use pcd at it's full potential. I would recommend doing it the hard way to learn first ( really not too hard ).
        Edit: Found it http://techtalk.parts-express.com/sh...coustic-Offset

        Comment


        • #5
          Re: PCD7 help? My project has stalled. SEOS12

          On your 3rd diag., shoot for each driver to be -6dB down at the cross point. If they're in phase at Fc, they'll sum +6dB (and "fill in" to make your line straight across, like a bridge).

          Comment


          • #6
            Re: PCD7 help? My project has stalled. SEOS12

            He says he hasn't taken impedance measurements. He needs to do that before worrying about the xo.

            Comment


            • #7
              Re: PCD7 help? My project has stalled. SEOS12

              I did for the woofer, but did not save the file as I didn't know how to export ZMA or whatever at the time, and I have the free ARTA version. I will make that a goal to redo the measurements so I can have that info ready! I have been worried about learning the programs, making sure they work than trying to get the data all collected and organized. I am getting to the data collecting phase right now...

              Chris, thanks for that little bit. Little tips like that I can save for later. I can re-open the Holmimpulse files to get phase and save the FR/phase picture to post. I just didn't want a messy looking picture for starters. The phase on the Holmimpulse can get rather messy when you have woofer, driver and combined response all on one graph!

              Henry, I will go over the steps again to re-measure with time alignment set up. I was only messing with time alignment last time to see how it worked. I still need to practice getting the impulses lined up after locking the zero time with woofer. I want to get the tweeter physically time aligned first. I will measure at the CTC space. I do plan on setting up the speakers so the CTC is at ear level. So, I will measure this way.

              Here is pic of the CV impedance.

              Oh, crap. Just found screen shot, and I stopped the measurement at 1k, because I only needed for looking at T/S for box design...

              Wow, I was not expecting answers this fast. If I have time tomorrow, I will get my measuring rig set up. I will re-do the woofer Le and re-measure the woofer and tweeter to get proper Z offset.

              I also need to investigate "minimum phase" more, as I know it's important.

              Whatever you guys need, just tell me...

              Thanks a lot!
              Attached Files

              Comment


              • #8
                Re: PCD7 help? My project has stalled. SEOS12

                Killa, thanks for the link. I added it to my list of links that seems to keep on growing. :D It's in the omnimic forum, but am going to read it over. Jeff added a pdf to his post, so I will go over it tomorrow. Too late for me to do much of anything tonight, anyways. Lots to chew on...

                Comment


                • #9
                  Re: PCD7 help? My project has stalled. SEOS12

                  As far as measurements it would be helpful to know things like measurement distance and what axis measurements are taken on. It seems like you have a decent grasp on this part but it is hard for us to know how much you know.
                  The more info the better

                  Comment


                  • #10
                    Re: PCD7 help? My project has stalled. SEOS12

                    Originally posted by tmhutson View Post
                    Killa, thanks for the link. I added it to my list of links that seems to keep on growing. :D It's in the omnimic forum, but am going to read it over. Jeff added a pdf to his post, so I will go over it tomorrow. Too late for me to do much of anything tonight, anyways. Lots to chew on...
                    It works for any mic

                    Comment


                    • #11
                      Re: PCD7 help? My project has stalled. SEOS12

                      I just skimmed over the pdf to see what the basics were, and I think I get the basic idea. I will investigate tomorrow. I only commented on omnimic as I would hate to ask questions in that thread as I don't have omnimic. But I can still learn from the thread, though! Thanks. That pdf will help!

                      I will measure from the CTC of woofer and tweeter. It's actually measured and marked on the speaker box :D I will measure the mic distance from front of horn and surround of speaker.

                      I am going to use the zero lock to get the tweeter and woofer time aligned from listening axis first.

                      Comment


                      • #12
                        Re: PCD7 help? My project has stalled. SEOS12

                        If you use the Holm time lock and physically align the drivers then you do not need to use the method described in Jeff's instructions. Your offsets would end up being zeroes as you already will have time aligned (physically instead of with the crossover) the drivers.

                        Any time you use the time lock feature your acoustic offsets will be included in your Holm measurements and you don't need to use the method in Jeff's PDF. The downside to using time lock is that you won't be able use pcd to it's full capacity. This means you will not be able to use pcd to predict power response or of axis response. Of course you can always measure these things afterwards.

                        Comment


                        • #13
                          Re: PCD7 help? My project has stalled. SEOS12

                          Originally posted by killa View Post

                          Any time you use the time lock feature your acoustic offsets will be included in your Holm measurements and you don't need to use the method in Jeff's PDF. The downside to using time lock is that you won't be able use pcd to it's full capacity. This means you will not be able to use pcd to predict power response or of axis response. Of course you can always measure these things afterwards.
                          Since he is using a waveguide the off axis will not sum properly in PCD. You will need to measure that on your own. It is accurate on axis which is the most important axis. From about 5kHz on up you may get some peaks and dips, due to reflections in the guide, that will probably go away by 20% off axis so I would check that before trying to fix any of it. If you do fix say a peak on axis it could become a dip off axis. Looking at the tweeter graph the 6.6 kHz dip will probably not be there 20% off axis. The one at 15 kHz may disappear or turn into a large notch. Either way not much you can do and personally it's out of my hearing range.

                          Dave
                          http://www.pellegreneacoustics.com/

                          Trench Seam Method for MDF
                          https://picasaweb.google.com/101632266659473725850

                          Comment


                          • #14
                            Re: PCD7 help? My project has stalled. SEOS12

                            Good call. I didn't think about the waveguides off axis being different tan a typical driver. I usually take off axis measurements anyway to double check things.

                            Comment


                            • #15
                              Re: PCD7 help? My project has stalled. SEOS12

                              Thanks. See, I am learning a boat load already.
                              I was going to take off axis measurements when I was playing around, but I changed from the speaker sitting on a swiveling office chair, to a small work bench. So now I have to get a lazy suzan. I might be able to turn the speaker, though. Not a big deal. I wanted to do off axis once I got better at measuring...

                              I can only combine 3 response measurments on holmimpulse. If I want to do more, I will have to use MSpaint or something :D. I think REW might let me do more. One more program to learn.

                              Will set up the FR measuring stuff right now...

                              Comment

                              Working...
                              X