Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 31

Thread: What measurements matter to you?

  1. #21
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    No idea, I've always rolled my own. I would think two-channel would be pretty standard at least. If you could run multiple instances of a particular acquisition program, and used routing software to feed each instance with the right pair of channels, then you could probably do it. Issue might be syncing, which could be handled with an external trigger.

    BTW - I've used these MC boxes (not this one, but similar http://www.mccdaq.com/usb-data-acqui.../USB-7204.aspx). Multichannel input, also output, and digital control lines which I can imagine could come in handy. This one is a little pricey, perhaps, and I'm not sure whether I like it powered off a brick instead of the bus, but peruse their site. There's some basic o-scope software that comes with it... they may have improved upon it over the years though, based on their description.
    Last edited by MarkZ; 09-22-2011 at 04:40 PM.

  2. #22
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    With soundeasy you have to input in to the left channel. the right is always used as a reference. the only way I could see using multi-mics actually working is to swap the xlr cable in to the pre-amp yourself each time. I don't see how you could do it any other way unless you got really crazy. IMO, it's a waste. If you have the funds to buy extra mics and stands and cables, then it might be easy on you and you'd never have to worry about the mics' positions changing between measurements... but that's the only advantage I see. I made a DIY "polar jig" with a piece of scrap birch and some PVC pipe.

    My setup was like this: computer sound card - rca out - rca in to behringer mixer with xlr phantom power (iow: mic preamp) - xlr out - mic


    With HD testing you have to use both channels of the soundcard (left & right) and you do this by splitting the mic preamp's rca output.


    *note: your soundcard may not use rca in/out. sub xlr if it uses that instead*


    Edit: Yea, if you used a card with multi-ins, it might work. I just don't know how SE will recognize it because you have to use the left channel as mic in and right channel as reference (probe).
    Your ears: The best tools you have... and they're free, too!

  3. #23
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    Quote Originally Posted by MarkZ View Post
    Fair enough, but I don't think 5 degree increments would be necessary at all. I was thinking more along the lines of ~22.5 (I'd actually probably do a logarithmic scale with ~1.2 multiplier, off the top of my head). The point is to get an accurate picture of the off-axis response profile, which I think can be difficult to reconstruct with only two points.

    .
    With anything less than 15*, I'd fear you may miss out on something. IMO, if you're going to do polars, you need a lot of data sets to do it with. I wouldn't do less than 15* but would hope for 5* for greater accuracy.

    All in all, I think a simple 2-d 0, 15, 30, 45, 60* measurement plot will suffice. Of course, that gets ugly quick so 0, 30 and maybe 60* would be best (I think this is what I settled on).
    Your ears: The best tools you have... and they're free, too!

  4. #24
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    You really think there's fine structure in the polar plots that would be essentially aliased by using larger increments? What are you basing this on? Is there some data that you're referencing? I would have expected polar response to vary much more smoothly than that. If you're correct that it doesn't, then doesn't that mean that the two-point directivity data is even more worthless than even I thought?

    FWIW, your list of values isn't terribly far off from what I'd choose. So are we agreeing, or what?

  5. #25
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    I'm recalling the few polars I've seen in the past where there's a suckout in the response. I believe that if you weren't doing fine increments, these issues might not show up. I really do feel that for accuracy you'd want 5*... and no less than 15*. Maybe 10* is something we could settle on.

    The big thing for me though is the time vs worth.
    Your ears: The best tools you have... and they're free, too!

  6. #26
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    back in...

    I've been giving it some thought and talked to the Mrs. about this yesterday. I'm going to build me a new portable test baffle and start doing FR, HD, and maybe CSD testing again.
    I'm also going to change things up a bit with the hardware/software. I'm thinking about going back to the Dayton OmniMic setup. I had some serious issues with this the first go round and sent it back. I felt for accuracy, SoundEasy was best. I've even bumped a thread on tech-talk regarding this and have gotten some replies that have helped qualm some issues I had initially; it appears that they have either fixed some things with the setup or I just had a bad mic. My concern lies with the HD portion of the system and it's non-stability due to the frequency of the sweeps. They simply didn't give the environment time to reach stasis and therefore the results fluctuated. That could have been an issue with the mic, according the system developer, who has been sharing PM's with me a good deal to resolve my issue. I've got a way to overcome this, though, if it is true and I'll share it when I get it finished.

    The plan for the baffle is a reduced sized baffle and I'll be doing nearfield/farfield merging from here out. That alleviates any issues in the lower end and therefore allows the baffle to both be smaller and the room treatments to be minimized. Rather than an 8ftx6ft baffle, I'll be chopping it down to 4x4ft, or thereabouts. It will be on 4" casters, braced with 4x4's and will be stored in the attic space while not being used. This frees up a ton of space and allows me to do testing like I want.

    I need to sell off my soundeasy gear first but I hope to start testing again by mid-October. If Adam and I can tag-team this, we should have some great data to present you guys in the future. Supplemented, of course, by the Klippel data I'll be posting on DIY. (Adam, we'll have to work out details of this).

    The OM setup does not allow polars, however. So, if you guys really want that done, you'll have to talk to Adam about this.

    - Erin
    Your ears: The best tools you have... and they're free, too!

  7. #27
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    I'm glad that you guys are going to work in parallel. That's the gist I got from your post. Hopefully there's very little redundancy between your setups, so that you can each offer something unique. Am I too far off?

  8. #28
    Senior Member cvjoint's Avatar
    Join Date
    Mar 2011
    Location
    Northern California
    Posts
    801
    Real Name
    George
    Can anyone of you get the GedLee metric? I presume it involves an approximation of the nonlinear behaviour with a functional form. From that certain parameters are plugged into the GedLee equation.

  9. #29
    Devil's Advocate Adam_MSS's Avatar
    Join Date
    Aug 2010
    Posts
    1,578
    Real Name
    Adam
    I think there are probably two ways to leverage the two setups:

    1) do different testing and exchange drivers so each gets the full course

    2) do the same testing and test a greater number of drivers (we would exchange a couple sets for calibration purposes)


    Erin's looking to shrink his baffle which is probably a practical choice, and I think it's the wise choice. I'm not quite there yet but am open to being convinced. All I'd need to see is a comparison test of the same speaker on the two different size baffles to verify that we don't lose much. I think that exercise would serve not only to convince me but also to lend credibility to the whole exercise.

    I'm probably a couple months out from starting as I am moving in the next 30-60 days.
    You don't use science to show that you're right, you use science to become right. - R.Munroe

    The important thing in science is not so much to obtain new facts as to discover new ways of thinking about them. - W.L.Bragg



  10. #30
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    New setup pictures
    Your ears: The best tools you have... and they're free, too!

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •