Announcement

Collapse
No announcement yet.

GANDI Design Build Thread

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

  • donradick
    started a topic GANDI Design Build Thread

    GANDI Design Build Thread

    Hope I'm not violating the terms of use for Tech Talk.
    The driver is NLA, and I'm not posting any links.

    A few weeks, ago another distributor had a closeout sale of Gradient Acoustics drivers of various sizes.
    After a bit of Googling, it appeared that these were well-regarded, if not really great, so I picked a quad of the 4 inch
    mid-woofers in kevlar/paper - the AXT-04

    Another user here suggested a TM with the ND20 and I committed to the design.
    I may use a pair of ND25 tweeters I have, or I may decide to use a better tweeter.
    Early days though, I'll make that decision after I've listened to them a bit.

    Here is the manufacturers info:
    Click image for larger version

Name:	AXT-04 summarysmaller.png
Views:	1
Size:	112.5 KB
ID:	1346546 Click image for larger version

Name:	AXT-04-FR and materials--smaller.png
Views:	1
Size:	115.6 KB
ID:	1346547 Click image for larger version

Name:	AXT-04 mechanical and TS-smaller.png
Views:	1
Size:	75.4 KB
ID:	1346548
    Last edited by donradick; 09-26-2017, 05:57 PM. Reason: added tag

  • donradick
    replied
    Originally posted by rpb View Post

    Where's the 10k peak shown by the manufacturer? Do you suppose some diffraction effect eliminated it, or did you possibly measure a little off axis?

    It would be interesting to see a waterfall on this driver too. Are there any surprises 45 degrees off axis?
    OK, as requested, FR responses for both samples (on axis 24 inches and 39 inches), and wavelets for both. I'm somehow not inspired to do off axis at this time.

    Click image for larger version

Name:	AXT-04-sample-1-far-field-25-inches-FR.PNG
Views:	1
Size:	71.6 KB
ID:	1351773 Click image for larger version

Name:	AXT-04-sample-2-far-field-39-inches-FR.PNG
Views:	1
Size:	79.1 KB
ID:	1351774

    now wavelets for both - pretty darn good.

    Click image for larger version

Name:	AXT-04-sample-1-far-field-25-inches-wavelet.PNG
Views:	0
Size:	0
ID:	1351781 Click image for larger version

Name:	AXT-04-sample-2-far-field-39-inches-wavelet.PNG
Views:	0
Size:	0
ID:	1351782
    Last edited by donradick; 11-06-2017, 05:27 PM. Reason: fix graphics

    Leave a comment:


  • rpb
    replied
    Originally posted by donradick View Post
    Tested the AXT-04 in a proper box with a port.

    Darn guy is totally flat even without a crossover.
    The full range devotees would swoon over this response.

    [ATTACH=CONFIG]n1351073[/ATTACH]
    Where's the 10k peak shown by the manufacturer? Do you suppose some diffraction effect eliminated it, or did you possibly measure a little off axis?

    It would be interesting to see a waterfall on this driver too. Are there any surprises 45 degrees off axis?

    Leave a comment:


  • ceiol
    replied
    that tymphany BC25SC06-04 looks like a great performer -- especially considering the $14 price tag!

    https://www.parts-express.com/peerle...eter--264-1028

    Leave a comment:


  • donradick
    replied
    Originally posted by rpb View Post
    Do you have a waterfall plot for the BC25SC04-6?
    As requested - looks very, very nice. Maybe that's one reason why I continue to like and use this tweeter.

    Click image for larger version

Name:	BC25SC06-4-wavelet.PNG
Views:	1
Size:	99.4 KB
ID:	1351561

    Leave a comment:


  • donradick
    replied
    Originally posted by Millstonemike View Post

    I think they made PVC for speaker builders and then discovered it could be used for drainage

    Thought you might want to you might want to use more of the ATX-04 stellar upper band and a ring would match it's polar response higher up and be clean beyond audible (just testing my learning here).
    I think you are correct, the upper end of the AXT-04 is incredibly smooth, and would easily work with a high crossover to an XT25 for instance.
    I used to be intrigued by tweeters that were flat to 30K. Now I'm much more concerned with 2--5K. If I can get that right, I feel like
    I've done a good job.

    Leave a comment:


  • Millstonemike
    replied
    Originally posted by donradick View Post

    test setup, easily change port length.

    re: ring radiator

    Hunh? not in this design.
    I think they made PVC for speaker builders and then discovered it could be used for drainage

    Thought you might want to you might want to use more of the ATX-04 stellar upper band and a ring would match it's polar response higher up and be clean beyond audible (just testing my learning here).

    Leave a comment:


  • donradick
    replied
    Originally posted by Millstonemike View Post

    That looks nice. I almost bought a pair the morning I read the post about the closeout sale.

    Are you going to HP the TM. I could never find a power spec. I assumed 30 W. At that they breach under 150 Hz and under 50 Hz at half that power (greater than double xmax). A 100 Hz 1st order HP seems to take care of that fairly well at 30 W.

    A ring radiator?
    I've tested them briefly with a selection of pop music at reasonable volume and they have not shown any distress.
    These will be for desktop use quasi near field with boundary reinforcement, so 80 dB, maybe 85 dB peaks should be fine.

    What I will do is to test real world behavior with various music when tuned higher or lower.
    The Unibox model shows the best bass extension / xmax protection when the box is tuned very low,
    but I"m going to verify that empirically.

    test setup, easily change port length.

    Click image for larger version

Name:	IMG_0968.JPG
Views:	1
Size:	200.4 KB
ID:	1351114

    re: ring radiator

    Hunh? not in this design.

    Leave a comment:


  • Millstonemike
    replied
    Originally posted by donradick View Post
    Tested the AXT-04 in a proper box with a port.

    Darn guy is totally flat even without a crossover.
    That looks nice. I almost bought a pair the morning I read the post about the closeout sale.

    Are you going to HP the TM. I could never find a power spec. I assumed 30 W. At that they breach under 150 Hz and under 50 Hz at half that power (greater than double xmax). A 100 Hz 1st order HP seems to take care of that fairly well at 30 W.

    A ring radiator?

    Leave a comment:


  • donradick
    replied
    Originally posted by ceiol View Post
    Looks great. Where would you plan the tweeter crossover?
    I'd like to do it about 2.8 or 3K. Gotta look at polar response and try to match the drivers.
    I could cross the tweeter lower if needed (3rd order, natch)

    Leave a comment:


  • ceiol
    replied
    Looks great. Where would you plan the tweeter crossover?

    Leave a comment:


  • donradick
    replied
    Tested the AXT-04 in a proper box with a port.

    Darn guy is totally flat even without a crossover.
    The full range devotees would swoon over this response.

    Click image for larger version

Name:	axt-04-alone-21-inches.PNG
Views:	1
Size:	83.2 KB
ID:	1351073

    Leave a comment:


  • donradick
    replied
    Click image for larger version

Name:	front view w separate baffle.PNG
Views:	1
Size:	40.2 KB
ID:	1351069 Click image for larger version

Name:	1st slot port layer front side detail.PNG
Views:	1
Size:	26.2 KB
ID:	1351070 Click image for larger version

Name:	2nd slot port layer 5.5 inches long front side detail.PNG
Views:	1
Size:	13.5 KB
ID:	1351071 Did some box modeling - this may change a bit, but for my limited Sketchup skills, it's workable.

    Leave a comment:


  • donradick
    replied
    Originally posted by rpb View Post
    Please do, I like this tweeter, but maybe not as much as some that I own that are more expensive. I'd like to see if anything in the waterfall stands out.

    Ok, I'll remount a BC25. The wavelet for it should be much better now that my gating is correct.

    Leave a comment:


  • donradick
    replied
    Originally posted by Wolf View Post
    ....I also see your gate is set after the first reflection, so that will also list more CSD/waterfall gremlins into your graphs.

    Later,
    Wolf
    Thanks Wolf -

    you were correct, and that mistake affected the Wavelet graphs more than the other measurements.
    Here's the corrected Neo3 for comparison. Much better than the first test.

    Click image for larger version

Name:	Neo3-Wavelet-2nd-test.PNG
Views:	1
Size:	108.7 KB
ID:	1351066

    Leave a comment:

Working...
X