Home Forums DIY Speakers and Subwoofers Lii Audio Crystal 10 build

  • Lii Audio Crystal 10 build

    Posted by ajc9988 on December 16, 2021 at 9:17 pm

    So, I am finalizing the computer simulations for my own builds, but I will be doing simulations and building a Crystal 10 set of towers for my friend. Construction of all three sets will start in the first week or two of January.

    This is how they were shipped and what they look like.

    I’ll be uploading the DATS information later today, along with a frequency sweep while unboxed using DATS to send the signal for the sweep.

    If anyone is interested in these, they are cheapest if bought on AliExpress over Lii Audio, although there are US distributors if you want to buy the finished product.



    Lii Audio has already removed the crystal-10 from their website. This is what the frequency response looks like on the Crystal 10 and their replacement:

    (silver 10)


    I think I can see why the Crystal-10 was removed from their website. So, there is a good chance that when the existing stock of Crystal-10 is out, they are done. All she wrote.


    Decware is a US distributor for Lii Audio. Finished Crystal-10 reference speakers are about $2,800 there.

    Now, I won’t be doing a curved design, but I am looking forward to building these. DATS and Sweep data soon to follow.

    Also, modeling off of spec sheet puts it at 14W to X-max with no insulation, 18W to X-max with insulation.

    Unfortunately, they removed the old designs for this speaker off of their website. I don’t need it, but my design is not far off from it either (meaning it was a decent design work for it). Maybe the wayback machine. . .

    123toid replied 8 months ago 3 Members · 6 Replies
  • 6 Replies
  • ajc9988

    December 17, 2021 at 12:36 am

    DATS linearity test, set from high to low, 10dB start, 2dB steps, 15 steps.

    Passes the rub and buzz test at the tightest settings without ever tripping it.

    For the T/S parameters.

    Now, does anyone know with the surround inverted if this effects where you measure the diameter from? @123Toid @TVOR-Ceasar

  • tvor-ceasar

    December 17, 2021 at 5:13 am

    From a mechanical standpoint, it works the same way, normal or inverted. As long as the edges of the surround are attached in the same spots as they would be normally, then no change in measurement.

  • ajc9988

    December 17, 2021 at 4:05 pm

    Update on specs: the Xmax is 10mm. For power, it is unclear. What they have printed on the driver is 50-90W. That doesn’t give much to go on.

    For the crossover, it has an L1 of 0.19mH, an L2 of 0.37mH, and an R2 of 2.87 Ohms. That is what was supplied for the spec sheet. I do wonder if there are screws under the label for the backside of the driver where, if one were so inclined, they could go in and swap out the components. But, these are not mine and I will not risk destructive investigation for curiosity sake on a discontinued driver whose replacement, if not prompt, could be problematic.

    Today, I plan to run DATS multiple times to generate averages, from which I can get an idea of exactly where the driver lands. Then likely a couple frequency sweeps and maybe plugging an amp into these to see where that bass is on the driver.

    Either way, modeling puts it at just under 115dB at 50W, meaning it’s got some power. So if it does need DSP’d down for the high end to match sensitivity of the low end, it still would hit 112dB, if the spec sheet is correct. I’ll find out soon enough.

    Note: the FR seen will not be in test conditions for a great reading, I just need to understand the trend lines and if the 3dB down under 200Hz is roughly accurate, which no reading is accurate under 200Hz without specific testing conditions (like high up outside, no cement around, removing reflections, while also looking for a low wind day, etc.; or build an anechoic or semi-anechoic chamber; or a Klippel near-field scanner). So this will not be that, but instead just to get an idea of what I am looking at.

  • ajc9988

    December 17, 2021 at 9:07 pm

    Found under the Edit tab “open the driver editor”, then this jewel popped up with information that isn’t seen along the side. Figured I’d let people know.

  • ajc9988

    December 17, 2021 at 9:56 pm

    So, driver 2 with variance from driver 1. They lie on the frequency response (even if it may do better in a box, there is a sharp drop between 2,000 and 3,000, then the depression from 40 to 700). This is 1/12 smoothing.
    So, maybe their new flagship is more honest on performance.

  • 123toid

    December 18, 2021 at 4:17 am

    @ajc9988 It does not affect it.