Announcement

Collapse
No announcement yet.

CX120-8 easy build

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

  • djg
    replied
    Baffles drying, maybe tomorrow. Excited to hear a coax speaker.

    Click image for larger version

Name:	IMG_0007.JPG
Views:	2473
Size:	419.5 KB
ID:	1453660

    Leave a comment:


  • djg
    commented on 's reply
    BOB, I just ordered 2 from Meniscus, thanks.

  • djg
    commented on 's reply
    Well, the product page says 11/06. Orphan boxes, hope not.

  • BOBinGA
    replied
    I hope they come back in stock. The similar CX150 seems to be discontinued, but Meniscus still has a few. I was planning to use the CX150s for atmos height speakers, but maybe they will be stocked again. If not, I guess I will use SB’s. But that’s a project for next year.

    Leave a comment:


  • djg
    replied
    This project wasn't in my mind until BOBingGA posted the XO. Always wanted to do a coax, so here's the last build this season. I won't do a build thread, just post a few pics here. Prebuilt the innards, the actual driver is back ordered for a month. Some fairly nice 1/2" ply, veneer with a 1/4" MDF baffle on all or part of the front. I plan to get them all built and waiting for the drivers in the next week. This is a pair of ported speakers for a 2.1 setup, the original being a single center channel.

    Click image for larger version  Name:	IMG_0002.JPG Views:	0 Size:	449.9 KB ID:	1453225

    Click image for larger version  Name:	IMG_0003.JPG Views:	41 Size:	866.0 KB ID:	1453243

    PSA mahogany veneer left from last year Zingers build. 1/2" MDF baffle, glued to 1/2" ply front panel, 1" baffle, overkill.

    Click image for larger version  Name:	IMG_0004.JPG Views:	0 Size:	445.9 KB ID:	1453406

    CX120s should be here today.

    Click image for larger version  Name:	IMG_0005.JPG Views:	0 Size:	920.2 KB ID:	1453497

    Click image for larger version

Name:	IMG_0006.JPG
Views:	2444
Size:	950.5 KB
ID:	1453524
    Last edited by djg; 10-07-2020, 06:06 PM.

    Leave a comment:


  • wogg
    replied
    Wow, that's way simpler than I expected! Well done Bob.

    Leave a comment:


  • BOBinGA
    replied
    Click image for larger version

Name:	CX120-8 as built photo.png
Views:	2784
Size:	649.4 KB
ID:	1452888 Here is the final XO:
    Click image for larger version  Name:	CX120-8 as built large.jpg Views:	0 Size:	19.3 KB ID:	1452883
    It has a very slight 1 to 2db upward bias in the 800 to 2000hz range to help with voice intelligibility. Otherwise, its flat.
    C1= 3.3uF
    R1= 3.0 ohms
    L1= 1.2mH
    C2= 4.7uF
    Last edited by BOBinGA; 09-29-2020, 01:45 PM.

    Leave a comment:


  • davidroberts
    replied
    Did you ever post the complete xover schematic?

    Leave a comment:


  • BOBinGA
    replied
    I have been using this as my living room center channel with the PEDS as left and right, PS95s as rears and an 8” sub I have not documented since the driver is long discontinued. It works great for this purpose since my wife and I don’t listen too loud. But the CX120 center does exactly what I need - it makes dialog very easy to understand.

    I have considered using the CX120 in a LCR sound bar. I think that would be good high value build with a subwoofer and an inexpensive 5.1 AVR. But I don’t need this now, so I will leave that to others.

    Leave a comment:


  • brianbunge
    replied
    I too am looking at using this driver for a center channel (and maybe mated to a larger woofer for L/R speakers) except I am in a huge room with 20’ ceilings. So I’d like to build a center shaped more like a sound bar with a quartet of 5.25”-6” woofers using the Dayton Designer series paper cones. Let me know if you have any ideas concerning this.

    Leave a comment:


  • wogg
    replied
    This is fantastic Bob! I have a plan in my head for one of these CX120's with a pair of 4" woofers flanking it for my someday HT build. I'll have to bookmark this for a starting point of the mid to tweeter crossover.

    Leave a comment:


  • BOBinGA
    replied
    Finally for tonight, here are some shots of the crossovers I made using the PE small perforated boards. Even on the small 3" X 5" boards, I mounted both the flat and low BSC crossovers at each end of one board. I didn't have exactly the parts I needed, so I had to create them by paralleling some parts. But the actual values are all standard values that you should be able to source at PE.

    Click image for larger version

Name:	XO layout small.jpg
Views:	2905
Size:	172.2 KB
ID:	1426490

    I will try to add an actual wiring diagram later, but that's all I have time for tonight. Let me know if you have questions.

    Leave a comment:


  • BOBinGA
    replied
    Finally, here are the actual measured response of the finished speaker using the flat and reduced BSC XO circuits. (Disregard the stuff below 70 hz. I left my subwoofer on set to a 50 hz crossover for my regular speakers.)

    First, with 1/3 octave smoothing to show you the response warts ad all.

    Click image for larger version

Name:	CX120-8 low and full BSC 1-3 octave.png
Views:	2876
Size:	18.4 KB
ID:	1426487

    You can see that there are lots of room modes below 600 hz. I could try gating them out in the measurements, but I find it somewhat easier to just mentally average out the peaks and dips to get a response that is closer to what I hear. The peak/dip/peak above 10k hz is a product of the coaxial tweeter and is not really audible.

    To get an idea of any response problems that I can work out in the crossover, I also look at the one octave smoothed response. Its obviously not as accurate, but it gives you a better sense of the octave to octave balance and is pretty indicative of what you actually hear. Here it is:

    Click image for larger version

Name:	CX120-8 one octave smoothing.png
Views:	2843
Size:	17.6 KB
ID:	1426488

    How about them apples! Pretty close to dead flat with only four crossover parts. I don't think you can do any better.

    Leave a comment:


  • BOBinGA
    replied
    You can see the different crossover versions I tried. The increased bass version was a little too bassy and lacked highs. The flat version is what I would use if it will be out from the wall a couple of feet on stands used as a stereo pair. The reduced BSC version works best when used directly below a large TV screen or when placed on a shelf close to a wall.

    To illustrate the difficulties getting to make these work, here is the measured response in my large room. First the woofer:

    Click image for larger version

Name:	woofer response WinPCD.PNG
Views:	2930
Size:	473.4 KB
ID:	1426483

    You can see that the in box, in room uncorrected response has a large hump at 1700 hz followed by a big dip down to 4k hz followed by the cone breakup going back up to 8k hz. Luckily, this was massaged into a pretty close to 2nd order butterworth low pass slope at 2300 hz with just the 1.2 mH inductor and 4.7 mf capacitor.

    Here is the tweeter:

    Click image for larger version

Name:	tweeter response WinPCD.PNG
Views:	2906
Size:	443.5 KB
ID:	1426484

    The unmodified response is about 5db higher than the final woofer response. Again, I was fortunate to get the down to match the woofer with only a 3.3 mf capacitor and a 3.0 ohm resistor. This resulted in a 4th order butterworth high pass at 2800 hz.

    Now finally, here is the predicted system response using the flat crossover:

    Click image for larger version

Name:	Full BSC combined response WinPCD.PNG
Views:	2929
Size:	316.6 KB
ID:	1426485

    You will notice that there is a small dip in the response at 2500 hz. This is because the woofer and tweeter crossover points are different - 2300 and 2800 hz. There was simply no way to get them closer together since these are the limits for the individual drivers. Fortunately, this crossover only creates a 3db dip, which is not even noticeable in use.

    Leave a comment:


  • BOBinGA
    replied

    This was a bit difficult to get working well. I first measured the T/S parameters of the one I received. It was a little off of the published specs, but it still modeled just like the published specs.
    Published Fs = 87.5 hz Qts = 0.48 Vas = 3.0 liters
    Measured sample Fs = 88.8 Qts = 0.53 Vas = 1.71

    Its not unusual for drivers to have higher than spec Qts and lower Vas. Indeed, this was the case, but for this closed box it worked out to the same volume of 0.066 cu ft (114 cu in).

    Click image for larger version

Name:	CX120-8 measured closed and ported.PNG
Views:	3227
Size:	288.4 KB
ID:	1426479

    The Box size was pretty much determined by the height needed to allow the driver to fit the front. Using a regular rectangular box, we get a Qtc of 0.69 with full stuffing. Indeed, measuring the final box with my DATv2, I get a Qtc = 0.71 and an Fb = 118 hz. That's pretty close to predicted.

    The WinISD graph above also includes a ported box 225 cu in with a 1.5"PVC port (1.59" actual diameter) X 7.25" length. You can see this will give you output down to 70 hz, but the size is double.

    Here is my notebook page with the box details:

    Click image for larger version

Name:	IMG_0685 small.jpg
Views:	2987
Size:	180.0 KB
ID:	1426480

    I didn't build the ported version, so that's up to you to try. I will be using the close box with a sub.

    Leave a comment:

Working...
X