J-Box III Review

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Reet
    Senior Member
    • Dec 2007
    • 816

    #1

    J-Box III Review

    The J-Box III Review

    The J-box is a device that I’ve often looked at on AliExpress, but I don’t see much or any talk about it really on DIY forums, not English speaking ones anyway. It looked quite interesting to me as an all-in-one measurement system; the poor man’s alternative to a Clio Pocket.

    Thanks to tktran for the donation to help make this review possible.

    The J-box is effectively a USB audio interface with a built-in dual channel configuration for measuring impedance as well as providing a timing reference for acoustic measurements. It provides a small built-in amplifier as well, so it can avoid a lot of the clutter of wiring up a USB audio interface to a measurement jig to a power amp. It’s perfectly functional for any measurement system like REW or ARTA, and is plug and play on both Windows or Linux systems (probably Mac too, but I don’t have one to confirm).

    What’s in the box?
    Click image for larger version  Name:	image.png Views:	1 Size:	993.3 KB ID:	957532
    The J-box is available in 3 different bundles. One includes just the device and power supply, not case or microphone. It can also be purchased with a case, and with a case and calibrated electret microphone. For testing, I have the full kit with case and microphone.

    Click image for larger version  Name:	image.png Views:	1 Size:	473.4 KB ID:	957533
    It comes with a reference resistor, mine was labeled as “9.976 ohm”, I don’t have an accurate enough meter to verify that, my Fluke 179 reads 10.0. The power supply unfortunately has a very short cable, only 3 ft, so you may find yourself wanting to take on a DIY project to extend the cable length. It also comes with a RCA cable for the mic about 3m long which is decent, as well as some alligator clip leads for impedance measurements, and a USB A to B cable. There is also a USB stick, which contains nothing more than the mic calibration file.

    Click image for larger version  Name:	image.png Views:	1 Size:	539.2 KB ID:	957534
    Click image for larger version  Name:	image.png Views:	1 Size:	570.6 KB ID:	957535
    The microphone came with a serial number hand written on the foam insert in it’s case, and the serial number is hand etched on the mic body. The mic itself is a simple electret mic capsule inserted in a brass tube. It came with a 3D printed camera mount. Unfortunately I don’t have a camera mount, so I 3D printed a mount to fit a 20mm mic clip.

    The I/O
    Click image for larger version  Name:	image.png Views:	1 Size:	547.5 KB ID:	957536
    Click image for larger version  Name:	image.png Views:	1 Size:	573.3 KB ID:	957537
    For audio output it includes a selector switch for 1.41V, 2.83V, or 4.9V. Output can be connected directly to a speaker via the "PORBE" binding posts, or used as a line level output via the “aux” RCA connector. The Aux output is labelled “0.5x” and will be precisely that, 0.5 x the output on the speaker binding posts, so maximum output on the aux connector can be 0.5x 4.9V = 2.45VRMS max line out.

    For input there is an RCA mic input and a XLR mic input, and an adjustable input gain control. Note that the RCA mic input is powered for ~6VDC for electret microphones, and the XLR/TRS combo input is powered with 48VDC phantom power for condenser mics. Unfortunately there is no switch to turn off the phantom power, and it is powered on both the XLR and TRS pins, so this is a “condenser mic only” input.

    The J-box included no documentation whatsoever, however there are fairly clear diagrams on the AliExpress sales pages for the device.
    Click image for larger version  Name:	image.png Views:	1 Size:	246.8 KB ID:	957538
    The “cal” selection is simply a loopback connection on both channels, for “soundcard calibration” in REW. The “Z” selection operates the same as my dual channel measurement jig, but with a 12 ohm sense resistor. The SPL selection uses the Left channel as the mic input, and the right channel as the feedback loop, for a dual channel measurement. Unfortunately, because the loopback is built-in, there is not an opportunity to inject things like a high pass filter capacitor in the feedback loop, but it’s still great to have a good timing reference with no additional effort required.

    What’s Inside?
    Click image for larger version  Name:	image.png Views:	1 Size:	1.22 MB ID:	957539
    Inside the J-box there are two TDA2050A amplifier ICs, which I found curious as this is a mono device. After staring at the circuit board for a few minutes, I determined that the second TDA2050A chip is being used to generate the split voltage rails. The J-Box receives 24VDC from the external power supply, which it then splits into +/-12VDC rails using a TDA2050A. Like this:
    Click image for larger version  Name:	image.png Views:	1 Size:	6.6 KB ID:	957540
    The second TDA2050A is the audio amplifier. A trim pot next to it can be used to fine adjust the gain setting. The rightmost trimpot adjusts the aux level output. 4 more trimpots to grouped together at the bottom for the voltage dividers for the loopback operation. At the mic input, there is an NE5532 and a OP1662. I believe the NE5532 is for the RCA mic input, and the OP1662 is for the XLR input. The USB and audio functions are all built into a single SSS1701B “USB Headset / Line-In Controller”. This chip combines the USB connectivity as well as DAC and ADC functions. I found it’s datasheet here for those interested:
    SSS1700B1_USB-Headset-Line-in-Controller-Datasheet_v1.pdf

    Attached Files
    Last edited by Reet; 18 May 2025, 13:02 Sunday.
    I'm not deaf, I'm just not listening!
  • Reet
    Senior Member
    • Dec 2007
    • 816

    #2
    Does it work?
    Yes, it works! It’s not perfect, but it has certainly met my expectations for performance. For comparison, I used my own test set-up which consists of a Motu M4, and included a basic class-AB amplifier with a linear power supply (Audiotrak AT300).

    The following tests were completed with the J-Box in “cal” operation, with no load connected to the amp. I found that with 0dBFS output, the loopback signal was around -5dBFS at the input. For an equal comparison, I set the output of my Motu M4+amp to 2.83V at the amplifier terminals, adjusted the input gain for -5dBFS at the input, using my dual channel jig for the feedback loop.

    For a quick and easy comparison, I used Rightmark Audio Analyzer. It complained that the -5dBFS input level was too low, but continued anyway, knowing that all my measurements would be at the same input level.
    Click image for larger version  Name:	image.png Views:	0 Size:	29.1 KB ID:	957545

    Voltage level
    Unfortunately, I found that out of the box, the voltage level was a fair bit lower than expected. I was getting 2.36V when I expected 2.83V. I connected my multimeter and adjusted the trim pot next to the amplifier IC to achieve 2.83V using REW’s generator, 60Hz sine wave at 0dBFS output. Aux level tested right where it should be. At least, it was an easy solution requiring no soldering.

    Frequency Response
    The frequency response of the device is very flat, however it includes a brick wall low pass filter at 20kHz. Despite it’s ability to operate at 96kHz sample rate, there is really no point in using this device beyond 48kHz. There are some small wiggles at high frequencies, however if you run a "soundcard" loopback calibration, the frequency response of the device can be completely compensated for.
    Click image for larger version  Name:	FR Spectrum.png Views:	0 Size:	10.3 KB ID:	957544

    Noise floor
    The noise floor can be described as “okay” or “good enough”, but it’s not state of the art by any means. Despite the fact that the device uses a switch mode power supply, there is some low level mains hum coming through. A higher quality power supply may help with that. Motu shows the same mains hum, from the linear power supply in my amplifier. If I switch to a different amp with a SMPS, the mains hum will disappear.
    Click image for larger version  Name:	noise floor.png Views:	0 Size:	20.7 KB ID:	957546

    Distortion
    Distortion is “not great”. If we consider that this device’s primary purpose would be for impedance and frequency response measurements, we can disregard it’s distortion capabilities, and conclude that this device is not suitable for detailed distortion evaluation.
    Click image for larger version  Name:	THD 1kHz.png Views:	0 Size:	20.2 KB ID:	957547
    Click image for larger version  Name:	THD multi.png Views:	0 Size:	9.7 KB ID:	957548

    The distortion level of this device is confusing. I completed additional testing in REW in addition to the RMAA test. It shows a high 3rd order distortion level, very consistent across the entire frequency range, so I'm not sure about the results that RMAA provides, but they were consistent if I re-ran the test.

    REW 0dBFS loopback:
    Click image for larger version  Name:	image.png Views:	0 Size:	131.1 KB ID:	957549
    Click image for larger version  Name:	image.png Views:	0 Size:	108.5 KB ID:	957550


    I'm not deaf, I'm just not listening!

    Comment

    • Reet
      Senior Member
      • Dec 2007
      • 816

      #3
      Mic response
      I compared the included electret mic to a Line Audio Omni1 microhpone.

      The J-Box microphone includes a calibration file. Inspection of the mic calibration shows it is compensating for the high frequency response, and appears to be a valid calibration. It is not a noisy mess like you find in Dayton EMM-6 calibrations, this is a perfectly usable calibration file as-is.
      Click image for larger version  Name:	mic cal file.png Views:	107 Size:	29.7 KB ID:	957552

      Frequency Response
      For as much of an "apples to apples" comparison as possible, I 3D printed a mic clip adapter for the J-Box mic that was similarly shaped to the Line Audio Omni1, so body shape, and placement in the mic stand and clip will be equal.
      Click image for larger version  Name:	image.png Views:	107 Size:	744.6 KB ID:	957553

      I placed the mic in front of a speaker, at 315mm distance and compared the measured frequency response. Measurements were aligned at 1kHz.
      Click image for larger version  Name:	image.png Views:	107 Size:	44.3 KB ID:	957554

      The J-box microphone is not wildly different from the Omni1, however the Omni1 is measuring a slightly higher level in the top end. There is a maximum of 2dB difference between the two measurements.

      The J-Box mic perhaps was not calibrated with this mic clip arrangement in mind, so I compared again, with the mic clip moved to the back of the mic tube. Minor differences in this arrangement are primarily caused by the change in location of the mic clip and stand.
      Click image for larger version  Name:	image.png Views:	108 Size:	46.4 KB ID:	957555

      I would consider this mic to be "good enough" for most people for speaker frequency response measurements.

      Mic Distortion
      Unfortunately, the included electret condenser mic is not well suited for distortion measurements. It includes a high 2nd order distortion product, as well the limitations of the low power built-in amp prevent the device from being useful for distortion evaluation of high end speaker drivers. For detailed distortion evaluation, I recommend a good USB audio interface like Motu M2/M4, connected to a good quality amplifier and a microphone like Line Audio Omni1.

      I tested a high end speaker at 315mm, using the 4.9V output. Calibrated the mic SPL level, and adjusted the gain of both the M4 and Jbox to have 120dB of headroom. Actual SPL at the mic for this test is in the 96-97dB range.

      J-Box:
      Click image for larger version  Name:	image.png Views:	0 Size:	144.3 KB ID:	957588

      With the M4 and Omni1, ​3rd order distortion has suddenly dropped 10dB from 600Hz and up. Noise floor is lower, and then there's H2.
      Click image for larger version  Name:	image.png Views:	0 Size:	131.3 KB ID:	957587
      Last edited by Reet; 13 April 2025, 14:54 Sunday.
      I'm not deaf, I'm just not listening!

      Comment

      • Reet
        Senior Member
        • Dec 2007
        • 816

        #4
        Impedance
        For evaluation of the impedance capabilities, I tested a complete speaker system at 250mV level, or -21.08dBFS level with the output selector set to 2.83V. I compared to my own impedance measurement jig. The only real difference between the two is that the J-box uses a 12 ohm sense resistor, while my own jig uses 10 ohm.

        For this speaker system, the odd wiggle in the 1-2kHz range is expected. You can see that the two measurements are very comparable, making the J-Box perfectly well suited for impedance measurements.
        Click image for larger version  Name:	image.png Views:	22 Size:	46.2 KB ID:	957556

        Conclusion
        For a very reasonable cost, you get a full functioning impedance and acoustic measurement system with timing reference. Perfect for use to gather data for CTA-2034 based crossover design using VituixCAD, with no external amplifier required. I would definitely recommend the J-Box III.

        If there was a J-box 4, what I would change is a longer power cord, 48V phantom power switch, and disconnect phantom power from the TRS input so it can be used as a general line-in. I’d also change the USB audio codec for something that fully supports output beyond 20kHz.


        Last edited by Reet; 12 April 2025, 16:16 Saturday.
        I'm not deaf, I'm just not listening!

        Comment


        • JonMarsh
          JonMarsh commented
          Editing a comment
          A lot of useful effort into the review- gives a clear idea of what it's capable of- I started at the last post and went back to the beginning, and I was thinking along the idea of my Pocket CLIO and how would this compare... as a poor man's version, how much do you save?

        • Reet
          Reet commented
          Editing a comment
          JonMarsh, the regular price for this kit to me is CAD$204 with free shipping. I checked the price with shipping for a Clio Pocket from SoundImports, total cost CAD$958!
      • a4eaudio
        Junior Member
        • Nov 2023
        • 4

        #5
        Originally posted by Reet
        If there was a J-box 4, what I would change is...
        and "PORBE" to "PROBE" 😁

        Comment

        • Reet
          Senior Member
          • Dec 2007
          • 816

          #6
          Nah, I think we should amend the English language, and we all start saying porbe instead, it's fun to say
          I'm not deaf, I'm just not listening!

          Comment

          • theSven
            Master of None
            • Jan 2014
            • 1656

            #7
            Reet Thank you for writing up this detailed review for the members here on HT Guide! That is a considerable amount of savings, $240 vs $958 CAD or $147.06 vs $690.67 USD. Where did you order your Line Audio Omni1 from?


            Painter in training

            Comment

            • Reet
              Senior Member
              • Dec 2007
              • 816

              #8
              I purchased from the Canadian dealer. https://lineaudio.se/contact.html
              I'm not deaf, I'm just not listening!

              Comment

              • Reet
                Senior Member
                • Dec 2007
                • 816

                #9
                Comparing to a Clio pocket may not be entirely fair, the Clio appears to include much more capable hardware, and you are also paying for the proprietary software development.

                On an equal price scale, the basic J-box kit without carrying case and mic sells for the same price as a Dayton DATS, so you could look at it as a better value alternative to a DATS + other hardware for acoustic measurements.

                One thing to consider with any of these cost/value comparisons, is that I have been unable to find any trace of "Song007.SK" as a company or person, so the J-box is an "as-is" device without any warranty or support. Support path is internet forums and your own DIY efforts.
                I'm not deaf, I'm just not listening!

                Comment


                • JonMarsh
                  JonMarsh commented
                  Editing a comment
                  Good information for folks to have
              • Reet
                Senior Member
                • Dec 2007
                • 816

                #10
                Updated with a few images detailing the mic distortion.
                I'm not deaf, I'm just not listening!

                Comment

                • Reet
                  Senior Member
                  • Dec 2007
                  • 816

                  #11
                  I'd prefer both of these mics to match, so I created my own calibration file. Not too bad for a mic swap comparison.
                  Click image for larger version  Name:	image.png Views:	0 Size:	40.3 KB ID:	957591

                  Comparison of the calibration file the mic came with (solid), vs the one I created (dashed).
                  Click image for larger version

Name:	image.png
Views:	863
Size:	30.4 KB
ID:	957592
                  I'm not deaf, I'm just not listening!

                  Comment

                  • letiennam
                    Member
                    • Mar 2016
                    • 55

                    #12
                    Thanks for your detailed review. I have discussed and also bought this product from Song007, who developed this jbox from hifidiy.net forum (Chinese diy audio forum). He uses Wechat, a Chinese chat software developed for themselves. I can get his email if you need support. Song007 - he is quite open and willing to guide me from the first steps in measurement. This product is sold in China mainly and he is an individual so he only has Chinese manuals for Lspcad and Rew installation instructions. I can send them to you if you need.
                    An interesting thing is that because I am closer (I am in Vietnam), I only spent about 70$ for the whole product like you. One problem I often encounter is that Jbox often measures the Fs of the unit a few dB lower than the Dats V3.

                    Comment

                    • Reet
                      Senior Member
                      • Dec 2007
                      • 816

                      #13
                      Thanks for the info letiennam. If you have contact information / email address for Song007 I would be grateful for it so I could contact him. I don't need support per se, but perhaps I can provide some assistance for a J-Box IV, if he decides to continue to improve the product. It's a great piece of kit, Dayton should be taking notes!
                      I'm not deaf, I'm just not listening!

                      Comment

                      • letiennam
                        Member
                        • Mar 2016
                        • 55

                        #14
                        Originally posted by Reet
                        Thanks for the info letiennam. If you have contact information / email address for Song007 I would be grateful for it so I could contact him. I don't need support per se, but perhaps I can provide some assistance for a J-Box IV, if he decides to continue to improve the product. It's a great piece of kit, Dayton should be taking notes!
                        I sent you a PM. Please check

                        Comment

                        • Reet
                          Senior Member
                          • Dec 2007
                          • 816

                          #15
                          Attached the manual to the first post. It's in Chinese, so feel free to run it through your favourite translation tool to review. As mentioned the device is pretty straight forward to use if you can understand the block diagrams provided. Main item found in the manual is the input impedance is only 5kOhm, for setting Rinput in REW for impedance measurement.
                          I'm not deaf, I'm just not listening!

                          Comment

                          • Reet
                            Senior Member
                            • Dec 2007
                            • 816

                            #16
                            Something identified from some further investigation, due to an inquiry at DIYAudio forum. I believe that the primary limiting factor in distortion analysis for the J-box appears to be a relatively high level of cross-talk between the audio output and mic input.

                            Below image shows a comparison of noise floor and fundamental measurement in REW, with no speaker connected, so the mic is simply measuring background noise during the measurement. The only difference between the "fundamental" and "noise floor" below is that the Jbox is playing back a sine sweep during the fundamental portion. If there is no cross-talk, these traces should overlay with no difference between them, however we see a significant increase in signal level in the fundamental measurement. This will be a limiting factor for low level measurements, ie distortion. For SPL measurement, this is not of any concern (hopefully everyone measures at a level >50dBSPL).


                            I'm not deaf, I'm just not listening!

                            Comment

                            Related Topics

                            Collapse

                            • Reet
                              Dealing with DC offset in measurements
                              by Reet
                              DC offset in measurements can be very frustrating, I thought I'd open this topic to determine best practices and ask what the recommended solutions are.

                              Equipment here is a Motu M4 interface, and sine sweep measurement in ARTA. Motu claims "DC coupled TRS input" so it tends to...
                              15 January 2023, 13:16 Sunday
                            • Brian Bunge
                              Mic pre
                              by Brian Bunge
                              I'm interested in getting an outboard USB device to be used for a mic pre and was wondering if you guys have any experience with the M-Audio products. I know one guy that's had good results with the M-Audio Mobilepre device but I also see that there is a device called the Audiophile USB. It has line...
                              11 June 2005, 14:26 Saturday
                            • Chris D
                              My BFD subwoofer EQ thread
                              by Chris D
                              Well, I'm FINALLY getting around to using my BFD to calibrate my two SVS PC-Ultra subwoofers. As expected, I'm running into a few hitches along the way as I'm learning and going through the process. Hopefully there are guys here that can help.

                              It took me a while, but I did go around and...
                              09 March 2006, 15:33 Thursday
                            • Reet
                              Single vs Dual Channel Measurements, or why friends don’t let friend use USB mics
                              by Reet
                              What does single channel or dual channel measurement really mean, and what are the implications of choosing one over the other? This document aims to help answer this.
                              Strictly speaking, a microphone is a single channel of audio input, in that it is a mono signal. USB microphones which include integrated ADC and USB codec can be considered a single channel device.

                              Dual channel would be a measurement which includes a second channel of signal input in addition to the microphone....
                              26 December 2022, 11:23 Monday
                            • marshahu_cr
                              8CH Analogue Input Splitters?
                              by marshahu_cr
                              From my previous post it seems that I have too high expectations to get a receiver with two 8CH analogue inputs for under £400! :cry:

                              Is there a device like an 8CH Analogue input splitter? - the idea being I hook this into the one 8CH input on the receiver and then hook up my two computers,...
                              28 December 2005, 14:09 Wednesday
                            • Loading...
                            • No more items.
                            Working...
                              Searching...Please wait.
                              An unexpected error was returned: 'Your submission could not be processed because you have logged in since the previous page was loaded.

                              Please push the back button and reload the previous window.'
                              An unexpected error was returned: 'Your submission could not be processed because the token has expired.

                              Please push the back button and reload the previous window.'
                              An internal error has occurred and the module cannot be displayed.
                              There are no results that meet this criteria.
                              Search Result for "|||"