Argos - TM and MTM bookshlf speaker design thread

Discussion in 'DIY Speakers and Subwoofers' started by Matt Grant, Dec 7, 2019.

  1. #1 Matt Grant, Dec 7, 2019
    Last edited: Feb 10, 2021
    The Argos design is a compact 5" two way bookshelf using the Faital Pro 5FE120 and the Beston RT002A ribbon tweeter. This small bookshelf speaker is capable of more output then its size would suggest and exhibits excellent dispersion characteristics with a good directivity match at the crossover resulting in a wide sound stage and solid imaging.

    I actually purchased a pair of each driver over a year ago to start working on the design but only got around to it just recently.

    If there is enough interest I will consider making up some PCBs and/or CNC'd baffles for this design. Or it could possibly be turned into a kit on DIYSG.

    Edit: -PCBs here-


    I'd been itching to try a ribbon design for a while and decided to go ahead and start with a simple two way bookshelf design. The only real goal for this design was was good price/performance ratio, as with most of my designs I like to stay with more value oriented drivers so that the final design is accessible to those who are on a budget.

    I started looking for a ribbon tweeter to use and the Beston RT002A caught my attention for it's ability to be crossed fairly low for a ribbon (2500Hz). I also liked the implementation on a shallow waveguide which I suspected would give much better control of horizontal directivity then flat surface mounted ribbons.

    With the tweeter selected I moved onto the woofer. I knew I wanted to stay in the 5 - 5.25" range in order to prevent too much beaming at the crossover. I've had my eye on on the Faital Pro 5FE120 for a number of years as it checks nearly all the boxes, very good sensitivity for a 5" woofer, decent xmax, high power handling, nice flat frequency response free of major cone breakup and has low inductance with a shorting ring in the motor as well. The only downside is it's bass extension is somewhat limited, a trade off of it's high sensitivity and ability the be used in compact enclosures. So no deep bass extension from this driver.

    The result:
    A speaker very impressive for it's size. The high power handling and xmax on the 5FE120 means you will be hard pressed to find it's limits on regular AVR power especially if crossed to a subwoofer. I find the woofer handled itself extremely well when pushed hard, I noticed no modulation of the midrange even under very bass heavy content being pushed to and perhaps a little beyond xmax. These won't wow anyone with their bass extension as it falls off below 60hz but it is enough to provide decent kick in smaller rooms with most music when used without a subwoofer. That said a subwoofer is recommended if you want full range extension.

    Now the RT002A doesn't have the best distortion performance I have seen from a tweeter but I don't find it too objectionable. It seems most of the harmonics are pushed up above the vocal range so this gives it a slightly grainy or sharper sound which again I don't find causes any listening fatigue the same way distortion in the vocal range does. Dynamics and transparency however are great and it seems to maintain composure better then many domes do which tend to get nasally and compress when pushed hard loosing their detail and separation.

    I found the sound stage to be excellent, the very wide and even off axis response helping there, imaging good as well. Overall I'd consider these to definitely be a step above my Nexus TM design in terms of output capability, dynamics and clarity while maintaining roughly the same footprint and cabinet size.

    The waveguide of the RT002A shows excellent horizontal directivity control maintaining near constant directivity down the the crossover where it transitions fairly smoothly to that of the the 5FE120. Horizontal dispersion is roughly 130 degrees (+-65 degrees).

    Argos off axis response.png

    I ended up with a mostly flat and even response which I found sounded best. There is slight rise at the top octave which when flattened seemed to take some of the life out of treble. This is similar to what I read others have found when using this tweeter and I suspect it has to do with that rise giving a more even overall power response since the vertical axis sees a reduction in output output in the top octaves off axis.

    There is a decent amount of BSC built in but it's just enough that the balance is not too dry when used free field and not to bloated when places near walls.

    Since both drivers have a fairly smooth response the crossover topology is pretty straightforward, a simple 2nd order lowpass on the woofer and 3rd order highpass on the tweeter with a series resistor providing a little padding and contouring. Electrically the highpass on the RT002A is just above the 3khz range but the bump in it's frequency response on the bottom end pushes the acoustic center of the crossover down towards the 2.5khz range. This is also where I found the best phase relationship between these drivers. The actual forward lobe is pointed slightly upwards so the listening window for the flattest response is between directly on axis with the tweeter and about 10 degrees above axis, it's not a sharp cutoff above or below that window.

    A 10uF cap can be substituted in for the 12uF, this will result in a flatter response in the 2-3k range. I did go back and forth on those parts quite a bit, the 10uF does give it a little more presence or detail in that part of the midrange but I found it to cause some music to sound more shouty. I like the more laid back open sound the 12uF value gave with a slight dip in that range. (differences are very subtle)

    The 12 Ohm resistor can be swapped to a 10 Ohm which will result in a slight increase in the treble range for more detail, or a 15 ohm which will result in a slight decrease in treble output. You could alternatively place a 75-80-82 ohm resistor in parallel with the 12 ohm which will equal ~10 ohms, or place a 3 ohm resistor in series with the 12ohm which will give you 15 ohms.



    Xsim response + impedance:


    Box Design:
    The enclosure is 6.5" wide x 10.5" high x 8.5" deep. I used a 3/4" baffle and 1/2" for the rest. Internal volume is approximately 0.2cuft. I placed a pair of cross braces centered between the drivers going front to back and side to side. This cabinet uses the 1-7/8" x 5-5/8" flared port from parts express which results in a tuning of ~65hz. The port is placed on the back and is centered behind the tweeter. Cabinet damping should be light fill of the back half (stay behind the port) I also places some damping around the bottom half near the driver lining the walls.
    Argos TM box.png


    Drivers, crossover components and the port come in at just over $110 per speaker. I suspect a pair could be built for right around $250 including the materials and odds and ends especially if a coupon is used at parts express.

    Argos Parts.png

    Attached Files:

  2. #2 Matt Grant, Dec 7, 2019
    Last edited: Sep 21, 2021
    Argos MTM

    Finally got around to finishing this one up. Much of what has been said about the TM applies to this one as well.

    The main differences are the second woofer which is run in parallel, this makes the nominal impedance 4 Ohms. The crossover is also slightly lower on this MTM version. I also added a few extra components to smooth the frequency response out a little more, partly because I couldn't get an ideal response shape using a similar filter to the TM with adjusted values.

    The second woofer brings distortion down in the bass and midrange compared to the TM but the ribbon tweeter is still the main weak point when it comes to distortion. Although like I had said before most the harmonics are pretty high in the treble frequency so it doesn't sound terribly obnoxious even though the distortion numbers looks pretty bad.

    Surprisingly the horizontal off axis response from this MTM does not looks quite as good as the TM did, I'm not sure what the reason for that is, I may have to remeasure the TM with the new turntable rig.

    Since the tweeter is rotatable this design could be used as a horizontal center channel though ideally you would want to stay within +- 10 degrees, beyond that you get the normal cancellation/lobbing from the woofers which can be seen in the vertical off axis graphs/polar.

    I measured full space sensitivity at ~88.5dB/2.83v/1m or 85.5dB/1w/1m.

    Cabinet dimensions:
    Argos MTM box.png

    Crossover sim:
    Argos MTM X-Sim Response.png

    Argos MTM X-Sim Impeadance.png

    Argos MTM Schematic 9-20-2021.png

    Crossover Parts List:
    Parts can be substituted with others of matching value.
    Note - woofer inductors should be at least 18 gauge, tweeter should be 20 gauge.
    L1 - 1.1mH
    L2 - 0.4mH

    L3 - 0.1mH
    L4 - 0.15mH

    C1 - 18uF
    C2 - 4.0uF
    C3 - 8.2uF
    C4 - 33uF
    C5 - 0.68uF
    C6 - 1.0uF
    R1 - 4 Ohm
    R2 - 10 Ohm
    R3 - 10 Ohm
    R4 - 3 Ohm

    Port Tube

    Measurements - tested out my new turntable so there is a bunch of graphs below, also experimenting with using VituixCAD since it's easy to dump the frd files into it and generate nice looking graphs and polars with minimal effort compared to how I was doing it before and with better consistency.
    Argos MTM Measurements Power+DI new1.png

    Argos MTM Measurements Directivity (hor).png

    Horizontal Normalized:
    Argos MTM Measurements Directivity (hor) Norm.png

    Horizontal Polar:
    Argos MTM Measurements Directivity (hor) Polar.png

    Argos MTM Measurements Directivity (ver).png

    Vertical Normalized:
    Argos MTM Measurements Directivity (ver) Norm.png

    Vertical Polar:
    Argos MTM Measurements Directivity (ver) Polar.png

    Distortion at 80, 90 and 100dB @ 1m:

  3. very interested. it seems the faital 5fe120 its been used a lot now.
  4. I have an existing enclosure that is almost perfect for this project - does it matter if the tweeter is on the top or the bottom?
  5. The tweeter can be on the bottom. It won't really change anything as long as your listening level is roughly equal to the tweeter. As I have noted the ideal listening position or axis for the best frequency response is level with the tweeter or slightly above. If you put the tweeter on the bottom it would then be level with the tweeter or slightly below. You can easily aim the speaker up/down if needed.
  6. On the existing enclosure, the reflex port is located directly behind the lower driver cutout, so I thought that having the back of the 5FE120 point right at the vent would be a bad thing. Is that a problem you think? Maybe I can just replace the straight port tube with a bent one?
  7. As long as there is enough clearance in your cabinet between the internal port opening with the port at the correct length and the woofer magnet it should be ok.

    It's why I often put the ports behind the tweeter on compact rear ported designs like this, there usually isn't enough space for the port to be placed behind the woofer without hitting it or butting right up against it.
  8. A loaded, but perhaps obvious question: how is the performance of the Argos distinct from Paul Carmody's Speedsters? They are in roughly equivalently sized enclosures, and the Speedster's Tang Band woofer supposedly has an f3 in the mid 40's. The Speedster also crosses over higher at around 4.2khz, so maybe the horizontal dispersion doesn't transition as smoothly as the Argos?
  9. I can't speak to sound quality differences since I have not heard Paul's Speedster design.

    From a performance perspective I suspect the Argos are capable of much more output. They use a larger woofer, and have both higher sensitivity and greater power handling. The Speedster in turn has much more bass extension from nearly the same enclosure size, these are definitely lacking if you try to play anything with even moderately deep bass notes.

    I've also heard the speedsters have a fairly narrow vertical sweet spot, that may be in part due to the 4.5k crossover point. With a 2.5k crossover point these should have almost double the vertical coverage before lobing becomes an issue.
  10. Interestingly, I read the same about the very small vertical sweet spot for the HiVi DIY 3.1A, which also crosses to the ribbon around the same point, but the mid is a 2" driver so there shouldn't be much lobing in effect. It sounded like a general problem with ribbons.

    In any case, the parts for an Argos are on their way from PE, so I'll probably have a load of noob questions in a couple of weeks when I have time to work on the build. Looking forward to this!
  11. #11 Matt Grant, Mar 14, 2020
    Last edited: Feb 10, 2021
  12. Is there anything at DIYSG that youd compare it to?
  13. I'd say It's closest in size and performance to the S2000 kit but obviously the use of the ribbon tweeter in this design is a major difference between the kits.

    I should note that after taking new higher resolution measurements on my new turntable both this design (the TM model) and the AlTi will be getting some crossover updates in the near future to fix some issues in the response that had been exposed with those measurements.
  14. I feel like I'm being rewarded for procrastinating on this build! I guess I'll just work on getting the driver cutouts and port worked out for now.

Share This Page