Argos - TM and MTM bookshlf speaker design thread

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.





Background:
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 for the above average sensitivity the little 5" midbass gives.

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 when crossed to a subwoofer. I find the woofer handles itself 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 hard 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.

The RT002A doesn't have the best distortion performance I have seen from a tweeter, most inexpensive ribbons don't seem to do well distortion wise but I don't find it too objectionable. Most of the harmonics are pushed up above the vocal range which gives the highs a slight sense of false detail, again I don't find causes any listening fatigue the same way distortion that lands in in the vocal range does. Dynamics and transparency however are great and it seems to give a quicker attack then many cheap 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.


Measurements:
I ended up with a fairly flat and even response which I felt worked and sounded best in this design. There is slight rise at the top octave which could not be pulled down without using another notch filter, however it is not problematic in normal listening.

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. The woofer response does narrow some in the 2khz range which I believe is due to some diffraction from the cabinet rather then beaming from the driver as the woofer widened back up to match the tweeter at the 3kHz range. Besides that slightly narrowing the horizontal dispersion is roughly 140 degrees (+-70 degrees).

-New full measurement set for the Argos MT here-





Crossover:
2022 update - I have revamped the crossover for this speaker like I have for several of my other designs after updating my measurement process and jigs. The previous measurement method I was using had limited resolution which was masking a peak in the midrange.

For the updated crossover I have completely revamped the low-pass network on the woofer. It is now 3rd order with the the RC acting as a broad contour filter of the upper midrange while the parallel LCR filter in series with the driver acts as a notch of the ~950hz peak which had been previously masked. The tweeter still uses a 3rd order filter with a series resistor for padding but the crossover point has been shifted upwards to 3kHz which takes some strain off the ribbon and actually results in a smoother directivity transition between the drivers.

Crossover Schematic and Parts List:



Note: If using the PCB a 5w 20 Ohm resistor is required for R3 this will have to be purchased elsewhere as Parts Express does not carry one otherwise the 15 Ohm (015-15) they carry could be used instead and results in just slightly more output near 950hz. If you are not using the PCB you can use the 10w 20 Ohm resistor (016-20).

I also forgot to add the 1-7/8" x 5-5/8" flared port tube (260-476).

The total cost for the above is ~$130.

Optional Crossover PCBs for this design can be found here.


VituixCAD simulated response:


Cabinet 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 or lining of the top half surrounding the port tube but staying behind it. I also lined the walls on either side of and behind the woofer. The crossover is placed on the bottom of the cabinet pushed to the back.

 

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:


Crossover sim:


Impedance:




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.


Horizontal:


Horizontal Normalized:


Horizontal Polar:


Vertical:


Vertical Normalized:


Vertical Polar:


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

 

very interested. it seems the faital 5fe120 its been used a lot now.

 

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?

 

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.

 

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?

 

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.

 

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?

 

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.

 

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!

 

PCB's for this design posted on ebay:
https://www.ebay.com/itm/283815380748

 

Added the Argos MTM.

 

Is there anything at DIYSG that youd compare it to?

 

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.

 

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.

 

Hello, is there a PCB designed for the Argos MTM version? Just wanted to check before I pull the trigger on this build.

 

I did not design a PCB for that one yet, but if you are interested I could possibly develop one and have a few made?

 

I appreciate it but don't worry about it. I can use some prototype PCB boards that I have. I just figured I would check beforehand.

Thank you though!

 

I originally planned to build a pair of the Argos TM's just as Covid hit, and was planning on doing it as a covid project. Turns out that I had way too many covid projects, and this project just got made as a birthday gift for my oldest son, who has starting playing his music on the stereo a lot. Waiting over 3 years meant my old crossover PCB and parts were obsolete, and Matt redesigned the crossover, perhaps in response to the more intense measurements that ASR has promoted over the years. In any case, the new crossover seems to be giving some really nice results, but mounting it was a challenge in enclosure I had.



I ended up replacing the original back plate with bi-amp-able binding posts with a 3D printed bracket that elevated the crossover above the binding posts as you can see above. The enclosure I had was just a little smaller than Matt's original design, it was just above .19 cuft, where I think the original design was just above .2cuft. The built in port tube was also 1/8" narrower and about 1/2" shorter, but it seemed close enough overall. The speakers are getting a Voxel subwoofer to cover the bass anyway, so fudging a little on the enclose and port seemed to be fine.
The walls are lined with melamine foam pads, and I put some loose fill around the port tube and over the crossover before closing up the speakers.


After hooking them up to the amp that my kid is getting with the speakers, they sounded great and looked pretty nice as well!


The front baffle cutouts weren't ideal (and the 5FE120's front isn't going to win any beauty contests either), but I have grills for both enclosures and the sound quality is what matters. Thanks Matt for creating this design and making it available for hobbyists! I have additional enclosures and may make another pair for the office.

 

Matt: Any changes over the past couple of years to the MTM version's crossover? These have come on my radar as a possible backup for my RS180 Denham build. I'm thinking I'll build a pair or two and just wanted to check that they didn't get updates like the TM version.

Also, is there any good way to get CNC'd baffles, or even flat packs, for the MTM version? I have the ability to route by hand, but measuring and setting up 6 routes per baffle adds up when making two pairs.

 

No changes on the MTM.

That was the first design I did with the updated measurement rig, the more detailed measurements from that rig is what caused me to go back and revise several of my older designs like the Argos TM.

I can CNC some baffles for these.

 

Awesome. PM me with details. Going to start gathering up parts for the winter building season.

P.S.:
Your Volts and a sub are what keep another family member's house going

 

@Matt Grant Hi Matt, I had to register again after a decade of following you and DIYSG and many more, just to revive an old thread, as this project has really peaked my interest. If the admins allow, I will lay down a few questions, if not I will send you a PM and hope for the best. But here it goes:

1) I really like the MTM design even with the truncated drivers
2) 5FE125 is round and seems to be the same driver but round (easy to cut), still awaiting price+availability from distributor
3) The tweeter is nowhere to be found in Europe, at least I don't know where so I've had great results with HiVi RT1.3WE + I have a custom wave guide avail for it to try something different (and maybe lower XO point)
4) I have limited experience with XOs, but have an UMIK 1, REW and DATS3 and most of the other SW used for design
5) Would you say that 3KHz cross point is acceptable for the MTM version
and finally
6) Would it be rude to try and modify the design with the help and knowledge of the group

I made speakers for my parents some years ago which are now destroyed by our new dog and I want to one-up by making them new speakers. I have access to tools and knowledge about cabinet building and most interesting is the custom wave-guide available for the HiVi RT1.3WE, which is made locally by a hifi shop. My only concern is that the ribbon is rated 10W RMS, recommended 3KHz and above. There are other plannar options from the same co, but I don't want to break the bank and not confident that if going all out will justify the experience and cost. I like this ribbon as it sounds really nice, airy and effortless, but 2x Pro Audio mids might be too much for it to handle.


I think it will be a nice addition to the series and although I like the look of the truncated 5FE120, baffle cutouts take alot of time to be done precisely by hand. It's gonna be either 18mm MDF or Birch PLY all over, maybe 16mm except the baffle.

 

The round frame 5FE125 does certainly appear to be the same driver with the different basket.

With a 3khz crossover I would expect the vertical dispersion to be a bit more narrow but it should still work.

I would not be too worried about the 10w power handling at a 3khz crossover as the tweeter should see very little continuous power and the horn/waveguide should help load the bottom end which will further reduce power requirements.

I certainly don't mind if you try and adapt the design with the drivers you have available.

 

Great news.

I will def update with progress. Currently half the drivers and WGs are on the way. Still waiting for the Faital quote.

Best regards

P.S. Regarding the vertical dispersion, I'm planning for the speakers to be sitting at ear height when sited and maybe a 10deg upward pads underneath them.