Completed project: The Microliths - a small RS125 / RS28 MTM speaker.

[fjhuerta]

The idea:
It all started when a friend asked me for a small minimonitor for his home. I initially thought of an RS28 / RS125 speaker. After running a couple of sims, it was obvious this was not going to work, since a single RS125 could not handle too much load. The speaker grew, as my friend's interest got smaller (due to the size of it).

The speaker itself isn't that big. But Bose and the like have made people think they can get awesome sound in a 3x3x3 package. Since I wasn't very interested in building a bad speaker in the name of size, I built this speaker to be used at my office. It still had to be built as small as possible, but not to the extreme.

From previous experience, a 4" driver isn't exactly a great idea if dynamics and maximum power are desired - especially dynamics. Then again, this is a speaker for smaller rooms - a 6" would make the enclosure too big, which defeats its purpose. The RS150 driver is a better deal than the RS125, IMHO- but this design really requires the smallest driver possible. As it is,the box may be a bit big for some rooms!

Bass alignment:
It was time to find out what size a pair of RS125's would be comfortable in. To my surprise, they needed less volume than a pair of W4-1370s, which was good. I tried to shoot for 11L, so the external size (considering 3/4" MDF) was 45 x 18 x 26 cms. This would give me an 11.72 liter volume, which I'd tune to 55 Hz for an F3 of 54.98 Hz. Nice.

I used a 2" P-E flared port I had around. At 12.5 cms, the size would not be a problem.

I can't complain about FR - considering the driver is more like a 4" one!



At 25 Watt, I can expect the driver to survive. I wouldn't want much more power than that, though. Don't expect this speaker to work well in big rooms - although there's full baffle step compensation, there's a very real risk of smoking the woofers.

[fjhuerta]

The enclosure

The enclosure:
I was going to go for a regular boxy baffle, but since 1) I can't get my hands on P-E boxes, and 2) I wanted somethingg a bit different this time, I decided to go for a slightly different look. Instead of wood laminates, I decided on white polyester, so that the RS drivers could stand out from the box - and I decided on chamfers instead of rounded corners.

After some drawings (I'll post them if there's interest in them - they are at my other PC!) I arrived at a design that emphasized the drivers, while (hopefully!) helping with diffraction. Neither EDGE or the FRD tools could model my idea, so I basically hoped for the best.

2254309045_ceb5988f29.jpg Image not available

(full pics coming later).

It was now time to measure FR in box.

[fjhuerta]

In-box FR:

The chamfers worked as intended... from earlier trials on a box with 90° corners, the chamfer smoothed out the diffraction to a nice degree.

The woofers' response is not as flat as I'd have liked to see, but this is obvious from P-E's graphs. They have a couple of slight bumps in the 1 KHz region. Hopefully, they won't affect the overall design that much.

Here are the upper and lower woofer measurements.





The tweeter is a champ. I love the RS28 - after this design, I prefer it to the SEAS 27TBFC. It's a bit smoother. If you prefer bright sound, I guess the SEAS is a better bet.

[fjhuerta]

Crossover design

Crossover design:

Given the size constraint of the box itself, the crossover had to be as small and simple as possible.

Also, it would have to somehow deal with the break-up of the RS125 itself. It has two peaks.

After initial trials using 2 KHz, 1.6 Khz, and 1.7 KHz, I settled on 1.7 KHz. Phase was a bit better than with the 1.6 KHz version. At 2 KHz version, I couldn't roll off the tweeter fast enough with a 3rd order electrical. Also, the woofer's peaks were more apparent at 2 KHz.

I also tried 6th order acoustical crossovers. This required 4th order electrical networks. Response was a bit better, phase tracking a bit off, but crossover count was HUGE. This wasn't a good alternative - the crossover would never fit in the box!

I decided to use on-axis measurements of the woofers and tweeter, and then use the XYZ distance tools and the piston size parameters on LSPCad to calculate off-axis response as accurately as possible. I also took 15° off-axis measurements of the speaker, so I could simulate them. From listening, I decided that flat 15° off axis sounded best to me, so this is what I modeled. There's also a crossover mod that will get you as close to flat on-axis response as possible.



There isn't anything out of the ordinary here... I decided on 18GA inductors to raise as much as possible impedance in the bass region. This is a 4 Ohm nominal design, and it drops to 3.5 Ohm (or less) on the bass, as you will see. The roll-offs are 4th order L-R, so break-up can be minimised. Both drivers respond quite well to this topology.

The LC shunt on the woofer is a notch filter focused on the biggest woofer peak. It's modeled to take off as much resonance as possible. The values here are important - any different will not be as effective. Peaks are reduced to -40 dB, so they really shouldn't be any issues at all.

Shunt caps are all electrolytics, in order to save $ and space. Space is tight in this box!

The tweeter's filter is a very simple 3rd order. You could possibly replace the 0.140mH inductor for a 0.150 mH one - 0.140 will yield a flatter response.

The 2.5 Ohm resistor yields a flat on axis response. 2 Ohm will yield a flat off-axis response, so pick the one you prefer.

On axis FR is as follows...





Given this constraints, this is the model I ended up with.

[fjhuerta]

Other measurements:

Other measurements:

Impedance.


Drops to 3.5 Ohm. Actually, measurements indicate it's closer to 3.6 Ohm ().

Off-Axis.



A good thing about using a 4" driver... off-axis response is quite nice! Vertical response varies a bit, although 15° is not that bad. This is a speaker that needs to be listened to on tweeter axis.

Transfer functions.



No surprises here - the only notable thing being the shunt filter.

Phase.



The woofers and tweeter are in phase for most of the crossover region - the null confirms this.

[fjhuerta]

Bom

BOM:

As much as I'd like to say these are inexpensive speakers, due to their size, they aren't. The RS28 and RS125s aren't cheap - but they make up for it due to their performance.

Contents of Shopping Cart
Part Number In Stock Description Qty. Price Ext. Price Remove
027-320 Yes 1.0uF 100V Non-Polarized Capacitor .. $0.35 $0.35
027-328 No 3.3uF 100V Non-Polarized Capacitor .. $0.40 $0.40
027-330 Yes 4.0uF 100V Non-Polarized Capacitor .. $0.40 $0.40
027-350 Yes 33uF 100V Non-Polarized Capacitor .. $0.85 $0.85
027-432 Yes Dayton DMPC-15 15uF 250V Polypropylene .. $5.23 $5.23
027-438 Yes Dayton DMPC-25 25uF 250V Polypropylene .. $6.58 $6.58
005-2.5 Yes Mills 2.5 Ohm 12W Non-Inductive Resisto .. $3.50 $3.50
255-022 Yes Jantzen 0.15mH 20 AWG Air Core Inductor .. $2.59 $2.59
255-254 Yes Jantzen 1.2mH 18 AWG Air Core Inductor .. $8.97 $8.97
255-198 Yes Jantzen 0.025mH 18 AWG Air Core Inducto .. $2.28 $2.28
295-360 Yes Dayton RS125S-8 5" Reference Shielded W .. $23.66 $47.32
275-130 Yes Dayton RS28A-4 1-1/8" Aluminum Dome Twe .. $48.65 $48.65
Subtotal: $127.12

This is the price per speaker. Add the cost of your enclosure (mines were $100 each, due to their finish) and you get $227.12 per speaker - $454 per pair - and I'm not considering your choice of terminals and wiring.

When all is said and done, I'd say $500 per pair would be right.

:E

[fjhuerta]

Finished speaker

Finished speaker:

2255108472_28472d9318.jpg Image not available

2254309045_ceb5988f29.jpg Image not available

2254308623_22da3439b2.jpg Image not available

[fjhuerta]

Enough... how do they sound?

Conclusion:

Although these speakers don't have the deepest bass, what they do have is impressive - clean and powerful, up to a point. In a small room, I'm certainly not disappointed at their performance - they play as loud as I'd want them to.

Just for kicks, I tested them in a big room - their tonal balance shifts to the lean side, but they still sound pretty nice.

What really surprises me is their midrange - it's not harsh at all - it's smooooooth, yet detailed and, well, real. Although I thought I preferred paper cones, now I'm really not that sure - I fell in love with the RS125's midrange. These, I know, are all subjective terms, so I'll try to put this in perspective. My Peerless Exclusive 8" based speakers (the Monoliths, the design is in this forum) do not sound as detailed as these speakers. And I prefer the sound of the RS125 midrange. If they had some real bass, I'd prefer these mini-monitors.

Although the design is not inexpensive, I feel it's money well spent. The only other commercial speakers I own with metal drivers are the Monitor Audio Silver RS-8 speakers. They have a tilted-up FR (I guess that makes them more exciting), and the tweeter has a major resonance at around 18 KHz or so. The Monitors are more than twice the price of my design - and quite honestly, I prefer the RS125-RS28 combination. It will not become a tiresome speaker, and it has as much (or more) detail than the RS28s, all while being a lot more neutral.

I know we knew about this, but really - Dayton Audio's RS series are winners at a great price!

[Operandi]

Nice work :T .

I'm going to be building a pair of mini-monitors for my boss and myself with the RS125 & RS28 so this is particularly interesting to me. I was planning on using a TMM version of Roman Bednarek's Microbe SE but this looks like another good option.

[69Stingray]

Nice write-up, thanks for taking the time to provide this information.

I like the looks of your speaker. I can hardly make square boxes, so I can appreciate anyone who steps outside that realm.

I am glad you like the sound. I am typically a paper cone fan also, probably just because that is what I am use to hearing.

Any plans for another design using this as the "midrange"? Maybe a 3-way tower?

[fjhuerta]

Stingray,

My original plans were to build an MTMWW with a set of RS225's. Since my friend asked for a small speaker, I scrapped the WW part.

Curiously enough, I was wondering whether I should add a bass bin with the 225's now. This would require a complete crossover redesign. So maybe I'll just build that set of WMTMW's with the RS150's I've been thinking about for a year now

Given the size constraints, though, I think a separate subwoofer is a better idea. I'm using a 15" sealed Tempest with them, and they sound superb... these little drivers really like it when you stop feeding them bass!

[---k---]

Curiously enough, I was wondering whether I should add a bass bin with the 225's now. This would require a complete crossover redesign. So maybe I'll just build that set of WMTMW's with the RS150's I've been thinking about for a year now !

Yeah, a WMTMW with the RS225/RS150/RS28 is a killer package.

[Undefinition]

Nice look. How did you cut the chamfers gradually like that?

[Jim85IROC]

Nice look. How did you cut the chamfers gradually like that?

This thread should shed some light on cutting the facets:

https://www.htguide.com/forum/showthread.php?t=28981

[fjhuerta]

After some discussion regarding a couple of perceived flaws in the RS125's midrange response, I decided to take some measurements again.

I discovered that I made a mistake, quite possibly during the nearfield / farfield merge. The levels in the woofer are a bit off. I double checked this with the stock PE files with BS and diffraction added.

Apparently, there are two very quick and easy fixes for this mistake. The first one is just to add a bit more inductance and take out some capacitance from the 2nd order filter in the woofer. The second one, the one that seems to be a better choice, is to re-do the whole woofer filter and turn it into a 3rd order. This completely gets rid of all the woofers' peaks, and has a very flat response.

I'll post this updated design as soon as I take some measurements again.

[kevinp.]

how would these compare to the Lineup D44's? I'm looking for a smaller speaker with killer midrange for a small room, music and movies... I'm not concerned with bass response as I have front placed subs. There are so many options I'm overwhelmed! Do I go with a high-end MT with ushers or scan-speaks? Slightly lower end MTMs with Dayton references or TBs? Open baffle? Sealed? Ported? arg!

[h-bar]

The enclosure:
...I decided to go for a slightly different look. Instead of wood laminates, I decided on white polyester, so that the RS drivers could stand out from the box...

Please excuse my ignorance, but what do you mean by white polyester? Did you apply the finish after assembling the enclosures or build the enclosures out of materials that had the polyester finish pre-applied? If you applied the finish to the completed enclosures, can you provide a little detail about what you did? Either way, the result looks great.

h-bar

[fjhuerta]

Please excuse my ignorance, but what do you mean by white polyester? Did you apply the finish after assembling the enclosures or build the enclosures out of materials that had the polyester finish pre-applied? If you applied the finish to the completed enclosures, can you provide a little detail about what you did? Either way, the result looks great.

h-bar

Hi,

I had a professional carpenter apply the finish. I wouldn't dare do it - even he has problems with the finish now and then.

As for the speakers... they are great in small rooms. I learned a lot with this project - mainly, that I do prefer smaller woofers as midranges. I haven't yet learned enough what possibly could be happening - my guess is that a small speaker, when used with a subwoofer (while I had them I used them with a sub crossed over at 100 Hz) has better transient response than a big woofer. I don't know.

[Jed]

how would these compare to the Lineup D44's? I!

The TB W4-1337 is one of the best mids I've ever heard. The RS series tend to have great bass however.

[fjhuerta]

The TB W4-1337 is one of the best mids I've ever heard. The RS series tend to have great bass however.

I've built an MTM with the W4s, too. As a matter of fact, I built this MTM because I gave the W4's MTM away to my sister, and I wanted to build a similar speaker. It also used the RS28 tweeter. (I don't think I ever posted the design, though).

I have to agree. Subjectively (because I never took any distortion measurements), the W4s are a lot better than the RS125s in the mids. I don't know if it's the distortion profile or the lighter mass of the W4s, but they are quite a bit more relaxed sounding. It's one of my favorite designs, actually.

The RS125's model is quite a bit more economical, though.

[fbov]

... Apparently, there are two very quick and easy fixes for this mistake. The first one is just to add a bit more inductance and take out some capacitance from the 2nd order filter in the woofer. The second one, the one that seems to be a better choice, is to re-do the whole woofer filter and turn it into a 3rd order. This completely gets rid of all the woofers' peaks, and has a very flat response.

I'll post this updated design as soon as I take some measurements again.

Javier,
Any chance you've got an updated crossover you haven't posted?

I'm seriously considering this design for a new center and surrounds. The appeal is the 125mm driver form factor (my center is limited to 6" tall) and the use of Dayton RS-series components similar to those used in Natalie P (my goal for new mains). My plan is to build two surrounds in sealed enclosures as you did here, but modify the third, center enclosure, adding depth to maintain volume at a 6" finished height.

Thus I actually have three crossover requests:
- do you have an updated XO design like that refereneced
- do you have one without BSC? The center will be flush-mounted.
- anything different in a 45x15.25x28.6 cm box?

thanks,
Frank

[fjhuerta]

I need to post the updated crossovers for two designs now... I'll do it later today or tomorrow morning.

[fjhuerta]

Here it goes... second version of the crossover. Much better sounding, and it's the definitive version.

Frank, as for a version without BSC, I could work on one - as long as your center channel will be vertically placed. I doubt this speaker used horizontally will work well. I don't think the box you are planning may affect FR that much, but the crossover may need to be tweaked a bit.

[fbov]

Thanks, Javier. As you may know, I'm working on a different track for a center speaker, but these are still under serious consideration for surrounds.
Thanks again,
Frank