Heeman's Dynaudio C1 + DIY Scan-Speak 13" Woofer Cabinet

[Flint]

Heeman’s 13” Scan-Speak High-Fidelity Bass Speakers
Best bass I've ever heard!


INTRODUCTION AND BACKGROUND

Starting in April 2017, my good friend Heeman decided it was time to do something to address the limitations of his Dynaudio Contour C1 stand mounted loudspeakers (henceforth referred to as “C1” or “C1s”), perhaps some of the finest small speakers on the market. While they excelled in their ability to create a stunning stereo image and provide extremely high dynamic range in the treble and midrange, they clearly suffered in the main bass region below approximately 200Hz. They also struggled to blend completely naturally to the two huge PSA subwoofers which could shake the skin off your legs at the subsonic frequencies, but which suffered at providing tight, clean, dynamic, and natural-sounding bass above about 60Hz. Clearly there was a gap in how his main stereo system performed in the frequency range above the subwoofers’ best low frequency range and below the limits of the small 6” speakers in the Dynaudio system below about 200Hz. Taking crossover regions into account (where two speakers overlap and still need to perform well) he required about four octaves of ultra-high-performance with high dynamics, low distortion, flat response, and smooth group delay from about 25Hz to nearly 400Hz, not an easy thing to fulfill given the nature of the industry’s love of small satellites and huge subwoofers which create this gap in performance.

To solve the issue, one option was to replace his main speakers with a pair of larger Dynaudio towers with more woofers to address his demands in the bass range, but the next speaker up in the Dynaudio line was the Contour C2 which was basically two C1 speakers stacked on top of each other. The physics of added a second 6” woofer meant an extra 6dB SPL of output in the bass range at issue, but the overall performance below 100Hz was still similar. They would just not give up until the volume control was much higher. Also, the cost of that upgrade was huge – like really huge – to get one more 6” woofer. Another option was to look at other brands of speakers, but Heeman loved the sound of the C1s in the midrange and treble, the imaging was spot on, and though it has nothing to do with performance they looked terribly attractive in his HT. Really, he just wanted to, somehow, magically solve the problem of the bass suffering at the listening levels at which he loved listening to music and concerts.

So, a third option was explored, which was to add a single bass enclosure to each channel in the stereo system with an adjustable active crossover to make the complexity of incorporation into the system as simple as possible. If Heeman could build a couple of large-ish enclosures from MDF and mount high quality woofers in them which would perform magnificently in the 25 to 400Hz range, the issue would be addressed, and he could continue to enjoy the performance he loved from the C1 satellite speakers as well as his carefully placed and calibrated PSA subwoofers. So, we got to work.

After researching the available best-in-class raw bass transducers on the market, together we narrowed it down to either enclosures with dual 9” woofers in them, or enclosures with a single 12” to 13” woofers in them. Surprisingly there are not very many bass speakers which are completely optimized to deliver the best performance in this range as the popularity of two-way satellites married to large subwoofers has made the demand for amazing large woofers appropriate for a three-way speaker relatively small. So, the selection process was inherently simplified by the catalog choices available to us.

To be clear, there are a ton of amazing subwoofer drivers meant to sound great in the 60Hz and below range, sometimes reaching down to 15Hz. But when asked to provide musically clear, clean, loud, dynamic and controlled output in the 120Hz to 400Hz range, those high excursion, high power handling behemoths struggle to fit into the definition of “high-fidelity.” Likewise, the plethora of mid-bass speakers which are designed to perform well in the midrange up to 6kHz all struggle to provide high SPL clean bass below 200Hz. So, the naturally small selection of speakers designed to play in the primary bass frequency range and provide outstanding high-fidelity performance makes choosing a tad easy in comparison to any other class of speaker.

After comparing computer simulations models for about a dozen high-end large bass speakers, it became clear that the best possible options would be either a pair of 9” woofers from Eton, a very well-respected German manufacturer, or a new to market single 13” woofer from Scan-Speak, one the finest transducer manufacturers in the world out of Denmark. Heeman and I hemmed and hawed over the pros and cons of each until we decided on the brand new 4 ohm 13” woofer from Scan-Speak being the best bet. Since it is new to the market, we didn’t have many industry examples of its use, but the specifications exceeded our goals and Heeman felt the high price was worth it for his needs, especially considering the next best option was upgrading to the Dynaudio C2 (something we calculated after selling the current C1s to cost about 4x more than building these new bass bins and all the extra components necessary to make them work). So, I got to work designing bass cabinets for Heeman to build.

 

[Flint]

THE DESIGN PROCESS

Since the basic modeling demonstrated the 13” drivers from Scan-Speak would be more than sufficient to solve Heeman’s concerns while delivering what might turn out to be the best bass performance possible in his room, the work of designing an enclosure fell down to fine tuning the parameters then adjusting for real world materials, displacements, and common issues like enclosure wall vibrations and such. Using a version of WinISD I had customized to match my real world experience with designing enclosures by downloading the code when it was in beta and asking a physics junky friend to code my tweaks for me, I adjusted the effective volume and tuning frequency of the enclosure to get a modified-B4 alignment which would balance well with the moderate room gain of his HT and perform excellently to as low as 27Hz with a bass limit (-10dB) of about 22Hz. I had envisioned Heeman insisting his subs be included in stereo listening, so I knew the crossover on his HT processor would go no lower than 40Hz. So, the 13” speakers would, I supposed, not need to reach into the depths of bass the woofer was potentially capable of delivering. This is a good thing, because designing for an extra low bass extension comes at the cost of losing out on max SPLs and increasing distortions in the frequencies just above the tuning frequency. So, this approach made sense to me. The models suggested that when using the amplifiers he already owned, Heeman would get as much as 115dB SPL from each 13” speaker (for a max SPL of 121dB before room gain) with acceptable distortion levels from 40Hz to 400Hz – the target range for these speakers.

I then went about determining the hard requirements for height and width as they play a role in the dimensions of the enclosure. I started with height, as these would be replacing his current speaker stands and the C1s would be place on top of each bass bin with a small isolation pad, most likely foam, and a small wood shelf to stabilize the very narrow C1 base. I also wanted to allow for some form of rubber or foam isolation from the floor, so I settled on an external enclosure height of approximately 20 inches. The next dimension of consequence was the width, which needed to be wide enough for the driver, but more importantly I was looking forward to the possibility that Heeman might build custom speakers to replace his C1s, and I am convinced that a wide, large baffle is superior to the narrow baffles that currently fill the market for speakers. I had done real research on ideal widths on my custom speakers I made about a decade ago, even building about four test baffles and measuring the results. The ideal practical width was almost 16 inches, so that was what I chose for these bass enclosures. The depth would then be determined by the necessary internal volume of the enclosure, and initially it was to be nearly 23 inches deep (later I’ll explain how modifications were made as we adjusted to real world volumes of the woofer, port, bracing and such).

Drawing 1: Artist rendition of the new woofer cabinet with Dynaudio C1 on top to show relative size.

The next steps were to determine the necessary materials and parts and acquire them. I hedged my bet on acoustic dampening (more on that later) and we figured that two sheets of ¾ inch MDF, Tightbond yellow wood glue, Precision Sound flared 4” components for the ports, simple speaker wire binding terminal cups, quick connects for the internal wiring (to allow for future work), mineral impregnated rubber sheets to dampen the enclosure wall surfaces, some flexible silicone caulk, and the mounting screws for the woofer. The final finish for the MDF was left up in the air for Heeman to decide on after he saw how the enclosures looked in his theater. Heeman also needed some tools such as a router, circle cutting jig, and such to build the enclosures effectively and easily.

Before the first cuts were made, Heeman got a friend from work involved, a smart mechanical engineer who enjoyed modeling the project in CAD and getting extremely precise volume displacements for the parts inside the enclosure. In the end the depth of the enclosure was altered about an inch from my original plans ensuring the ideal realization of the calculations and simulations I had made. I’ll discuss how precise that extra work was in my measurements.

 

[Flint]

THE BUILDING PROCESS

After days of preparation and fear, and about a dozen phone calls with me, Heeman and his friend started cutting the MDF and assembling the enclosures. I highly recommended using only glue to hold the MDF seams together, avoiding nails, brads, screws or staples. The result is a more stable enclosure with no metal potentially sticking out to hinder router bits, sanders, or make painting more complicated. Since Heeman and his friend decided to cut dados at every joint, the assembly went easily. They even used a round-over router bit to soften the edges throughout the inside of the heavily braced enclosure and the outside edges around the enclosure. This made for an easier to handle production and will cut down on any potential turbulence which could occur inside the box when the woofer is working hard.

Below is an exploded engineering drawing Heeman’s friend made of what they actually cut and assembled, in mirrored pairs (since the braces are not internally symmetrical).


Drawing 2: Engineering Drawing, exploded, of cabinet with braces and cutouts


Drawing 3: Engineering drawing, solid and transparent, of assembled enclosure

(I am not supplying detailed design plans for the cabinet free on the internet since this work was done with proprietary tools and skills. I don’t want to see my work stolen or turned into a product sold by others. I intend to sell speakers in the future and this may become part of my product portfolio. If you wish to make these speakers for yourself, contact me and we can work together in private.)​

We were all surprised at how quickly the enclosures went together, and once the glue fully dried it was easy to mount the extra parts and woofers and hook up the bass bins. Great attention was placed on damping the cabinet walls to reduce vibrations and no acoustic stuffing or other acoustic absorption was put in the enclosure to get the highest efficiency from the enclosure – this should increase dynamics and tighten the bass.

 

[Flint]

INITIAL TUNING

Heeman and I worked over the phone, through email, and with texting to get the crossover tuned for initial listening. Using the highly regarded miniDSP HD crossover, we focused on the theoretical goals of the speaker and initially set the crossover to 350Hz using a 48dB/octave Linkwitz-Riley filter. Heeman used his calibrated measurement mic and TrueRTA software to fine tune the levels and he was quite pleased with the sound from the start.

His initial impressions suggested the dynamic limits of the C1s were completely eliminated and the sound seemed to blend well enough that the overall tonality and performance of the C1s were not lost with the addition. Further, he could turn the levels up significantly higher without any sign of stress from the speakers – one of his biggest desires. In fact, he learned the hard way that he could easily damage his hearing well before he would perceive any limitations of the speakers’ loudness levels.

After several months, I visited Heeman and used my measurement tools to fine tune the system even further and address some acoustical and equipment limitations of the system to get what I perceived as the best possible results for the system. More on that in the measurements section below the listening review.

 

[Flint]

CRITICAL LISTENING EXPERIENCE & REVIEW

Once everything was tuned as perfectly as possible, and all the measurements were completed, I sat down for a highly-focused listening session. As always, I used my fabulous Speaker Auditioning dual CD set which includes high quality tracks chosen for how they challenge speakers and for my familiarity with the content. Right away I was completely blown away by the clarity of the bass performance. It was tighter than anything I’ve ever heard, extremely clean, amazingly dynamic, and at no point did the sound seem to be coming from a mere pair of speakers. I was hearing the music, the instruments (most of which were real instruments, not synthesizers or samples), and the raw energy of the tracks.

I was, and am, impressed! Heeman’s incredibly thorough job of treating the room to achieve perfect bass with massive quantities of absorption material compromising well placed bass traps, his construction of non-parallel side walls, and his care to detail on placement of speakers was fully realized with the installation of these absolute state of the art 13” woofers in proper enclosures and tuned to perfection.

I have never, in my entire life, experienced bass performance like Heeman now gets to enjoy on a daily basis. It is darn near impossible to describe other than it just works. Much like we love to talk about guitars or singer virtually appearing in the room, with Heeman’s rig, the same thing happens with the most complex and difficult bass, like Peter Gabriel’s “Don’t Give Up” from the album, So…, or the incredibly dense mix of drums, bass, and keyboards filling the low end of “At the End of the Day” from V by Spock’s Beard. Everything remains detailed, separated, clear, and… well… just like being in the room while the music was being performed. Gone were all symptoms of bloated bass, or lack of definition, or muddying of sound. One hears exactly what is on the recording, and if I played a song with muddy bass on the recording, that’s what I’d hear. The closest thing I can think of to compare it to is an excellent set of headphones for their clarity and detail – but with the speakers you can feel the bass and get the real impact of the performance. That visceral experience we often seek and tend to experience at live concerts.

Of course, this sort of system brings out the worst of a recording as well. So, while an amazing recording will make one journey to Nirvana, and poor recording – especially one you’ve loved for decades for what you thought was great sound – can disappoint beyond belief. This tends to happen for me when watching concert videos where they over-compress the recordings and often sacrifice aural fidelity for surround sound or capturing the energy of the show. While many love that experience, I prefer to listen to the studio recordings of the music I love over watching a concert video. One of the nice things about watching old concerts, like old stuff from The Who, Jethro Tull, Yes, King Crimson, and so on, is that the audio is just good enough to enjoy the show and no pretense to high fidelity was made – thus allowing me to just watch and love the performance rather than internally bemoan the poor mix or lousy dynamic range.

 

[Flint]

MEASUREMENTS AND TUNING

Let’s start with real world ground plane measurements of the woofer system taken on Heeman’s driveway. In summary, I calibrated the SPL and amp output. Here are the measurements at the equivelant of 2W, 20W, and 200W (into 4 ohms) feeding the woofer:


Figure 1: Pseudo-Anechoic output from Woofer at 2W, 20W, and 200W into 4 Ohms at 1 Meter.

To make that data into information, as in how does it affect performance, here’s the same curves normalized to the 2W measurement so you can see the nature of the dynamic range of the woofer:

Figure 2: Power Compression for 2W (green), 20W (purple), and 200W (yellow)

Note that down to 50Hz it is damn near perfect and real compression with 200W input doesn’t occur until about 32Hz. Since Heeman is using a 12dB crossover at 40Hz, this should be negligable. I am very impressed with these measurements and cannot think of many high fidelity woofers which can manage such clean and dynamic performance.

To verify the loading and tuning of the enclosure as it relates to the woofer itself, I made close mic measurements of the woofer and port which are shown below:

Figure 3: Close Mic measurements of Woofer and Port

I was very impressed that the measured tuning frequency and amplitude exactly what was predicted when modeling the design. Fun stuff.


… test and measurement continued below....

 

[Flint]

…. Test & Measurement Continued....

We then returned the woofer cabinet to the main system in the listening room and tuned it to the system.

We compared the measurements of the woofer outside to the performance in the room to gather info on how the room impacted the response performance. Here's the response from one woofer at 2W then again at 200W in the room and outdoors in a pseudo-anechoic measurement:

Figure 4: In-room measurements at 2W (purple) and 200W (gold) and outdoors at 2W (green) and 200W (yellow)

Using these data sets I was able to create a transfer function curve that visualizes the impact of the room on the output from the woofer:

Figure 5: Impact on amplitude of woofer from the room reinforcements

A room gain of 18db to 24dB in the deep bass is always expected (which is why small subwoofers can sometimes be impressive). But the cancellations from room modes and gain from room nodes show the impact of the room from 10Hz to about 200Hz. Above 200Hz the impact will vary dramatically based on where the microphone (or listener) is placed. Extreme nulls, like those at 45Hz and 90Hz can also be impacted by listener placement, but they are more likely to be just a fact of life in the room.

I was also aware of a bump in the response in the 180Hz range which mic placement didn't seem to affect. It was also not in the pseudo-anechoic measurements, so it was a room issue as seen in the room curve above (fig.5) which I decided to address with a parametric EQ in the crossover centered at 187Hz and cutting just over half the measured peak level. which I decided to address with a parametric EQ in the crossover centered at 187Hz and cutting just over half the measured peak level:

Figure 6: Effect of 187Hz Parametric EQ on response

One immediately obvious issue was the blending of the subwoofer to the 12” woofer cabinet, so after playing with phase and levels and even crossover frequency settings, we decided to see what happens when we go through the range of subwoofer “distance” settings in the preamp/processor. It was quite amazing how this delay setting affected the performance, and we went through every possible setting and measured the results. Here’s a chart showing where we started and what we settled on:

Figure 7: Difference in response in bass from different subwoofer distance settings in Preamp/Processor

….. Test and Measurement Continued Below...

 

[Flint]

…. Test and Measurement Continued...

So, after fiddling, testing, listening, measuring, fiddling some more, and getting things to the best balance between measured and perceived sound quality, this is the frequency response of the left speaker at the listening position with and without a parametric EQ cutting the peak in the high end (about 16,000Hz) of the tweeter:

Figure 8: Final response of auditioned system for review, above

This was tuned to my preference which was a fairly flat low end sub-bass response. I believe Keith preferred it with more bass and later turned up his subwoofers which are quite significant at frequencies below 70Hz.

Then, as a final check we ran a series of response measurements of all the speakers in the system to ensure the levels were on point and their overall timber was matched, as they should be:

Figure 8: Measurements of all speakers in surround system using test tones in Preamp/Processor

As you can clearly see, the overall balance between all the speakers is the same. The subwoofer was measured at the "ideal" level as shown in the Navy Blue curve, then as Heeman liked to set it with the Green curve. I prefer the lower setting.

 

[Flint]

Ultimately, the sound was unbelievable once we got done and I was very, very impressed both with the design, build, and testing, but also the listening.

I will let Keith add photos of the speakers as he chooses. Including build photos, in room photos, and his final competed photos now that he's blacked the cabinets with a nice finish.

 

[Wardsweb]

Wow, that was a lot of time and effort, but it sounds like it was all worth it. Congratulations.

 

[PaulyT]

Wow, thanks for the detail. Still reading through it... BUt I can personally attest to how awesome those speakers - and room - are.

 

[team_tsp]

Very cool to see the design process that went into this. Just curious @Flint, is there a reason you didn't use the same driver for your own, new speaker build? Different goals, room acoustics, etc?

 

[heeman]

Thanks again to Flint on his support and dedication to this great project!!

For those of you that have not seen the build thread, it is here:

https://www.theaudioannex.com/forum/threads/c1s-13.12144/

 

[Flint]

Just curious @Flint, is there a reason you didn't use the same driver for your own, new speaker build? Different goals, room acoustics, etc?

The reasons for using two 10" drivers over a single 13" driver in my system are many. The most important difference is my desire to always have more than anyone else. While the a stereo pair of this particular 13" driver can generate gobs of output at low distortion and very little power compression, I had to go one better than Heeman to stroke my ego. So, I used two stereo pairs of 10" drivers from the same family which are together capable of more output, lower distortion, and higher output before power compression sets in. Also, while I discussed this approach to Heeman, his layout limits how deep a speaker can be. In order for the height and width of my 10" enclosures to be appropriate for my goals, I had to make the enclosures very deep. This would not have worked in Heeman's room. This is made worse by my using a nearly 20" width on my speakers where his was 16" wide - if the width is narrower than my, but the height is the same, then the depth would be even greater than I had to settle for. So, it wasn't practical. In Heeman's case, the single 13" woofer per channel is the best option. In both systems the proper bass range can generate massive amounts of output without distortion or compression, so we are both good.

 

[Dentman]

Outstanding read, thanks.
The funny thing about Flint is, if you didn't know better and only listened to his systems most would wrongly think he hates bass. That is when playing 80% of most rock/pop type music. He keeps his bass freq extremely flat with no boost what so ever.

Great work boys!

 

[Razz]

great work!!!
I really would love to hear that system of yours Heeman! Now put up some pics.

Ok, No laughing at this next question...
On the final check on the TrueRTA, the SPL levels in the low frequencies are around 55 and then go up to 80 in the high end. Why is that? Don't you want it flat? Could this be why Heeman likes the Subwoofer set at a higher SPL level?

 

[Flint]

great work!!!
I really would love to hear that system of yours Heeman! Now put up some pics.

Ok, No laughing at this next question...
On the final check on the TrueRTA, the SPL levels in the low frequencies are around 55 and then go up to 80 in the high end. Why is that? Don't you want it flat? Could this be why Heeman likes the Subwoofer set at a higher SPL level?

I was using the test tones from the preamp/processor which are Pink Noise for the subwoofer and white noise for the 5 main speakers. White noise has that shape - each higher octave is 3dB louder than the lower octave. So, if 100Hz is 55dB, then 200Hz is 58dB, and 400Hz is 61dB, then 800Hz is 64dB, and so on. A perfect curve would be a straight line which goes up 3dB each octave.

 

[Flint]

The reasons for using two 10" drivers over a single 13" driver in my system are many. The most important difference is my desire to always have more than anyone else. While the a stereo pair of this particular 13" driver can generate gobs of output at low distortion and very little power compression, I had to go one better than Heeman to stroke my ego. So, I used two stereo pairs of 10" drivers from the same family which are together capable of more output, lower distortion, and higher output before power compression sets in. Also, while I discussed this approach to Heeman, his layout limits how deep a speaker can be. In order for the height and width of my 10" enclosures to be appropriate for my goals, I had to make the enclosures very deep. This would not have worked in Heeman's room. This is made worse by my using a nearly 20" width on my speakers where his was 16" wide - if the width is narrower than my, but the height is the same, then the depth would be even greater than I had to settle for. So, it wasn't practical. In Heeman's case, the single 13" woofer per channel is the best option. In both systems the proper bass range can generate massive amounts of output without distortion or compression, so we are both good.

I will add, one of the driving factors for most of my decisions when replacing my previous speaker builds is to get 10x better performance from the new speakers. I did a huge upgrade going from my old speakers, the Audax/Dynaudio blend now residing with another member of this forum, to my previous set with custom midranges and top end Morel tweeters and Focal W-Cone 12" woofers. The Focal woofers were very good, and while a single 13" Scan-Speak like Heeman's would have been better than the focal woofer, I wanted to be 10x better, so I went with two 10" Scan-Speak woofers which gets me about 15dB SPL more peak output before power compression and stronger output below 35Hz.

 

[heeman]

great work!!!
I really would love to hear that system of yours Heeman! Now put up some pics.

Ok, No laughing at this next question...
On the final check on the TrueRTA, the SPL levels in the low frequencies are around 55 and then go up to 80 in the high end. Why is that? Don't you want it flat? Could this be why Heeman likes the Subwoofer set at a higher SPL level?

You can find the pic's here:

https://www.theaudioannex.com/forum/threads/c1s-13.12144/

 

[heeman]

@Flint

I am curious to understand the relationship of how a system (rig) measures vs. how it actually sounds.

You have made some pretty serious statements about the bass performance of my system (which of course includes the room), however the measurements really don't correlate with your statement.

Can you please help me, us, understand this???