[matevana]

It’s been two years since the last Hestia project and there are a number of concepts I’ve been waiting to test. I plan to start building the prototypes this week and will document as the build progresses. As always, feel free to join in.

The Hestia Hybrid uses a primary full range driver mounted in an open baffle configuration with LF support from a secondary driver. It is envisioned as an alternative to well-crafted studio monitors, combining the sonic benefits of dipolar / full range loudspeakers with the manageable size and stand-alone characteristics of traditional bookshelf speakers. A compact, self-contained loudspeaker system with a modern/industrial aesthetic.

System type: A single dipole point source combined with a larger monopole up firing driver that is mounted in a tuned and damped slot loaded enclosure. Both drivers operate without a crossover.

Drivers:

Primary – Jordan Eikona II / 100mm aluminum cone
Secondary – Peerless 830869 / 8” Nomex cone

Theoretical benefits and testing objectives: (Evaluations 1-6 below).

Both drivers operate full range electrically and avoid the monopole to dipole null that would occur around a fixed crossover point. This helps maintain a symmetrical figure eight pattern associated with classic dipolar dispersion. (Eval 1). 

The right angle driver coupling helps reduce frequency modulation distortion (Eval 2) and is a preferred method for combining two full range drivers. 

A tuned and heavily damped front firing slot provides low frequency reinforcement that sums constructively with the dipole driver. (Eval 3).   

The tuning frequency is centered at the -10dB rolloff point of the dipole driver for cohesive (non- destructive) low frequency support (Eval 4).

An adjustable driver mount is used to better align the acoustic centers of both drivers and optimize phase coherence (Eval 5).

The adjustable dipole mount provides micro toe adjustments as well as a way to help manage floor and ceiling reflections (Eval 6).

-More to come-

[JCarney]

I know I will enjoy what I see and read in this thread. Can't wait for more.

Thanks matevana,
JCarney

[Woodsage]

Looks like a cool build!

When I first looked at it I pictured a 10” full range and a 15” woofer.

Is the woofer active? Also, do you think the up firing bass could effect the sound of the full range driver, particularly at high volumes?

Mike

[matevana]

Is the woofer active? Also, do you think the up firing bass could effect the sound of the full range driver, particularly at high volumes?

Mike

Hey Mike,

The theory is that both drivers are passive (full range) and summing them at a right angle should be more constructive than trying to do this on the same plane. While there is no electrical crossover, the larger driver is mechanically rolled off trough a tuned slot and is heavily damped with acousta-stuff filling. When you put your ear to the slot, you should only hear low frequencies emerging from the front of the enclosure. If I did the math correctly, these low frequencies should reinforce the low end of the top mounted dipole driver. I will likely need to tweak the slot length and the amount of fill, once the test cabinets are built.   

When you see the photos as the cabinets are built, it may be easier to visualize what I am trying to accomplish.   

[Woodsage]

Hey Mike,

The theory is that both drivers are passive (full range) and summing them at a right angle should be more constructive than trying to do this on the same plane. While there is no electrical crossover, the larger driver is mechanically rolled off trough a tuned slot and is heavily damped with acousta-stuff filling. When you put your ear to the slot, you should only hear low frequencies emerging from the front of the enclosure. If I did the math correctly, these low frequencies should reinforce the low end of the top mounted dipole driver. I will likely need to tweak the slot length and the amount of fill, once the test cabinets are built.   

When you see the photos as the cabinets are built, it may be easier to visualize what I am trying to accomplish.   

Very cool, thanks for the clarification matevana.

I’ll be watching with interest!

Mike

[matevana]

The Hestia Hybrid is first and foremost an open baffle loudspeaker that uses a full range driver mounted in an open baffle.  The ‘hybrid’ concept introduces a 2nd driver in a slot loaded enclosure that is primarily intended to compensate for some low frequency loss due to dipole roll-off. To accomplish this, it is necessary to tune the slot to a frequency that sums coherently with the output from the open baffle driver.

I realize this is an open baffle forum, but for the purpose of this specific project it might be helpful to share some information about slot design in an enclosure. 

The tuning frequency (Fb) of the slot is a function of the slots cross sectional area (Av), its effective length (Lv), and the enclosure volume (Vb).  You can replace variables and solve for unknowns using this formula: 



For the Hestia Hybrid, the target frequency of the slot is 70Hz, which is where the open baffle driver plays approximately 10dB lower than its average sensitivity mounted in its baffle. I am using heavy cardboard to experiment with different tunings (simulating slot length and width) before the plywood slot gets cut and installed. There may be some additional challenges in trying to couple the slot’s output to that of a dipole driver, i.e., different roll-off properties and subsequent phase abnormalities, but I am hoping to minimize this through experimentation.     

[matevana]

Here's some construction photos. Next on to evaluating and tuning.

[bfstrysik]

Very interesting design. Looking forward to your thoughts as to how they sound

[Woodsage]

Looking good!

And a very clever use of a grab bar.......

Mike

[matevana]

Looking good! And a very clever use of a grab bar.......

LOL, Good eye!  I do Procurement for a living and like sourcing materials to be used in different ways. Not only are grab bars very strong, but they float the dipole securely over the up-firing driver with plenty of room for aligning the two. It also hides the speaker wire going to the back of the dipole and makes it easy to lift and move the speaker. I'm going to use thumb wheels instead of lock nuts so adjustments can be made more easily.



 

 

[matevana]

A half pound of Acousta-stuff seems to yield the best output at the slot, taming any peaks in a less than optimally sized enclosure. I also tried ¼ and ¾ pounds. A mesh zipper bag holds the fill so the individual strands don’t work themselves down the back of the cabinet and into the slot.





The Peerless 8” Nomex driver is no slouch. It was selected among several 8” drivers due to its flat-ish response since it will operate without a crossover. Even though the Peerless driver has higher sensitivity than the Jordan (90.2 dB 1W/1m vs. 85.2 dB 1W/1m), experience tells me that I will still have to attenuate the Jordan’s output when level matching the two drivers. In reality, I’ll be level matching the output of the slot with that of the Jordan and factoring-in dipole cancellation.   

[matevana]

The next and perhaps most important tuning is the forward adjustment of the dipole driver. The dipole mount allows for 5 inches of lateral travel that can be dialed-in precisely. In principle, you want the acoustic center of the dipole driver to float directly above the dust cap’s center of the up-firing driver. Finding the exact acoustic center of a driver is easier said than done. However the point from where sound originates is generally close to the voice coil. Cone drivers typically have their acoustic center closer to the back of the driver. The Jordan’s acoustic center is located in the rear third when viewing the frame from the top.

Dipole mount (rear)



Dipole mount (front)



This would be a straight forward adjustment if the Jordan was not mounted in an open baffle and if cancellation was not an issue. But due to the minimalist baffle, the Jordan will act more like a midrange/tweeter in this design. Because of this, it can be advantageous to compensate for the time arrival of higher frequencies at the listening position, and adjust the driver’s position to coincide with the lower frequencies coming from the slot. All things being equal, higher frequencies will arrive at the listening position earlier than lower frequencies. This can be offset by sliding the dipole driver back in relation to the up-firing driver’s center.

Using tone bursts of various frequencies, I was able to clock the arrival of the burst in milliseconds at the listening position, which in my case is 10 feet from the front of each speaker. The time difference between the average set of low frequencies and the average set of high frequencies can be represented by approximately 2 inches of rearward travel on the dipole mount. For the purpose of initial listening sessions, this will be my default setting.

[matevana]

The final piece is a carpet tile inlay applied to the top baffle. This will help damp any possible reflection from the dipole driver. The front panel was cut 1/4" oversize so the inlay sits flush from the front.

[matevana]

And here's the finished project and stand detail. My preferred listening height is having the full range driver slightly higher than ear level when seated.

Once the shiny new toy syndrome passes, I'll post some listening impressions.

[rollo]

Nice, Kinda looks like a Gradient without the sloped Mid. Enjoy looking VG.


charles

[Woodsage]

Looks very good and, aside from the Jordans, it looks to be a pretty inexpensive build.
 
Is the box about 13-14" cubed? I'm trying to get an idea of the overall size.

Looking forward to your listening impressions.

Mike

[matevana]

Nice, Kinda looks like a Gradient without the sloped Mid. Enjoy looking VG.


charles

I got to see Gradient's Helsinki in-person. They actually use sheet glass as part of the design which I think is very creative.  I imagine it would be like owning a piece of art and a great conversation piece. The finish was also top notch. 

[matevana]

Looks very good and, aside from the Jordans, it looks to be a pretty inexpensive build.
 
Is the box about 13-14" cubed? I'm trying to get an idea of the overall size.

Looking forward to your listening impressions.

Mike

One of the cool things about the dipole baffle being so minimalist is that you can have several and swap drivers in minutes. There are several good full range drivers in the 4-6" range (Tang Band, Fostex, Seas, etc.) that should work well in this design. I'd be more concerned about substituting for the Peerless driver, as you essentially need a woofer that will play politely without a crossover and in my experience there are few. The Seas H1411 is another woofer that should work but is slightly larger at 10" It is designed to be used in Dynaco A25 replacements, which also did not use a crossover.

The cabinet is actually only 12" x 12" x 12". It has a net internal volume of about .70 cubic feet. It looks larger in the photos than in real life.

[wendelltate]

How much of the theory of this is like the LXMini?  I use those as my main speakers and they are incredible.  These look very similar.

[matevana]

How much of the theory of this is like the LXMini?

They are probably more different then they are similar. They both share the right angle coupling of two drivers to minimize frequency modulation distortion. I have not heard the LXMini, but I am a big fan of the larger LX-521. Here’s where I believe they differ: The LXMini operates as a cardioid over part of its frequency response, relies on electrical crossovers and correction from digital signal processing, requires 4 channels of amplification, and can be subject to pipe resonance if not constructed properly. They also may have less dynamic impact due to the small-ish drivers that are used (just physics).

The Hybrids are true dipole loudspeakers, as they operate with dipolar radiation throughout their frequency response similar to the LX-521. There is no electrical crossover on either driver (phase shift), no need for digital signal processing and are driven by a single stereo amplifier. The drivers are larger and can handle more excursion (higher SPLs) before distortion becomes audible. I wanted a system that allowed <source -> amplifier -> speakers> and nothing else in the chain. The Hybrids can also be placed like a bookshelf speaker, provided they have sufficient distance from walls and significant boundaries. (The stand is not an integral part of the design the way it is with the LXMini).
 
While I like the Linkwitz designs, I am not a fan of the MiniDSP that is often used with the speakers. With the LX-521 that I heard, I much preferred the ASP (analog signal processor) option as I didn’t feel like the ASP became the limiting factor in the design. Of course, the ASP was more expensive by comparison.