Periodic Audio Mg, Ti, and Be In-Ear Monitors

Periodic Audio launched last May and released their first three IEMs at CanJam NYC. I got to chat with founder Dan Wiggins and learned that he worked on the Microsoft gaming headphone and, interestingly enough, the first few Beats headphones when Beats were designed by Monster. He also co-founded Doppler Labs and was previously Principal Transducer Engineer at Sonos and Chief Transducer Engineer at Blue Microphones, among others.

The three in-ear models debuted were:

  • The Mg ($99) - high-magnesium content magnesium/aluminum alloy transducer
  • The Ti ($199) - pure titanium transducer
  • The Be ($299) - pure beryllium transducer

The bodies are all made of polycarbonate and the only difference between the three is the metal used for the transducer, which is a single wideband dynamic transducer design, eliminating the need for a crossover. The polycarbonate body construction makes them extremely light (9.2-9.8 grams, depending on the model) and manageable in extreme temperatures.

Another cool thing: each IEM comes with small, medium, and large tips (silicone single/dual flange, memory foam). After a few months of living with them, customers can reach out to Periodic to let them know which tips worked best for them and receive a few complimentary sets of tips. You don’t often find this level of service at the $99 price point. Cool.

It’s quite impressive considering that only eight months have passed since the company’s conception. Pre-sales will start on February 15 and shipping will begin in early March.

Their promotional material looked very appropriate for Inner Fidelity:


Here's the video of Dan introducing Periodic Audio:

Watch on YouTube here.

detlev24's picture

...only by changing the material of the diaphragm? Can somebody please explain?

I would have tipped on purposely-different tuned IEMs, just by looking at the promotional material. Anyways, they really were quick in going through the whole process of (developing; up to) releasing a new product... :)

Edelweiss's picture

I suppose the main motivation behind using these "exotic" materials is, discarding marketing reasons, to reduce partial vibrations which naturally occur in any material at higher frequencies. By this logic, using "stiff" materials would reduce the uncontrolled wobbling of your driver. Since stiffness is related to the geometry of your driver and far too complex for internet conversations, we will content ourselves with the elastic modulus, which you can find on the wikipedia pages of the materials.
Continuing this train of thought, using only the stiffest materials would produce the purest sound, since the transducer would have the least uncontrolled vibrations. This is a fallacy: Stiff materials tend to have far more mass than the usual diaphragm foil, and the high inertia of the driver reduces other parameters like impulse response.

With that information in mind, the far stiffer and lighter Beryllium should have a potentially more even high frequency response than the relatively heavy titanium, the information on their website is wrong in that department.

Though using these shiny materials is great for marketing and seems promising for enthusiasts, even the most rigid materials like diamond start to develop partial vibratrions at high frequencies and are therefor no remedy for the dynamic principle.

When purchasing products using beryllium, keep in mind that the production of it is highly hazardous for involved workers due to it's irreversably damaging effects on the lungs.

detlev24's picture

It seems very likely, that the three products differ not only in their diaphragm material. It is all about marketing, as at the latest Beats taught us! ;)

Comparing CVD Diamond vs. Beryllium, the latter material seems more balanced concerning the specific use for diaphragms. Density [g/cc] 3.5 vs. 1.8, velocity of sound [m/s] 16200 vs. 13000, Poisson's ratio 0.31 vs. 0.08. In other cases, polymer diaphragms, e.g., 'Teonex' [some JBL compression drivers] seem to work even better.


Edelweiss's picture

As far as I know, poisson's ratio is not relevant for transducer design, since we neither stretch nor compress the material very much. The same goes for the velocity of sound, since the transducer only moves at the frequency but isn't actually the medium sound travels through. To eliminate partial vibrations, the elastic modulus is important: for titanium it is 110 GPa, Beryllium 287 GPa and for Diamond an astounding 1050+ GPa.

Also, I did not mean to say that they lie about the products! I just think that you don't necessarily get better results from a beryllium driver than from a high performance diaphragm foil from the big brands. It always looks plain if you open your headphones, but a lot of time and consideration goes into the development and production of these foils.

detlev24's picture
Dan Wiggins's picture

Absolutely! It's a combination of several things:

- Mass of the diaphragm
- Internal damping (loss within the diaphragm)
- Stiffness
- Electrical conductivity

Mass of the diaphragm obviously affects not just the Fs (resonant frequency) of the transducer but the Q of the transducer around resonance as well as the overall sensitivity. The first two - Fs and Q - affect the response around resonance +/- about 1.5 octaves. With an in-body Fs around 2 kHz, that means the range around 700 Hz 6 kHz is strongly effected by the Fs and Q of the transducer.

Internal damping is also a rather important one, and something we hope to bring further forward into the conversation about headphones. Ever looked at a waterfall plot? That is a plot of the decay of the driver - how the frequency response changes over time as the sound is nominally "stopped". A poorly damped diaphragm (like a mylar unit, or a driver built on the concept of exceedingly high resonance like a balanced armature) has long "ridges" of resonance that can carry on for tens of milliseconds. These may not show up in the FR - but they are definitely audible.

Stiffness. A diaphragm does not move uniformly! This is why many people pursue planars and electrostats - rather than having a force applied over a thin ring (voice coil to diaphragm), the planar and 'stat apply the force of a larger area (planar) or the entire area ('stat). If your material is infinitely stiff, then the entire diaphragm moves as one. If it is not infinitely stiff, then the first part that starts moving is the part directly above the voice coil, then outwards in both directions. It takes a finite amount of time (which can reasonably be predicted by the speed of sound within the material) for the motion to propagate across the diaphragm. If the time it takes is too long, you get measurable phase lag between different parts of the diaphragm - and that leads to cone breakups and issues. This is most often seen quite easily (and very low in frequency) with compression drivers, which are about the closest cousin to an IEM dynamic transducer (in terms of construction and loading on the transducer).

Electrical conductivity. More conductive, the more eddy currents are induced in the diaphragm, the better damped the driver will be and the lower the inductance. It's measurable and audible. Shorting rings are normal in high end transducers - and with an IEM, because of the close proximity of the diaphragm to the voice coil and all other parts, it becomes predictable and very apparent in production. This is only an issue for metallic/conductive diaphragms, however.

All three units share the exact same body parts, motor parts, everything except:

- Diaphragm material (which runs for inexpensive to buy and work AZ31B Mg alloy to crazy expensive to buy and even more to work Beryllium)
- Color of the PVD applied to the rear cosmetic MIM cap on the main body (so you can tell them apart)
- Color of the PVD applied to the rear port tube on the top (to match the MIM cap)

Basically, it's either a cosmetic PVD on two parts, or the diaphragm. Everything else is the same - and the measurements AND audible differences are dramatic.

Last thing - we don't smooth ANYTHING in our measurements. We use an IEC 60318-4 compliant coupler, zero smoothing. We do have some waterfall plots we'll be putting up later this month - including some from IEMs from other manufacturers. Looking at what happens in time, 15-20 dB down in level, is pretty surprising!