A DIYers Delight: The Fostex T50RP Mk3

This story originally appeared at InnerFidelity.com

The Fostex T50RP Mk3 ($159) is the latest iteration of this famed planar magnetic family of headphones that reaches back to 1978 with the introduction of the original, but fairly different, Fostex T50. The more similar T50RP however, first introduced in 2002, bears a strong family resemblance.

These headphones have long been a staple at recording studios due to their clear rendering of the mid-range, and bullet-proof construction. But they've also found a home with headphone enthusiasts as they respond very well to modifications. There are even a few companies that commercially produce modified versions of this Fostex headphone. (Enigmatic Audio; ZMF Headphones; Mr. Speakers)

The new T50RP Mk3 is, to my ears, much better sounding than its predecessor, delivering excellent tonal balance and good clarity. However, this is not a headphone that will gain wide popularity for three important reasons:

  1. I wouldn't call them ugly, but their utilitarian style is about as appealing as a dump truck.
  2. They are extremely inefficient and cannot reach a solid listening level using most portable players and smartphones. You must use some sort of dedicated headphone amp, or A.C. wall powered device with a strong headphone amp built in to get solid listening levels.
  3. They lack sub-bass. People love their bass these days, and the T50RP rolls off significantly below 100Hz.

If you are not a hard-core headphone enthusiast interested in modifying headphones, you'll likely be better served by the Audio Technical ATH-M50x, which is fully sealed, about the same price, has better bass response, and is likely nearly as durable.

If you are a headphone enthusiast looking for an entry level open planar magnetic headphone, but are not interested in doing headphone modifications, you'll likely be better served by the HiFiMAN HE400S ($299), which has a somewhat more refined and open sound.

If you are an audio pro accustom to the T50RP Mk2, you will likely find the Mk3 version much better sounding, just as durable, but slightly less efficient than the Mk2 model. If you liked the T50RP Mk2 then, you'll probably like the Mk3 model even more now. But I still think you should check out the Audio Technica ATH-M50x as a slightly better sounding (mostly in having better bass response) and sealed alternative.


If you are a headphone enthusiast who's interested in playing around with headphone modifications, the Fostex T50RP Mk3 is a terrific place to start—or continue—your headphone modifying activities!

This article is less product review and more about the T50RP Mk3 and the differences between it and the previous iteration, written specifically to provide some initial information about innards of the model, and to posit about what modifications might be useful.

But before I go on, I need to thank Dan Clark at Mr. Speakers for, on very short notice, sending me a T50RP Mk2 to dissect and compare with the new model in this article. Thanks, Dan!

Differences Between the T50RP Mk2 and T50RP Mk3

In the above and following pictures, the older version of the T50RP will be on the left, the new Mk3 version on the right. You can see in the photo above that the two versions remain quite similar looking. As we'll see in more detail below, other than the graphics the two capsule housings are identical. This brings both good and bad.

The bad news is that Fostex did not change the locking jack on the left ear piece. This jack is reliably unreliable. I would say fully half of all T50RPs I've had here in the lab have displayed some sort of intermittent connection while in use. Sometimes it cuts in and out and a simple twist of the connector will clear it up. But often it will persist as an intermittent connection, cutting in and out seemingly at will or, amazingly, randomly lowering the volume in one ear. (How can it even do that!?) This is a well documented problem in the enthusiast community, and frankly I'm quite disappointed Fostex didn't address it in the product refresh. Fortunately, the headphones are quite easy to disassemble, and jack replacement—or jackless connection by soldering the cable directly to the driver terminals—is fairly straightforward.


The above picture shows the new T50RP Mk3 has a new padded headband as opposed to the rubber molded over a metal strap headband of the previous generation. You can also see the earpads of the Mk2 are attached by a flap that goes around the capsule housing, while the Mk3's pad edge is inserted into a gap between the baffle plate and capsule housing. This is an important observation, as we'll soon see, because the air-gap between the baffle plate and capsule housing serves as an acoustically tuned vent.

Here's a close-up of the pad edges.


Speaking of pads, here's a quick look at them.


Other than the fabric screen of the Mk3, the pads are quite similar. They're about the same size and feel, which means that they are a bit thin and can get uncomfortable in long listening...especially if you wear glasses. The new pad also uses a slightly different type of foam internally. The good news is that they can be fairly easily replaced with a variety of after-market pads. I will link to T50RP threads in the "Resources" section at the end of the article, which delve into the various replacement pads and how users perceive performance changes.

Once the pads are off, we can have a look at the front side of the baffle plate.


The baffle plate is the major change from the previous Mk2 model. The Mk2 (left) has a permanently attached screen/filter over the driver; the Mk3 has a recessed driver with a foam piece mounted in front. You can see a small hole in the Mk2 driver at about the 6:30 position. This vent does not exist in the same way on the Mk3.

The Mk3 baffle plate is actually a sandwich of front and back plates that enclose a space between. Here's an edge-on photo of the baffle plates.


You can see that the Mk3 plate (right) has openings that essentially vent the edges of the foam insert. I suspect DIYers may want to play around with occluding these holes in varying degrees in conjunction with pad selection to control and linearize bass response.


Here is the rear of the baffle plate. You can see the small vent hole of the Mk2 just to the right of the bottom corner of the driver mounting square. On the Mk3 baffle plate you can see to sections of filter paper at the top and bottom of the plate. These cover holes that exit into the space sandwiched between the front and back parts of the baffle plate. These holes probably serve the same purpose as the small hole on the Mk2 to control bass by using some of the signal from the back of the driver to cancel out the signal in front of the driver at low frequencies. But the Mk3 looks to me like a bit more sophisticated acoustic circuit than the simple hole of the Mk2.

Here's the rear of the baffle plate with drivers attached.


Fostex claims the drivers themselves are identical but, as you can see, the filter paper on the back of the driver is different. In the T50RP modding thread there are numerous accounts of changing the density and configuration of this filter. From my reading, if you reduce the density of the filter, you allow air to more easily egress from the rear of the driver making it easier to perform the large excursions and therefor increase bass response. The trade-off, I suspect, is that diaphragm damping is reduced and may increase resonances and cause treble response problems. Again, the experts are in the modding threads.

Here are some close-ups of the driver itself.



From what I can tell with visual inspection, the capsule housing itself remains unchanged in the two models. Here's a picture of just the capsule housing.


In the lower right quadrant of the inside of the housing you can see some scuffing from the mold tool. This scuffing pattern is the same on both housings which leads me to believe both parts came out of the same mold and are identical. This might be very good news for folks who have already heavily modified a pair of T50RP Mk2 capsule housings. (DIYers will often add putty or epoxy to the inside of the housing to stiffen it and reduce stray vibration.) I think it may be possible, if one so chooses, to simply transplant the baffle plate from a new Mk3 model into the housing of a modified Mk2 housing.

Inside the T40RP and T20 RP Mk3 Models
Just to round out this exploration of the mechanical differences between models, here are a series of photos of the T20RP Mk3 and T40RP Mk3. The 20 is on the left, and 40 on the right in each photo.





The only difference I was able to see between all three models on visual inspection was the density of the paper filter over the vent to the outside of the capsule housing. (Circled in red in the above photo.) The T20RP Mk3 is advertised as the "open" model; the filter over the opening is very thin and permeable. The T40RP Mk3 is "closed" and essentially has a piece of impermeable tape over the hole. The T50RP is considered "semi-open" and has a filter with permeability somewhere between the other two.

Sonic Evaluations
To my ears, the T50RP Mk3 is a much better sounding headphone than its predecessor. The Mk2 sounded quite mid-emphasized with a rolled off treble and bass response. The Mk3 sounds much better balanced, erring mainly with a very slight sizzle in the treble, and loss of bass below around 100Hz.


The above graph shows the averaged raw frequency response plots of the left channel for the Mk3, Mk2, and Mr. Speakers Alpha Prime—a heavily modified T50RP Mk2.

To my eyes, the Alpha Prime has the best low frequency response with a 2-3dB rise in the bass below 200Hz. The Mk2 is flatter and better extended than the Mk3, but I very much like a slight measured bass emphasis and much prefer the meatier sound of the Mk3 to the Mk2 even though it's not as well extended.

If I were modding the Mk3, I might try to reduce the filter resistance of the baffle plate vents from the rear chamber to slightly reduce the bass emphasis at 100Hz, and then try to extend and reinforce the bass with better pads to get a similar response to the Alpha Prime.

The gentle rise from 400Hz to 2kHz of the Mk3 is highly desirable, in my opinion, and is quite similar to the Harman Target Response. My guess is achieving this rise is a tricky combination of a number of things, and may suffer during modifications. If you're a DIYer with a measurement rig, take care to note exactly what this area looks like on your measurements (they may look quite different than mine on a non-IEC-spec measurement system), and try to retain the profile between 400Hz and 2kHz during your mods.

The peak at about 3.5kHz on the Mk3 is maybe 2-3dB too high (there does need to be a peak here on raw measurements), but the other two headphones quite obviously don't have a high enough peak here. As a result they sound more veiled and laid back compared to the Mk3. The dip at around 6kHz, and peak around 10kHz on both Fostex models are not desirable, and likely contribute to the slight sizzle of the Mk3 model in the treble. The average levels in this area on the Mk3 are much more in-line with what I consider neutral response; the Mk2s sound quite dull and muffled due to the low level in this area. While the Alpha prime might be a little low in level through the treble region, they are also much smoother in response overall. My understanding is that this is a result of very careful changes to the filter paper behind the driver.

The last little bit of information I'll provide DIYers is a small .pdf booklet containing numerous measurements of the T50RP over time and measurements of DIY modified headphones based on the Fostex model. And with that, I'll point you towards the various sites and threads about this historic headphone and its modification.

Fostex International product pages.
Head-Fi T50RP modification index. (A great resource page with tons of links for DIYers!)
T50RP Mk3 impressions on Head-Fi and SBAF.

USA Distributor: American Sound and Music
925 Broadbeck Drive Suite 220
Newbury Park, CA 91320