Fostex T50RP Incremental Mods and Measurements

[bluemonkeyflyer]

Quote:

The other evening, I had a need to make a small custom cable, so I went ahead and tried to source part of the cable with current materials that I had on hand.  I reached for a spare Fostex locking cable that is the standard cable that comes with the T50RPs.  To my surprise, it was purely junk when I had cut it open.  The three conductor wires were literally the thickness of a piece of thread.  Each of the wires appeared to be coated with some sort of resin to insulate them as they didn't have their own dedicated insulation material covering each individual wire.  Then, all three wires were just embedded in the rubber cable itself when the cable was molded.  So, essentially, 98% of the standard Fostex cable is just the rubber component itself.  The other 2% is the meager wires that they have used for the cable itself. I've seen such a cable practice be used in the past in some of the lower end Sennheiser headphones. It's sad when they use such junk to make cables for consumer goods.
 
Here is a picture that I recently took of what I found:
 
 

 
wje,
 
Thanks for the photo. Those wires are tiny. It's surprising to me that they work as well as they do. I compared the stock cable with the V-Moda cable and cannot hear any improvement from V-Moda. I actually like the stock cable better because it's tough and doesn't tangle like the V-Moda cable. I have not tried any DIY or after market boutique cables but would like to.

 

[onefatsurfer]

The Fostex cable is pretty bad.  It doesn't really snap into the jack, which annoys the crap out of me. I'd like to find a decent replacement cable, but not sure what I should use.  I'll probably end up making my own at some point.  I also preferred the stock cable to the vmoda in terms of use, and heard no difference in sound

 

[micmacmo]

It think it's part of Fostex's just-good-enough design philosophy. The cable definitely isn't overbuilt or over-engineered, but it doesn't underperform. I recabled a set of T50RPs to dual-entry with Canare cable and didn't notice an audible improvement in sound quality. Granted, my ears are leaden, not golden, and YMMV.

 

[Armaegis]

Considering that your signal runs through micron thin pcb traces and voice coil wire (in headphone drivers) is almost too thin to see with the naked eye, those wires look just fine to me. 

 

[wje]

Quote:

Considering that your signal runs through micron thin pcb traces and voice coil wire (in headphone drivers) is almost too thin to see with the naked eye, those wires look just fine to me. 

 
I do hear what you're saying about the traces of wires that are embedded within the PCB.  However, it just seems that three neekid wires plastered with a light coating of some sort of shellack and then embedded in a rubber housing isn't what I'd call optimal.  I'm not a cable snob, by any means.  However, for about $20.00 - $25.00, one could easily make a very solid cable using Neutrik connectors and Canare mini star quad (26 gauge) wire that would beat the pants off of the stock, as well as the V-Moda cable.  The main benefit being with using the Canare cable, you have a nice, insulated braid that encases the conductor wires, and then the braid is covered in the rubber jacket.

 

[Armaegis]

I'm guessing it's just a price and durability thing. Pulling three wires through a rubber extrusion mold gives you something quite tough for probably pennies to the foot. 

 

[eltocliousus]

Just tinning the small wires when soldering seems to do the trick, although my soldering job isn't great, and the ground has a cold joint that I'll inevitably have to fix, I managed to solder it to a 3.5mm 3-pole REAN jack without any problems, but it is tricky.
 

 


I managed to tune DBV#3 for Staxpads, or so I believe, I went through a number of combinations, different cotton/fiberglass sizes, cup vents sealed/unsealed/felt overlay. After all that, just adding cotton to DBV#3 does indeed give the best results, taming the increased bass of the Staxpads, Occam's razer.. 

 

[Nevod]

There's a bit on REW's CSD plots I don't get. They all show decay above 100ms, even for the fastest mods. However, measurements done by other systems, as in purrin's measurements, have decay under 1ms on some phones which are supposedly similar. Is there something incorrect with REW's measurements? I understand that the home rig is not as sophisticated as proper equipment, but that difference is too high, IMO, to be explained by that, I suspect methodology differs somewhere in the chain..

 

[micmacmo]

Quote:

There's a bit on REW's CSD plots I don't get. They all show decay above 100ms, even for the fastest mods. However, measurements done by other systems, as in purrin's measurements, have decay under 1ms on some phones which are supposedly similar. Is there something incorrect with REW's measurements? I understand that the home rig is not as sophisticated as proper equipment, but that difference is too high, IMO, to be explained by that, I suspect methodology differs somewhere in the chain..

 
I've noticed that too and think it's either 1) the methods use different sound pressure thresholds/gains or 2) they're measuring something different entirely. In the case of 2), the REW measurements include resonance created by the cup. Perhaps Purrin's measurements only measure the driver? I'm not sure.
 
Here's something else. Compare Purrin's measurement of stock T50RP and his measurement LFF-modded T50RP Paradox (note that these are links to Purrin's site, not to Head-Fi). Granted, the waterfall plots of the Paradox are smoother, but the decay isn't hugely faster than the stock T50RP. The biggest difference is the general "flatness" of the Paradox. So I'm not sure why people say there's a strong ringing in stock T50RP. Purrin's measurements don't show a dramatic ridge anywhere that isn't found in both the stock and Paradox flavours. Any idea on what I'm missing in my interpretation?
 
EDIT:
BTW, getting back to your original question about the validity of REW waterfalls. I'm not sure they're trustworthy. I use the REW setup and almost all of my waterfall plots show similar decay rates, regardless of mods. 

 

[Nevod]

REW has just been updated with a new version, with improvements on CSD generation. I've tried it, and had a relatively successful result, by setting plot time to 6 ms, time window to 6 ms, and rise time to 0.1 ms. Although these parameters should have been in previous versions as well, as some people were already able to get proper plots with HPs..
 
Here's a plot made from DBV #2 mdat measurement. Note that floor is -60db, as I've not been able to set vertical axis properly (not very user friendly stuff..), so it's actually better than it seems. Note that most of decay above 1 kHz is below -40db by 1.2 ms already.

 

[wje]

I received my J$ Denon leather ear pads yesterday.  I haven't tried them on the Fostex headphones, yet.  I initially used them on my HifiMAN HE-400 though. I believe the J$ leather Denon pads are roughly in the same league as the Stax "007" pads that have been quite a good upgrade for the Fostex headphones.  The Stax 007 pads are approximately $150 to buy - new.  However, the Denon J$ pads are only $75.00.  Both prices don't include shipping costs.

 

[bluemonkeyflyer]

Quote:

REW has just been updated with a new version, with improvements on CSD generation. I've tried it, and had a relatively successful result, by setting plot time to 6 ms, time window to 6 ms, and rise time to 0.1 ms. Although these parameters should have been in previous versions as well, as some people were already able to get proper plots with HPs..
 
Here's a plot made from DBV #2 mdat measurement. Note that floor is -60db, as I've not been able to set vertical axis properly (not very user friendly stuff..), so it's actually better than it seems. Note that most of decay above 1 kHz is below -40db by 1.2 ms already.

 
Nevod,
 
Interesting. Are you using Windows, Linux, or Mac?
 
I need to download REW V.5 for Mac and see if I can get it to work for me. Did you use my .mdat files stored in Google Docs or did you make your own measurements?
 
Thanks for the post.


Edit: All Incremental Mods and Measurements were made with REW V.5 for Mac using REW default settings. All measurement data are available in my Google Doc files. Anyone can analyze the data using REW with finer Time and Window Time settings, as you have done.

 

[eltocliousus]

Quote:

I received my J$ Denon leather ear pads yesterday.  I haven't tried them on the Fostex headphones, yet.  I initially used them on my HifiMAN HE-400 though. I believe the J$ leather Denon pads are roughly in the same league as the Stax "007" pads that have been quite a good upgrade for the Fostex headphones.  The Stax 007 pads are approximately $150 to buy - new.  However, the Denon J$ pads are only $75.00.  Both prices don't include shipping costs.

 
I'm still trying to tune the o2 pads to DBV#3, I just can't get it right, I was certain I had it until I directly compared to the thin/airy HE-6s on the Beta22 3ch, Stax pads just add too much bass, I'm really running out of ideas, BMF gave me a really good list of what to try and I'm going to spend a few nights trying to get this just right
 
Let me know how yours go as the J$ have similar dimensions/depth/flange to the o2s, or so I can assume looking at photographs.

 

[bluemonkeyflyer]

Schiit Magni + Modi:  A New Contender?
 

I received these little jewels today. They look sharp with the amp stacked on top of the DAC. I connected the DAC to my MacBook Pro via a printer USB cable I had laying around. I bought a $20.00 pair of Schiit PYST RCAs to connect the DAC to the Amp. The DAC is Plug and Play but I had to go into "Audio Settings" and select 'Speaker-Schiit USB Audio Device' to make it active.

 

What follows is, of course, IMO & YMMV.

 

The sound quality from my Magni + Modi is very good, especially considering the price at only $99 each. I don't hear any coloration. There is no channel drop-out or imbalance at the lowest volume setting. Bass is tight and fast, midrange is clear, and treble sparkles nicely. I think this combo competes head-to-head with the O2 + ODAC. The Magni + Modi costs a bit less and the form factor is better than the O2 + ODAC. The power plug, USB port, and RCA interconnects are on the back of the Magni and Modi which leaves the front of the units "clean" and uncluttered. There is no gain switch on the Magni which may be a factor when using IEMs. I have not yet tried my IEMs with the Magni. 

 

The amp is plenty powerful enough with demanding tunes played through my modded T40RP mk II's. I expect my modded T20RP mk II's and T50RP's, modded vintage T10 and T20 v.2's, vintage TDS, and LCD2 will play at similar SPL's. Note that modded T20/40/50RP's are less sensitive than stock. Zero volume is at 7 o'clock. Comfortable SPL for me has the Magni volume pot set at 9  to 10 o'clock which generates approximately 75 to 80 dB. 11 o'clock is about as high as I want at about 85 to 90 dB. 12 o'clock on the volume pot produces around 90 to 95 dB. Full volume is at 5 o'clock....don't go there unless you are using high impedance headphones.

 

I'm at my office and have not A/B'd them against my Dacmini which is at home. I don't know if I will be able to reliably tell them apart. I'll have to recruit my wife to do the switching between my Dacmini and Schiit combo. News at 10:00...

 

O.K., Now for the obligatory word play on the company name: This is my first "piece of Schiit gear" ...  "this is some really good Schiit!"  

 

EDIT:  A/B comparisons and impressions between my Dacmini (with 1 ohm output impedance mod, providing DAC and Amp duty) and my Schiit Magni + Modi Combo....My wife took care of the switching from one to the other. I was actually blindfolded and used the same modded T50RP headphones which I kept on my head in the same position across comparisons. My wife changed the stock cables whose 1/4" TRS plugs remained in their respective amps. Sometimes, she removed the locking plug from the headphone and pretended to switch to the other one but, in fact, re-inserted the same plug/amp. Otherwise, she randomized the amp order and took notes about what I told her I heard and my guess about which amp was used.

 

I selected some of my favorite test tunes that I know very well for these comparisons, choosing certain segments from each. These recordings are high quality, complex, and detailed from bass through the mids and into the treble range:

 

Patricia Barber's "She's A Lady", "Constantinople", "The Thrill is Gone", and "Too Rich For My Blood"

Donald Fagen's "On the Dunes", "Morph The Cat"

Yo Yo Ma's "Atta Boy"

War's "Summer" and "All Day Music"

Sting's "La Belle Dame Sans Regrets"

Diana Krall's "I Don't Stand A Ghost of A Chance"

Metallica's "Enter Sandman"

Radiohead's "Jigsaw Falling Into Place"

Steely Dan's "Negative Girl" and "Gaslighting Abbie"

Norah Jones' "Stuck", "I Wouldn't Need You", and "Back to Manhattan"

Stan Getz' "One Note Samba" and "Desafinado" with Charlie Byrd

Miles Davis' "So What?"

Natalie Lungley's covers including "Change in the House of Flies", "The Greatest", and "Between the Bars"

 

The bottom line is that I reliably identified the DAC/amp used across these test tune segments significantly greater than chance. The Schiit combo is very good, especially for its priceerformance. If I never listened with anything else, I would be happy with them. My Dacmini, however, wins in several areas including soundstage width and depth, treble detail retrieval and clarity, and vocal/acoustic smoothness creating a more lush quality. To me, the Dacmini sounds better but costs 3.5 times more than the Magni/Modi. Is the extra expense for the Dacmini worth it? For me, yes.

 

I bought the Magni + Modi to give to my son for Christmas. I'm certain he will love the Schitt outta this gear. (There, I got it in.)
 

 
 

 
 
 

 
 
 
 

 

[Nevod]

Quote:

Nevod,
 
Interesting. Are you using Windows, Linux, or Mac?
 
I need to download REW V.5 for Mac and see if I can get it to work for me. Did you use my .mdat files stored in Google Docs or did you make your own measurements?
 
Thanks for the post.

Edit: All Incremental Mods and Measurements were made with REW V.5 for Mac using REW default settings. All measurement data are available in my Google Doc files. Anyone can analyze the data using REW with finer Time and Window Time settings, as you have done.

It's Windows version, V5.01 beta 10, there is also beta for lin/mac, but they're completely untested, as author says.
 
Have you properly level-matched Magni and Dacmini during your test? Safe margin for level matching is only 0.1dB.
I'm personally though more inclined for a very high-feedback gear, NFB takes care of any errors for sure, while in low-NFB stuff, there are some elusive things..
 
Yes, I've used your measurements.