Home > General > Inside FiiO D3

Inside FiiO D3

More on Fiio D3

  • Power supply tweaks: [link]
  • Fiio D3 Mods: [link]

Update 10/26/12

3.3V or 5V OPERATION?

According to the datasheet, the Cirrus 4344 DAC can operate at 3.3V and at 5V. At 5V operation, you get 2 dB better performance pretty much across the board.

The current operation of the D3 (original model), seems to be operating at 3.3v since the input voltage to the device is 5V.

However, by measuring the voltage on the power pin of the DAC (VA, Pin 9), the DAC is actually operating at 4.8V which is pretty much ideal for best performance.

In addition, the output stage opamp is also operating at the same 4.8V. Therefore, there is no need to do any modding to get better performance.

But a good and easy mod here is the input capacitor. Space is very tight, must use similar sized capacitor. Used a standard class ELNA 1000 uF capacitor, bypassed with a 22nF film cap. The original was 470 uF, 16V.

Detail after removing existing capacitor

After the large capacitor, power goes through a ferrite, then it is bypassed by two small value ceramic capacitors and connects to V+ in the opamp

Detail after replacing capacitor and bypassed with film cap

Update 8/9/12:

NEW VERSION: D3K TAISHAN

WM8805 replaced by Cirrus CS8416

As FIIO shared when they first introduced the D3, the WM8805 (perhaps) has been discontinued and they have replaced the chip with the CS8416. This is the new “D3K TAISHAN” model. (Taishan as in “Mount Tai”, a famous place in China where emperors used to worship the heaven and to pray for peace and prosperity)

A reader shared and internal photo (see comment #40):

Comparing with the original version, the layout has been completely redesigned. The DAC and the opamp remains the same.:

Backside of D3K Taishan edition. Date is 11 months after the original V1 version.

D3V2?

Same reader alerted me that there is “in-between” version between the original D3 and the new D3K. We call it D3V2. This version uses the WM8805 part but there are omitted components in (what I believe) is the power section for the opamp. You can check the comments for comments about the performance of this version.

I’ve made a composite image of the D3 Original and the “D3 V2″ which is shown below:

Here is the backside of the D3 V2:

The case is identical with V1. In the V1, the RCA connectors are silver color. The RCA connectors in V2 are gold plated. (V2 is the one to avoid)

***

Update 2/14/12: More information on the DAC, I2S to External DAC, see end of this post

***

WM8805-BASED DAC FOR $30

I was intrigued by this DAC based on the industry-leading Wolfson SPDIF receiver WM8805. According to the specifications it is capable of 24bit/192KHz operation and it only costs $30. (Even less nowadays)

Just like any good DIYer, the first thing I did was to take it apart…

Main Components

The famous Wolfson WM8805 SPDIF receiver. Even though it is capable of muxing up to 8 SPDIF sources, it is configured in hardware mode and therefore it is configured to have a single input into the receiver. The device has a coax and a toslink input and these are selected with a single pole  switch. Notice also the use of a seam-sealed crystal instead of the more traditional can (most likely for size).

The WM8805 interfaces a Cirrus Logic 4344 DAC (“344″ indicates fixed I2S configuration). This DAC has pretty decent specs at 105 db SNR and -90 db THD+N, a cut above the DACs found in this price range (in reality, you can’t find any DACs at this price range which is a new low). The CS4344 DAC is also found in the Apple Airport Express 802.11n version.

The analog output of the DAC connects to a TI LMV358 opamp

You can see the I2S lines (the 4 diagonal traces), ready to be tapped…

Power Section

The power is supplied by an external 5V DC switching  supply through a standard mini-USB connector. The 5V line (the uppermost trace from the USB connector) connects to a 6.8 (R37) ohm resistor and to the main PS capacitor (470 uF 16V). This provides a first stage RC filtering to the incoming power.

The 5V incoming is filtered through L5 and  further regulated with a simple discrete Zener diode circuit (basically a shunt regulator) to 3.3V which feeds the DVDD (pin 1, digital core suppy) of the WM8805. This is pretty good filtering and regulation. This line also feeds PVDD (Pin 11, PLL supply) filtered through L2, and feeds the Toslink module through L1. Good use of ferrites for noise filtering.

The zener regulated line also feeds the DAC, with a simple cap bypass. The 102 resistor you see in the photo powers the LED to show Power-on condition.  One of the advantages of the 4344 DAC is that it requires minimal external components. In the case of power, it only has a single power line.

I2S to External DAC

Linux-works has connected the I2S lines to a PCM 5102 DAC. In this case, only 3 lines plus GND are needed since the DAC does not require a master clock connection

More on the CS 4344 DAC

The CS 4344 DAC is not a very popular DAC among audiophiles. It is used in the Apple Airport Express (the 10-legged device near the center of the photo), a device not always appreciated by audio enthusiasts.

The New (2012) Airport Express changed to a new DAC:

According to RogueAmoeba:

More pertinent to our customers, the audio digital to analog converter is an all-new 24-bit/192khz Asahi Kasei AKM4430. This chip is similar to the Cirrus Logic CS4344 used in the previous model, but should be a improvement over the Burr-Brown PCM2705 used in the original 802.11g Airport Expresses

The CS4344 is also used in the iQube V2 DAC/headphone Amp. This DAC/AMP was designed by Guido Tent and Bruno Putzeys [link], which are highly regarded people among the audio community.

And also used in the FubarIV Plus [link].

More on the OPAmp

A user reported changing the opamp to a AD8656 with great results. The AD8656 is also a recommended component of the Gamma-2 DAC  and it can be used to drive headphones directly. The OPA2209 is also another pin-compatible rail-to-rail opamp with even better noise specification than the AD part. It is used in TPA’s Trident shunt regulator.

Comparison of TI LMV358 with AD8656 and with OPA2209

Parameter TI LMV358
AD8656 OPA2009
Input offset (max)
9 mV 0.25 mV
0.035 mV
CMRR (typical)
65 dB
100 dB
120 dB
Input Voltage Noise Density
39 nV/sqrHz
4 nV/sqrHz
2.2 nV/sqrHz
slew Rate
1 V/usec
11 V/usec
6.4 V/usec
Price
~$.95
~$3.40
~$4.29

 

Additional Reading

An excellent report on the device with high resolution photos (very nice photos), extensive measurements and listening impressions [link].

  1. 2011/06/25 at 05:29

    Great pics! Not exactly an ‘audiophile’ product with that opamp but I can see it becoming popular with diy/tweakers as a quick and cheap way to get access to decent (low-jitter) I2S. The WM8805 has a very good reputation so that alone could win it many friends.

  2. BlogGeanDo
    2011/06/26 at 05:58

    Hello Richard, Good enough for $30! :-)

  3. 2011/06/27 at 02:42

    Yeah, its a really attractive price. A Taobao kind of price :) Excellent find.

  4. 2011/06/27 at 02:46

    Yesterday though I found a DAC on Taobao which is even cheaper still. Its only a PCB and seems to have no power supply connector. But its under $8, so who can complain?

    http://item.taobao.com/item.htm?id=5096898989

    The guy selling it is in my city, so perhaps I need to befriend him :)

  5. BlogGeanDo
    2011/06/27 at 04:53

    Are you trying to save the 12 yuan shipping? :-).
    Yes, give them some tips on building better kits…

  6. 2011/06/27 at 11:45

    Yeah I was thinking maybe if I design something, he might build it for me very cheaply…. he must by components at amazingly cheap prices to sell a DAC for 49.5rmb

  7. goldkenn
    2011/07/09 at 03:26

    Are you sure the 8805 is congif as hardware mode by only because having a manuel selection input? Since as far as I knoe, 8805 will not sypport 192K and no PLL de-jitter function at hardware mode, but refer the D3 information, it does support 192K and did active the PLL

  8. BlogGeanDo
    2011/07/09 at 15:34

    Yes, it is in hardware mode. There are several indications: there is no uP on board, the pins for i2c are grounded, the switch is a toggle between the two lines…

  9. 2011/09/17 at 06:13

    Really nice article on the D3 there.

  10. niksego
    2011/10/16 at 04:41

    Simple question about the D3 operation. Hope you can help.

    Purchased the D3 to convert digital audio from a TV to analog for surround sound system.

    The D3 powers ok. The red power light comes on upon connection to the USB power supply. Manual says that blue signal light ON means its working. It currently lights up red. I cannot get any sound through my surround speakers.

    However, when I unplug the USB power supply and re-insert it, I see a quick flicker of blue before it turns red. Happens each time I do that.

    I am not sure if the “locked” blue light is supposed to be ON no matter the audio cable connections to and from the D3 or if it only comes on when everything is connected correctly and there is digital audio feed to the D3 and analog audio from it.

    In other words,what are the conditions for the blue “locked” light to come on. And why do you think it flickers when the power is connected? Could the unit be defective?

    Thanks
    Nick

    • BlogGeanDo
      2011/10/16 at 05:48

      Make sure you are sending stereo and not 5.1. The D3 can only receive stereo signals

      • 2011/10/28 at 17:40

        You, sir, are a bloody genius. Thanks.

    • Miguel
      2012/10/05 at 22:42

      Did it work? I was looking to buy it to play music and surround sound from my ps3, but have some doubs if the surround works from optical to rca. I also heard the D3 distorts when selecting other than PCM sound.

  11. Jennifer
    2011/12/06 at 11:19

    I am running this FIIO D3 DAC with a 12v Pyramid 3A regulated power supply ($23 on amazon) along with an LM7805 5v voltage regulator and .1 uF cap on output voltage.

    It sounds MUCH better than the stock switching power supply it comes with. Clearer, more separation, less muddy.

  12. 2012/02/07 at 19:30

    hey there, linux-works here ;)

    I tried your idea of tapping into i2s. success!

    DIY: pcm5102 burr-brown DAC and Wolfson wm8805 spdif receiver

    testing a new (to me) dac chip, the pcm5102.

    in a pinch, and IF you can tap into traces that small, its a good i2s source. not recommended for the fat wires I used, though, lol!

    • BlogGeanDo
      2012/02/07 at 23:35

      Hey linux-works good to see you here. Glad to see you got it to work. I will add your photo to this post (thanks in advance :-)). Also that new preamp of yours, really good looking.

  13. anthony
    2012/02/08 at 05:24

    If anyone is curious in upgrading the TI LMV358 opamp, there are several 8 pin opamps available in the same pin configuration which will work as a drop in replacement. I upgraded mine with a AD8656 which operates nicely at 5v.

    Be careful removing the LMV358 however as initially I was not being very careful and my soldering pads started to break apart. I would recommend just clipping the pins the best you can and soldering the new chip over top rather than trying to remove each pin.

    • BlogGeanDo
      2012/02/15 at 08:26

      Anthony, you got a photo of your mod?

      • anthony
        2012/02/18 at 07:25

        sorry I dont have a pic of the mod. I am planning on modding another one in the near future and will take a picture when doing so.

    • Milos
      2012/02/17 at 17:54

      Wow, thanks for a great idea. Did you notice an audible improvement? If it’s only marginal I won’t bother but otherwise it sounds like a good idea.

      • anthony
        2012/02/18 at 07:27

        I dont have a side by side comparison so it is difficult for me to say. The modded DAC sounds quite good though.

    • Milos
      2012/02/18 at 19:37

      The LMV358′s slew rate of 1V/us looks to be just good enough to handle CD quality audio. Looking at the AD8656 specs, its slew rate is superior (although I don’t claim to hear frequencies above 20 kHz) and its THD+N looks amazing (which is IMO the most important parameter). I will give this a shot. If I manage to damage the board, what the hell… given its price I might just order a new one.

  14. 2012/02/16 at 11:55

    Very good article !
    I own a D3 connected to a little E5 headphone amplifier, and it work very well through TosLink cable. I power the D3 with a single Lipo battery (3.6 .. 4.2 V), the blue led is not so bright, but it work perfect.

    • Tom
      2012/04/28 at 22:32

      Is the sound clearer than with the power adapter?

  15. Ashley Doyle
    2012/03/14 at 13:26

    I did the AD8656 mod using a Chipquik SMD Removal Kit other than having to
    reheat the pins a few times to seat the pins eaiser than i thought.

    • BlogGeanDo
      2012/03/14 at 15:43

      Send me a photo of the mod: hifiduino -at- yahoo

  16. Anonymous
    2012/04/29 at 15:32

    The noise voltage and slew rate specs for the op amp are reversed.

    • BlogGeanDo
      2012/04/30 at 01:39

      Thanks for the catch. Corrected…

  17. Tom
    2012/05/04 at 23:50

    Hello, I have power supply question. I have an old computer power supply laying around which has a 5v lead. Would this be better then the one which comes with the unit? Thanks.

    • BlogGeanDo
      2012/05/06 at 01:31

      Likely not better. If you have a 5V linear supply (those that weigh more because it has a transformer), then you might get improvements

  18. ruin
    2012/07/06 at 08:50

    Did anyone who changed the op-amp notice any difference in the highs, or do an RMAA? Is the roll-off further up the chain?

    • Emerge
      2012/07/10 at 19:54

      I changed the op-amp with AD8656 and did RMAA. The frequency response starts to fall off from 500hz. At 10.000hz its almost -15dB. Something is definitely wrong!

  19. Anonymous
    2012/07/08 at 18:34

    The link to the alternate op amp should be updated to http://www.digikey.com/product-detail/en/AD8656ARZ/AD8656ARZ-ND/998010 to match the package in the d3. The existing link is for the MSOP size.

  20. ruin
    2012/07/14 at 23:05

    Emerge :
    I changed the op-amp with AD8656 and did RMAA. The frequency response starts to fall off from 500hz. At 10.000hz its almost -15dB. Something is definitely wrong!

    Wow… That my RMAA finds the stock starting to drop around 3k, at less than -1dB at 10k, and -3dB at 20k.

    • Thomas O.
      2012/08/09 at 01:03

      Well, I tested my fresh fiio D3 with RMAA and it’s horrible! I really noticed the audio getting muffled when plugging a Lindy A/D and the fiio D/A into an analog chain (ground-free transport over optical).

      Checked with RMAA, I can confirm -15 dB at 10 kHz! By switching around with the converters and the ins/outs of my USB interface (Alesis io|2) I could isolate the fiio as the main culprit.

      Have a look at http://thomas.orgis.org/div/fiio-spectrum.png .

      Since I already did order the AD8656, I would like to know what is causing this extremely bad sound — I have to assume it is not the stock opamp, but something else that’s broken. Something simple like a bad 470 uF cap? Is there a good and a bad production batch floating around? My board differs at least in the color of the RCA block (see http://thomas.orgis.org/div/fiio-board.png).

      • Thomas O.
        2012/08/09 at 09:11

        Quick followup on myself: Updated the pic to show sloppy soldering on the underside and had a closer look at the upside: The color of the connectors is not the important difference, but the lack of the whole circuity near the cap is appaling. No diode, no transistors … What’s up with that? Was that stuff redundant? I don’t see other severe changes in the design. I don’t have much experience with audio electronics, but something tells me that this could be correlated with a now sub-optimal lowpass filter.

        (The 470 uF cap just jumped into my eye, because I had serveral device failures because of bad caps in the past … but it’s not part of the output stage here.)

        Emerge: Is your fiio also of the “D3MO2″ variant and has those vacant spots?

      • BlogGeanDo
        2012/08/11 at 02:43

        My board has those pads populated. I couldn’t figure out what kind of circuitry that was, but I suspect it is to regulate the power to the opamp. I am not sure what is the effect of omitting those components.

      • BlogGeanDo
        2012/08/11 at 02:40

        Thanks for the links to the photos. I just pulled your comment out of moderation

  21. Thomas O.
    2012/08/10 at 22:25

    Hm, my first reply is stuck in moderation, perhaps because of the pictures of rev2 I linked in. I just want to give an update about the current, third, revision: The FiiO D03K. The spectrum looks a lot better for that one; basically flat from 50 Hz to 10 kHz; going towards +-0.5 dB around 10 Hz / 20 kHz.

    A little bummer might be that they changed the receiver chip: It’s a CS8416 now. I also don’t see an external crystal. I’m not sure that this is a downgrade (as many folks seem to believe); that chip costs more than twice as much as the Wolfson around here (and about half the price of the D03K unit).

    Interested in pictures of the board? I’m not adding any links here yet to avoid the void of moderation. And, well, perhaps you’ll want to check out one for yourself anyway;-)

    But the dreadful second revision goes back to sender stante pede!

    • BlogGeanDo
      2012/08/11 at 02:44

      Yes, photos or links would be nice. I read somewhere that the wolfson part was being discontinued, I think that is the reason they switched to the cs device

    • Anonymous
      2012/08/24 at 12:30

      I’m not very happy about the D03K, the sound is missing in details…but I’ve not a direct comparison with previous version. I’m wondering why FiiO has changed something which was working pretty well (at least considering the reviews)…

  22. ruin
    2012/08/14 at 04:27

    Thomas O. :
    Quick followup on myself: Updated the pic to show sloppy soldering on the underside and had a closer look at the upside: The color of the connectors is not the important difference, but the lack of the whole circuity near the cap is appaling. No diode, no transistors … What’s up with that? Was that stuff redundant? I don’t see other severe changes in the design. I don’t have much experience with audio electronics, but something tells me that this could be correlated with a now sub-optimal lowpass filter.
    (The 470 uF cap just jumped into my eye, because I had serveral device failures because of bad caps in the past … but it’s not part of the output stage here.)
    Emerge: Is your fiio also of the “D3MO2″ variant and has those vacant spots?

    So I just opened mine, and I have D3M02 with those pieces missing. I don’t see the loss.

  23. 2012/08/22 at 02:58

    I’ve got one of these (at last!) and I’m using it to feed I2S to my prototype ‘LAID’ DAC. It didn’t sound as good initially as when I was using a DIR9001. But that was running from a linear PSU and the D3 comes with a switcher. Here’s the rub – the switching PSU sucks! Even though I’m not using the analog outs, changing to a locally regulated linear supply makes a worthwhile improvement to the ease of the sound. I’m guessing the SMPSU is spraying common-mode noise throughout my DAC.

    • BlogGeanDo
      2012/08/24 at 16:21

      Hi Richard, hopefully you got the one with the wolfson chip. Thanks for sharing your projects.

      • 2012/08/24 at 16:31

        Yes, that’s the chip I wanted. I was puzzled to hear it was going out of production because over on Wolfson’s website they give no indication at all that its ‘NRND’.

      • BlogGeanDo
        2012/08/24 at 19:39

        I should have said “perhaps been discontinued”. I can’t find the reference, but I do remember that Fiio indicated that the wolfson chip would be discontinued later.

  24. Anonymous
    2012/08/24 at 19:37

    Thanks for this great review!
    Is there any reason to use the E3 between the digital out of my airport express and Audioengine A5s (analog input)? You said they both have the same DAC so it doesn’t seem like the sound would be any different than using the Airport Express miniplug out into the Audioengine powered monitors.

  25. Anonymous
    2012/08/24 at 19:44

    oops I meant D3

    • Fernando
      2012/09/26 at 23:47

      I would appreciate if someone could answer this, I have the same doubt…

  26. 2012/09/27 at 03:18

    The sound you get is only partly determined by what DAC chip you use (assuming its not a multibit chip, which has a rather different kind of sound). Rather its more to do with noise and an external DAC has the chance of having its signal being less corrupted by noise than one in a digital product like the Airport Express. This is especially true if the optical connection is being used.

    • Fernando
      2012/09/27 at 04:19

      Thanks Richard!

  27. Marcelo
    2012/11/23 at 03:15

    This would help me a lot!
    I´ve just find out that I own the D3 V2 (the one to avoid). I´m converting from the AppleTV TOS output to the analog output. I noticed that the sound quality is not too good. The sound does not present a clear quality. The high frequencies are terrible. The question is if I got the original one will I get the sound quality improved. Thanks.

  28. scott17
    2012/11/23 at 18:19

    I have the D3v2. I changed the op amp to the AD8656, and the electrolytic to a Nichicon KA series audio cap bypass with a Wima 0.022uF.

    In my case I also replaced the zener 3.3V regulator with an ST Microelectronics LE33CZ 3.3V regulator.

    If anyone else wants to do this, you must remove the zener as well as the two 75 ohm paralleled resistors that are connected to the cathode of the zener. Then you connect the ground of the LE33CZ to the pad where the anode of the zener was, the “in” of the regulator goes to where the two 75 ohm resistors were, nearest the edge of the circuit board. The “out” of the LE33CZ goes to the pad where the cathode of the zener was.

    So far so good.

    • scott17
      2012/11/24 at 15:45

      After listening to this with the mods, I can say that the stereo separation and “spaciousness” (for lack of a better word), has definitely improved.

      • BlgGear
        2012/11/24 at 16:14

        Thanks for sharing. Any photos posted somewhere?
        I have the V1 with basic capacitor mods. Almost indistinguishable from the analog outputs of a DENON DVD/SACD player through an NAD amp and bookshelf speakers…

      • scott17
        2012/11/24 at 16:48

        I should have taken photos when I did the mods. I’ll open it and take a couple of photos in the next couple of days. I must admit it was a little tricky replacing the op amp.

        How would I post the photos here?

      • BlgGear
        2012/11/24 at 17:37

        Thanks. If you post them elsewhere, I can just grab them or send them to me: hifiduinoatyahoo

    • osexpert
      2013/01/01 at 17:07

      Don’t understand why changing op amp should matter, since the v1 uses the same op amp and has no problem? I have the v2 and all treble is missing. I have done the caps mods but no change (did not expect any either, since v1 uses same caps and don’t have the problem). Wondering about buying the v3 but FiiO doesn’t deserve it when making such a crappy v2.

      • Marcelo
        2013/01/01 at 22:46

        I do have the v2 and the treble is really missing. I bought the v3 and the sound is much better than the V2. I highly recommend buy another one. I was wondering to perform the op amp changes but now, will not spend my time on that. The 1st Christimas at my home with the v3, Apple TV and Bose speakers were great.

      • BlgGear
        2013/01/02 at 18:09

        I have a V1, so no experience with V2 or V3. Perhaps the different perceptions are due to age. I know my kids have a much higher sensitivity for high frequencies…

      • osexpert
        2013/01/04 at 18:59

        I mapped the circuit between dac and op amd and frequency simulation show the same plot as in comment #41: https://www.circuitlab.com/editor/#tu2ujy
        Trial and error’ed new values for the low pass caps until simulation looks good, but have not tested in real life (yet). Have ordered the D3v3, will update circuit when I get it.

      • BlgGear
        2013/01/05 at 04:28

        Thanks for sharing. I like the on-line circuit simulator.

      • osexpert
        2013/01/08 at 20:42

        Desoldered the 4 low pass caps and this fixed problem with missing treble. Removing low pass mean freq. above 20kHz won’t be filtered, but haven’t had any problems.

        http://flic.kr/p/dJDU3e

      • BlgGear
        2013/01/08 at 20:57

        Very nice detective work! Easy fix to those having V2. I’ll do a post about this fix…

      • osexpert
        2013/01/08 at 23:47

        If you wait til I get the D3v3 I will have an updated circuit diagram and hopefully discover correct values for the caps, in case anyone want to take on the challenge of not just desoldering the bad caps, but also replacing them with correct caps:-) (I will probably attempt doing this, just for the heck of it).

      • Marcelo
        2013/01/09 at 00:52

        I did the cap removal and it works very well. Now I have the V2 and V3 and there is no perceptive sound difference between both. The treble is there!!

      • BlgGear
        2013/01/09 at 01:41

        V2 still has the WM8805 which has better jitter specifications…

      • osexpert
        2013/01/09 at 21:24

        Got the D3v3 and I’m completely confused. Treble sounds ok. Circuit seems to be the same as D3v2 and all caps are the same, except the low pass caps between dac and op amp: they were 10nF but is now 5.4nF. I am confused because freq. simulation of the circuit does not improve when changing cap from 10nF to 5.4nF, and I was soo convinced the simulation showed the real thing. I measured the 4 caps I desoldered and found all to be 10nF! I should have known that measuring caps in-circuit is not reliable… I resoldered the 2 caps on the D3v2 that I previously thought was 26 nF, and the treble seem to be ok (will do more listening tests), so I was probably fooled by the freq. simulation. It may seem it’s only necessary to desolder the 2 caps close to the op amp (sorry to those who desoldered all 4). The next and last thing I will do is solder 5.4nF caps on the D3v2 (those 2 caps close to op amp) to check if what I think is correct.

      • BlgGear
        2013/01/09 at 23:55

        A 500 ohm and 10 nF RC filter has a cutoff at 30+KHz… should not be a problem. Thanks for reporting all of this. Very interesting

      • osexpert
        2013/01/10 at 01:17

        5.4nF was another in-circuit measuring error. I desoldered and measured 130pF! That’s pretty close to my trial’n err’d value of 108pF. With the updated values the freq. simulations doesn’t look so bad after all! Updated schema: https://www.circuitlab.com/editor/#tu2ujy
        So it was in-circuit meas. that was the main confusion.
        Final solution is: remove, or optionally replace caps next to op amp from 10nF to 130pF: new picture: http://flic.kr/p/dKdzHP
        I might do the replacement some time in the future to verify.

      • BlgGear
        2013/01/10 at 16:43

        So the caps in the low pass filter into the opamp were:
        V2: 10 nF
        V3: 130 pF
        What about the resistor of the LC filter. The same value for both V2 and V3 at 10 Kohm?

        How about the low pass filter after the opamp? Seems ok with R=500 ohm and C= 10 nF (gives you cutoff at 30+KHz)

        On V2: R=10K and C= 10nF the cutoff is: 1.5KHz! reducing C to 130pF, the cutoff is increased to 122KHz

      • osexpert
        2013/01/10 at 18:26

        <>
        Yes.
        <>
        Yes.
        <>
        Yes. The v2 sounds fine with them soldered back on. AFAIK, both v2 and v3 have 10nF here since I measured the same in-circuit value of approx 24nF on both. I did not desolder on the v3 to verify real value thou.
        <>
        Formulas seem to say 130pF give 122kHz. For some reason the circuit freq. simulation does not agree, as it plot a cutoff at approx 20kHz with this value. Also a weird coincidence that the value I trial and error’d from the freq. simulation was so close to the real value.

        BTW: the circuit schema has most of this info.

      • osexpert
        2013/01/17 at 16:47

        Didn’t find 130pF but RC formula says 31.8kHz cutoff with 10k need 500pF, so I guess anything between 130pF and 500pF is ok. I fitted 250pF and didn’t notice any difference in treble. Game over.

      • BlgGear
        2013/01/17 at 17:17

        Thanks for getting to the bottom of this problem and sharing your results. This post gets a lot of traffic. I am sure a lot of people will be thankful for your info.

  29. Marcelo
    2012/11/27 at 21:15

    Based on the comments what´s the best one? The original V1 or the Taishan V3? As I said, I have the V2 and I´m very disapointed with the sound quality. I will buy another one. V1 or V3?

    • BlgGear
      2012/11/28 at 17:00

      I don’t have experience with the V3, but as you can see in the photos, they have rearranged the circuitry quite a bit, to accommodate the new receiver chip and perhaps optimize the layout. If you care about having the well regarded WM8805 receiver chip, then get the V1. However, at this level of price and performance, I doubt there is a difference in sound between V1 and V3

      • Anonymous
        2012/12/01 at 02:23

        My experience with the D3 V2 is that I did not have any complaints with the stock unit that I received. After the mods it has definitely improved with regards to clarity and soundstage. Even though I have 56 year old ears, I’ve been a musician for about 40 years, I design and build top end hi-fi tube amps, and have been a discerning listener for a long time.

  30. 2012/12/01 at 03:03

    I don’t know why my ID did not appear, but this was posted by scott17.

    Anonymous :
    My experience with the D3 V2 is that I did not have any complaints with the stock unit that I received. After the mods it has definitely improved with regards to clarity and soundstage. Even though I have 56 year old ears, I’ve been a musician for about 40 years, I design and build top end hi-fi tube amps, and have been a discerning listener for a long time.

  31. Anonymous
    2013/01/01 at 22:57

    I have the V2 and mine sounded fantastic when it was stock, even better after the mods.
    Maybe there was a bad production run.

    • formatthias
      2013/01/14 at 21:26

      I modded the dac with ad8656. It is really better than the cheap TI one. Not so midcentric and non detailed.;)

  32. alex
    2013/01/22 at 16:36

    osexpert :
    5.4nF was another in-circuit measuring error. I desoldered and measured 130pF! That’s pretty close to my trial’n err’d value of 108pF. With the updated values the freq. simulations doesn’t look so bad after all! Updated schema: https://www.circuitlab.com/editor/#tu2ujy
    So it was in-circuit meas. that was the main confusion.
    Final solution is: remove, or optionally replace caps next to op amp from 10nF to 130pF: new picture: http://flic.kr/p/dKdzHP
    I might do the replacement some time in the future to verify.

    As a V2 owner with treble problem, I’ve followed your advice and removed the two caps.
    Frequency response is now flat and treble is back. Confirmed by a RMAA test.
    A side effect though: dynamic range went from 92dB down to 82dB according to the RMAA before/after test. Noise level and crosstalk increased also. THD has decreased a bit.
    So was this mod worth it?
    Any suggestions?

    • osexpert
      2013/01/22 at 17:51

      Worth it? With the original 10nF caps the cutoff was 1.5kHz! It’s was like listening to music over a telephone line. If you ask this question I wonder if your FiiO v2 in fact had the problem (several say their FiiO v2 sounds ok).
      When you first have them caps desoldered: can you measure them?

      • alex
        2013/01/22 at 19:52

        Of course, it was worth it! It sounds obviously better now!
        My D3 definitely had the problem and that’s why I was taken to this thread.
        Thanks a lot for your research and advice, BTW :)

        My question was more about the trade-off regarding dynamic range decrease… (According to RMAA measurements, I haven’t noticed this by myself)
        I certainly should have asked: Was this expected/expectable/normal?

      • osexpert
        2013/01/23 at 01:14

        Sorry for jumping the guns. All the differences you measured were unexpected. Removing the 10nF caps should from my simple understanding only let higher freq. thru, but I’m guessing if the dac emit high freq. > 122kHz (can’t understand that it should thou) it may cause the op amp to distort. Suggestions: as said before, FiiO v3 use 130pF, so you can replace with that (or the next larger available eg. 150pF). This will prevent high freq. from entering the op amp and it might improve the measurements.

      • alex
        2013/01/23 at 12:55

        I’ll post the detailed RMAA report ASAP.
        Yes, seems like the OP amp doesn’t like the mod. It sounds definitely better though, I didn’t notice anything bad at all (except the measurements).

        I don’t have anything to measure the capacitance of the removed caps and I don’t think I have the skill and tools to resolder such tiny caps without hurting something around :/
        The removal was the easiest part of it! :)

        Maybe a OP amp replacement like scott17 did would be easier and improve things?

        Anyway, I’m very satisfied with the improvement in sound at this point ;)

  33. 2013/01/25 at 03:42

    That’s all fine and well. Evidently, my D3 was not one of those affected with the treble response problem. Yes, the D3V2 does sound better with the new op amp and VR, However I recently purchased an Audioengine D1 and the difference in sound quality and response is nothing short of phenomenal. Not meaning to trash the thread here, but now I need to figure out how to make the D1 even better.

  34. Anonymous
    2013/02/22 at 17:36

    Just bought the v3 and did the cap mod and changed the opamp to ad8656, sound very good ! Now may I ask where is the DC bypass cap and I want to remove them, because the ad8656 have a lot better DC offset than the TI one so I think it should be no problem, thanks.

  35. Tea
    2013/02/22 at 17:37

    Just bought the v3 and did the cap mod and changed the opamp to ad8656, sound very good ! Now may I ask where is the DC bypass cap and I want to remove them, because the ad8656 have a lot better DC offset than the TI one so I think it should be no problem, thanks.

  36. Ralph
    2013/03/21 at 19:45

    What’s your thoughts on sound quality of D3K? Any mods possible to upgrade the sound?

  37. 2013/04/05 at 05:41

    I just finished modding a D03KM02 with a TI OPA1602AID and a Nichicon KA 1000uF 16V. It’s faster, sweeter and tighter.

    • 2013/04/10 at 02:03

      OMG… I just tried out the mod above with a Fiio E12 and Shure SE-535. It’s like being on stage. The sound is clean, powerful, full, no hiss, just simply perfect. My source is an SPDIF out from my notebook computer playing FLAC files that I purchased from HDTracks.com. This is phenomenal. Hurry and do the same. You can’t lose.

      • BlgGear
        2013/04/10 at 05:58

        Thanks for sharing. That opamp has real nice specifications.

  38. Progmeister
    2013/09/02 at 12:57

    Hi I was wondering if it is worth trying a higher quality linear regulated power supply with the Fiio D3, and also can any one advise what the operating voltage range is as I have a power supply lying around which puts out 4.5v or 6 v which is the nearest it goes.

  39. Keroronsk
    2014/07/01 at 14:28

    Hello. I’ve noticed mine d3 have some high frequency hissing if I set maximum volume on my amplifier. I’ve changed op amp (to OPA2209), replaced aluminium capacitor with tantalum one, and even tried to power d3 from battery pack instead line adaptor, still no luck. What else could be source of this hissing? Thanks.

    • BlgGear
      2014/07/07 at 19:12

      I would guess the ground connections? Is the amp completely silent if you power off the DAC?

      • 2014/07/19 at 08:42

        Well, if I unplug reciever it will be silent. But if I plug in anything (FiiO, cell phone etc) this hissing returns.

      • BlgGear
        2014/08/08 at 04:12

        So the hissing is not particular with the FiiO?. Try using a surge suppressor or power filter.

      • 2014/07/19 at 08:43

        If amp plugged into FIIO and no power at FiiO it also silent.

  1. 2012/12/01 at 17:56

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 208 other followers