MF X-DAC V2 Mod

[dejanm]

I already changed OPAs in this DAC. I put new 2111 and it was great improvement. It is by the way Limited edition of this DAC, which means it has a bit better caps installed (although I can see only these cheepo Jamicon electrolitics and a lot of tantalums on the signal path).

Now I would like to do some more mods of this DAC. My problem is that I do not have any schematic of the unit. Another problem is that I do not have values of these tantaliums .... I cannnot see anything written on them, either.

Can somebody help ?

[Deleted]

dejanm

I have sent you an email.

SandyK

[XTRProf]

So what have you done after Alex email?



Cheers!

[dejanm]

Jun 25, 2008 3:16:58 GMT XTRProf said:

So what have you done after Alex email?



Cheers!

Alex was very kind to send me schematic of DAC output stage that might be what I am looking for. Unfortunately it is not quite the schematic of MF X-DAC V2. It has some similarities, so I guess that it might be perhaps the version 1 of this unit. I do not know ...

Some resistor values are not the same and the topology is a bit different. I am now reverse engineering it but I will probably end up by simply changing caps for better ones. Most of these tantalums at the output stage are: 10 uF, 20 V.

[dotnet]

Jun 26, 2008 10:59:15 GMT dejanm said:

Alex was very kind to send me schematic of DAC output stage that might be what I am looking for. Unfortunately it is not quite the schematic of MF X-DAC V2. It has some similarities, so I guess that it might be perhaps the version 1 of this unit.

If that's the case I would be doubly interested!

Cheers
Steffen.

[dejanm]

by the way - with which caps should I replace these tantals ? Please have in mind the available place which allows some 5 mm at max. BlackGates series N are too large ... Any idea ?

[dejanm]

Jun 26, 2008 11:08:35 GMT dotnet said:

Jun 26, 2008 10:59:15 GMT dejanm said:

Alex was very kind to send me schematic of DAC output stage that might be what I am looking for. Unfortunately it is not quite the schematic of MF X-DAC V2. It has some similarities, so I guess that it might be perhaps the version 1 of this unit.

If that's the case I would be doubly interested!

Cheers
Steffen.

I do not have any problem to pass the schematic to you but send me your eMail address first ...

[dejanm]

I did some reading. It seems to me (if I understood it correctly) that replacing Tantalums is not so straight forward - there are some rules. If the tantalum is on the signal path than it should be taken the same value for the new cap. But if the tantalum is used as a filter, than it should be taken ~ 10 times greater capacitance for the new cap. Is my understanding correct ?

(new caps are certainly not going to be tantalums).

[Deleted]

dejanm
That is my understanding too, but perhaps not quite so important as regards Audio formulation electrolytics used for filtering?
SandyK

[dotnet]

Interesting. I wonder if there's any point in replacing tantalums as bypass caps? I thought they were quite capable at that.

I usually try to get away with low ESR electrolytics near the op amp to deal with ripple and audio frequency spills, and a small ceramic (virtually zero inductance) cap straight across the power pins, to zap any RF energy that may want to enter - or leave - the op amp...

Cheers
Steffen.

[Deleted]

Steffen
In my experience, tantalums and ceramics,as bypass caps, can lead to metallic sounding treble. Dervices like the fast AD8066 sound nicer if the manufacturers recommended 100nF ceramic is replaced with a 100nF poly, PROVIDED that the regulators are reasonably close by to ensure stability.
Alex

[dejanm]

Any cap recommendation ? Tantalums in signal path and used as filters (most of them) are: 10 uF, 20V. The place is though critical ... 5 mm for radial type for example ...

[XTRProf]

Jun 27, 2008 8:31:49 GMT dejanm said:

Any cap recommendation ? Tantalums in signal path and used as filters (most of them) are: 10 uF, 20V. The place is though critical ... 5 mm for radial type for example ...

For signal path, use the best possible. Forget about Tantalum or Ceramic as they are the worst caps to use although they measure well. I concur with what Alex had written on this above. Even the fame John Curl, who had designed Vendatta and now Parasound, don't like them and he's someone who likes to design things with no caps in the signal path!

For my experience on my X-Dac V3 last dc decoupling caps of 100 uf, yes, 100 uf and I don't know why, in the signal path b4 the rcas, to my ears the best sound was using the SUPER E-CAP configuration of Elna Cerafine 15 uf BP and BG NX Hi-Q 0.1 uf BP. Using BG BP as super e-cap will also be possible but the Cerafine is much cheaper and I wanted a sound not all from Elna but also from BG as well and so the mix of Cerfine and BG BP here. Don't let one type of brand sound dominate. I have tried the Solen 10uf MKP and it just sounds not as transparent, spacious and detail as the former combination. The Toflen at 10 uf will be better still but just too expensive for me to stomich. The Teflon will be head and shoulder above all claims of the best caps in the world.

For signal path dc decoupling caps, 10uf will be ideal although higher will be better for bass extension. But definitely not as high as 100 uf as used by MF in my V3! The minimum caps value recommended here will be at least more than 2.2 uf.

In my conclusion, the best e-cap, at least in my application, are from Cerafine -> BG -> OScon, with Oscon sounding the worst. Relatively quite close in type of sound for Oscon when compared with Cerafine or BG. However, i will still find an Oscon for my last PS cap to be replaced to have diversity of sound and not fixed on one particular brand sound.

For filtering, I will use the Wima MKP 0.15 or 0.1uf for the cheapest solution. Definitely way way better than the ceramic by as much as 30% and killer mode here! But for the best sound use Teflon or BG NX Hi-Q 0.15 or 0.1 uf. If using OPAs, and the OPAs doesn't have local PS caps for + and - supplies, beef it up like I had did with super e-cap cofiguartion of Cerafine15 uf and BG NX Hi-Q 0.1uf for supersonic filtering. Another jaw dropping expereince here! Well, by another 10 to 20 % improvement if needing any quantifying. After all this, u will have xtreme 3D sound.

[dejanm]

XTRProf, thanks for your reply. You gave me enough material at the moment to think about.... Especially about using BG NX as bypass cap.

[dejanm]

Hi,

I just did the mod of my MF X-DAC V2. I replaced the following parts:

PSU:

- I put Panasonic FM 1000 uF, 16V - 1 piece
- Panasonic FC 470 uF, 25V - 2 pieces
- Panasonic FC 150 uF, 35V - 2 pieces (originally Jamicon 100 uF,35v)

I bypassed all of them with Evox-Rifa MMK 0,1 uF, 63V.

I replaced also diodes and put Vishay-Telefunken SF4007 - 5 pieces.

Signal path:

- output resistor Caddock 47R - 2 pieces
- Cap between filter and buffer stage on the signal path Blackgate 10 uF, 50V - 2 pieces
- OPA2111 - 4 pieces

I was thinking a lot about tantalium caps that are all over the place but then I decided to replace only the cap that is directly in the signal path working as DC blocker. All other, mainly filtering tantals, I simply left. There are lot of different opinions about them but it seems to me, after reading some forums and discussions about them, that they are quite OK when used for filtering purposes.

Anyway, V2 unit is very small and there is no much place to play. Even putting Blackgates was not such an easy task.

Possible further improvements:

- using better wire (silver ?)
- better connectors (WBT ?)
- putting Lundahl LL1566 as decoupling between RCA SPDIF input and the rest (with Caddock 75R on the secondary) would be nice thing to do - there are some problems though with the place -> where to put it ?

So this is a brief report. Now I am listening to it and waiting some 100 or whatever hours for Blackgates to burn in and then ... we will see ...

[XTRProf]

Aug 30, 2008 11:09:47 GMT dejanm said:

- using better wire (silver ?)
- better connectors (WBT ?)
- putting Lundahl LL1566 as decoupling between RCA SPDIF input and the rest (with Caddock 75R on the secondary) would be nice thing to do - there are some problems though with the place -> where to put it ?

HI dejamn,



Silver wire, at least to my dewaxed ears, is not as good as the OCC copper. Ya, it has all the details at the expense of emotion and bloom of the real music. However, this is system dependent. So do try them at short length before committing further.

Ya, connector if you can get the copper version instead of the brass ones will be a worthwhile upgrade. However, the brass ones on my V3 are already very good and so I just skipped for my V3.

Ah, yes, if you can improve on the MF SPDIF pulse traffo that will be great! The MF uses a bare minimum traffo there for matching. Also, don't forget to change the coupling cap there as well. I changed mine to a Solen Teflon and I was blown away. Not cheap though b ut definitely worhtwhile to save up for at least that coupling cap.

Ya, how much you are paying for your pulse Lundahl trafo?

[Deleted]

if you can improve on the MF SPDIF pulse traffo that will be great! The MF uses a bare minimum traffo there for matching. Also, don't forget to change the coupling cap there as well. I changed mine to a Solen Teflon and I was blown away. Not cheap though b ut definitely worhtwhile to save up for at least that coupling cap.

XTRProf
Are you referring to the V3 here ? If you are, can you please elaborate on the location and value of that cap etc.
SandyK

[dejanm]

Concerning Lundahl ... I will contact Jac Music at:

www.jacmusic.com

and ask ... Very similar transformer there (but with two secondaries - splitting version -> LL1527A) costs there 53 euros. I guess that LL1566 or LL1572 should be a bit cheaper.

[dejanm]

LL1566 here for 20 euros:

www.tripleaudio.nl/webshop/pdf/prijslijst_lundahl.pdf

[XTRProf]

Sept 5, 2008 11:46:16 GMT @sandyk said:

if you can improve on the MF SPDIF pulse traffo that will be great! The MF uses a bare minimum traffo there for matching. Also, don't forget to change the coupling cap there as well. I changed mine to a Solen Teflon and I was blown away. Not cheap though b ut definitely worhtwhile to save up for at least that coupling cap.

XTRProf
Are you referring to the V3 here ? If you are, can you please elaborate on the location and value of that cap etc.
SandyK

Yes, C45 104 ceramic, I think, as I didn't open it up now and depended on my sometimes overloaded memory. Really worthwhile to have the best there as it's only one piece of 0.1uf.

[XTRProf]

Sept 5, 2008 13:23:11 GMT dejanm said:

Concerning Lundahl ... I will contact Jac Music at:

www.jacmusic.com

and ask ... Very similar transformer there (but with two secondaries - splitting version -> LL1527A) costs there 53 euros. I guess that LL1566 or LL1572 should be a bit cheaper.

Sept 5, 2008 14:39:08 GMT dejanm said:

LL1566 here for 20 euros:

www.tripleaudio.nl/webshop/pdf/prijslijst_lundahl.pdf

Thanks very much to share to all.

[dejanm]

Hi,

I need your help. I am struggling with cap values written on them.

104J63 - I am interpreting as 0,1 uF 63 V ... is it correct ?
100J 100V - Obviosly 100V small cap but 100J is 1 pF ?

Thanks in advance ...

[Deleted]

dejanm
104J63 is indeed 0.1uF (100nF) 63V - "J'' denotes +-5% tolerance.
100J 100V is most likely 100pF . More recent 100pF capacitors would be marked 101J.
Smaller ceramic capacitors are normally just marked with their capacitance in pico Farads (pf)
e.g. 47,or 47J for 47pF 5%.
SandyK

[dejanm]

Sept 6, 2008 8:22:20 GMT @sandyk said:

dejanm
104J63 is indeed 0.1uF (100nF) 63V - "J'' denotes +-5% tolerance.
100J 100V is most likely 100pF . More recent 100pF capacitors would be marked 101J.
Smaller ceramic capacitors are normally just marked with their capacitance in pico Farads (pf)
e.g. 47,or 47J for 47pF 5%.
SandyK

SandyK,

Thanks a lot for this ... I have one question more. There is a small red cap. The following is written on it: L222J. It is a filter cap on the minus input of OPA. What is the value of it ? Which voltage ? What type of cap would be a good replacement: MKP or SilverMica or something else ?

That's more than 1 question, I know ...

[Deleted]

dejanm

L222J - .0022uf = 2200pf (2.2nF) A 2.2nF 5% MKP (polypropylene) rated at 63V or higher would be a good choice, as well as being compact. I do not know what the L stands for.

SandyK