V-dac modifications

[spore]

Josep/Dutchamps,

I tried finding a suitable sized non-elco 4.7uF caps that will fit inside the stock but not able to do so. Good ones meant for audio do not fit in and are more expensive than the normal caps. I am trying to keep the cost of modifications low also.

I tried changing the dc output caps with normal caps of 3.9uF, 15uF and 47uF. 3.9uF sounds terrible with weak and mudy bass while 15uF improved a little with more defined bass. I got best sound with 47uF. However, I changed the 47uF back to 100uF as when I did a comparison previously between 47uF and 100uF, 100uF gives better imaging and bass and I preferred 100uF sound.

I know this may defy logic but seems like higher capacitance values are better. Dutchamps, since the 4.7uF non-elco cap sounds much better than 2.2uf, have you tried higher values of non-elco cap? I find it hard to accept that a 4.7uF non-elco cap will sound good when a normal cap of the same value sounds much worse.

Will a difference in materials will make such a big difference? Or is it due to our different op amps used? I read that tubes have worse distortion values in certain parameters than ss but seems like tube still has a large following and sounds better than ss.

I have also changed some other capacitors to my v-dac and my updated list is as follows:

Modification Lists:
C25 : Nichicon Muse 1000uF (H0112)
C5,C53,C21,C44,C34: SE 100uF (S.E.66)
C14: Nichicon Muse 47uF (H9945) [Original 10uF]
C22, C52, C3, C4, C45, C61: Nichicon Muse Bipolar 100UF (A0204) [C3,C4 Original 47uF]
C43, C46: ELNA 1uF (0418)
C18, C58, C12, C16, C20, C18, C30, C36, C29, C27, C28, C32, : ELNA Starget 10uF (0749)
C31, C28: ELNA Cerafine 0.47uF (Original 104M)
C38, C37: ELNA Starget 47uF (Original 10uF)
IC5:LME49740NA [Original MC33079]
IC7: LME49860NA [Original NE5532]

 

[praganj]

If you are 100% sure that your preamp has input caps, than you can REMOVE completely the output caps from the V-dac. If pre has not input and output caps blocking DC and your power amp has not input caps, than there is a danger that DC will destroy your speakers ! Be carefull !

 

[Dutchamps]

In serial placed caps into the Signal path if you use elco`s then you miss alot of transients.
If a elco does sound better, then your preamp is not so good as your vdac.
And what Praganj says is true too.
I put the v-dac right into my preamp, a e83cc tube.
With again no input cap, think your preamp is a high inpedance input then you need those big 47-100uF.
If your preamp is less high impedanced then you should hear a day and night differance with elco of mpk caps.
I even choise the wrong mkp from intertechnics those have a little overdue on high frequenties but stil 1000 % better soundings.
And yes then the cover wont fit if you use those mkp.
and locate those input caps in your pre-amp
remove that one of the input you use in the pre-amp (just put a jumper in the place)
place the mkp back into the v-dac and or link it into the v-dac and place the mkp in the preamp (more room)try it out for a 2-3 days befor you judge the sound quality course it can take this long. caps need to "burn in"
I am convinced about the bad sounding elco compaire to the mkp caps.
Back in the 1980`s i have read an article about a cap test.
The DA factor of caps, the lower the amount of loss the better the caps sound.
Even a bip. cap of 100uF 16v a 100uF 100v sounds much better.
Best caps are in line:

Polystyrene:
types like styroflex ideal for filtering uses like those caps near the 5532 mc33079p
or like RIAA for finyl.

Mkp:
couple caps for amps, filtering caps for speakers

Mkc:
Polycarbonate
couple caps for amps, filtering caps for speakers
Sounds little less in micro details, but a oke Cap

Ceramic caps:
Very good for high frequent distortion filtering, in dac`s
Im not pleased to hear those in my amps in the signal path.

Electrolyte Caps non polarity (bipolair elco)
Worse in details like the above named caps, not suitable in signal path applications.

Electrolyte Caps (elco)
Far worse in details like the above named caps, not suitable in signal path applications
Some view marked fabricates are good like blackgate Philips and so on.
In factory builded devices amps cdplayers etc many elco`s are bad, mainly becourse of the low amount of voltage limits. (1000uF/6.3V) A 1000uF/100V is too expensive to be used in mega manufacturing of those devices.

From the input pre amp, through the power amp together have 2 caps in the signal path and they are MPK.


This is what i think about it

 

[spore]

Pranji/Dutchamps,

Thanks for the advice and sharing. I am connecting the v-dac directly to my preamp Bryston BP25. A quick search on the web reveals that the input impedance is about 48k Ohms. If this value is considered as high impedance, then that will help explain my observations.

As I still have warranty on the Bryston preamp, I don't think I will try to bypass the internal input caps.

 

[spore]

I have added the original bypass capacitor 104M back into their original position at C31 and at C28 in parallel with the 0.47uF capacitor. There was not much difference in having the bypass capacitor although I think the high frequency and low bass is better defined slightly. However, it could be my mind playing tricks on me also. Anyway since there is no negative effect that I can perceive, I left them in the new position as they are.

My updated Modification Lists:
C25 : Nichicon Muse 1000uF (H0112)
C5,C53,C21,C44,C34: SE 100uF (S.E.66)
C14: Nichicon Muse 47uF (H9945) [Original 10uF]
C22, C52, C3, C4, C45, C61: Nichicon Muse Bipolar 100UF (A0204) [C3,C4 Original 47uF]
C43, C46: ELNA 1uF (0418)
C18, C58, C12, C16, C20, C18, C30, C36, C29, C27, C28, C32, : ELNA Starget 10uF (0749)
C31, C28: ELNA Cerafine 0.47uF in parallel with original 104M
C38, C37: ELNA Starget 47uF (Original 10uF)
IC5:LME49740NA [Original MC33079]
IC7: LME49860NA [Original NE5532]

 

[praganj]

I have v-dac since few days and i am very happy with the performance, but - IMHO there is some lack of dynamic, PRAT and details. I think the first thing i will do will be one regulated psu, than i want to exchange IC5 and IC7, and than some caps.

I hope that better psu will bring some more dynamics and PRAT. I am little afraid about changing the IC5 and IC7. Maybe the sound will become more detailed, but also i will loose some musicality and this beautifull full body vocals ? What is the sonic difference beetwen LME49860NA and OPA2604 ? Anybody tried ?

p

 

[spore]

I have been running in my modified v-dac for several hours. In addition, I have also did a comparison between the modded v-dac vs the dvd player denon 1740 which I was using previously before I got the v-dac.

The differences between the modded v-dac and dvd player can be clearly heard. The bass is more deep and robust, soundstage is slightly wider with more depth, vocals more smooth and airy. Tonality is excellent and music just sound so 3-d, clear and lifelike.

In comparison, the dvd player sounds a bit deaden and I cant imagine that I quite like the dvd sound previously. It seems like a heavy curtain has been lifted off the music.

I remembered the orginal v-dac is slightly better than the dvd player when I did a comparison the 1st time. However the modded one now excels the dvd player in all areas. The difference is very obvious. This was a good learning experience with excellent results.

praganj, if you are modding, why dont you trying fitting a adapter so that you can swap various opamp to try out and see which one you like? If you could report your findings, that will be great. Or you may want to try my modification list as I dont perceive any loss of musicality and full body vocals. The parts required to modify are not expensive either. It will be interesting to hear feedback on how the modded v-dac fares across different systems.

By the way, the reason I choose LME4860NA is because it is a direct replacement for NE5532 . In addition, good results are obtained for LME4960NA. Refer to this link http://www.sg-acoustics.ch/analogue_...distortion.pdf for opamp distortion results.

 

[Dutchamps]

What i have mesured and traced back till so far.
Now building me a power supply sepparated units for 3,3V 5V 9V

Excusses for the dutch

Gemeten spanning means mesured voltages in and output

Note for the drawing the 0,5 must be 5 volts i still dont get it
Im not realy sure about the lm317 setup course its a regulated voltage setup

 

[josep]

Duthcamps,

Do not miss in your measuements the -V analog rail, it is very important too and the weaker point of this unit as it is derived with DC-DC switching converters! You MUST have this too on the new PSU.

Re. the coupling caps, Spore, he is right... I think your Bryston is DC-coupled, so no caps already on its inputs; be careful then (look for it on its user manual) and do NOT short the output caps of V-DAC.

Cheers

 

[spore]

Dutchamps, nice drawing and work. It will be interesting to see your PSU when its completed.

Joseph, I wont remove the dc caps as I still have high dc offset.I am afraid of spoiling my speakers too! If anyone has ideas that can reduce the dc offset to low values such that the dc coupling caps can be bypassed, please share!

I have changed the capacitors around the op amps to non-elco type of the same values which I bought from a local electronics shop. They are no brand ones but changing to a non-elco results in increased clarity but music sounds a bit off. I had to change some capacitors to get back the music. Everything gels now and music is one notch higher!

Take note that with the ELNA Silmic 100uF, its just barely fits the original case and I need to squeeze it in. The sides of the capacitor got scrapped off a bit when pushing it in.

My updated Modification Lists:
C25 : Nichicon Muse 1000uF (H0112)
C5,C53,C21,C44,C34: SE 100uF (S.E.66)
C14: Nichicon Muse 47uF (H9945) [Original 10uF]
C22, C52, C45, C61: ELNA Silmic 100uF [C3,C4 Original 47uF]
C3, C4: Nichicon Muse Bipolar 100uF (A0204) [Original 47uF]
C43, C46: ELNA 1uF (0418)
C18, C58, C12, C16, C20, C18, C30, C36, C29, C27, C28, C32, : ELNA Starget 10uF (0749)
C31, C28: ELNA Cerafine 0.47uF in parallel with original 104M
C38, C37: Nichicon Muse Bipolar 100uF (Original 10uF)
IC5:LME49740NA [Original MC33079]
IC7: LME49860NA [Original NE5532]
C147, C151, C51, C47: Non-elco 3900pF
C160, C155, C55, C60: Non-elco 22pF

 

[iszatso]

I am running a 12v gel cel battery to the v-dac versus a switched 2500ma switched source. Is there something that should go between the batterry and the v-dac for filtering of the power source? Seemd better with the battery versus the stock ps and the switched 2500ma I purchased seperately. I have never used a soldering iron before, although i have welded steel.

Was hoping all these mods get filtered out to the best all around reasonable upgrade and then send it to someone for the upgrade. seems like the choices are stil in flux as to the most mprovement generally over the stock sound. The ps change certainly made it sound better, but that isone that I could do without burning the thing up.

when I run the battery full charge 12.8, charge the battery when it gets to 12.2. Any suggestions much appreciated.

 

[spore]

iszatso, I have not used a battery but I remember that someone did use a battery and reported good results in the main MF vdac thread. I think there the battery is connect directly to the vdac.

I made further modifications to my vdac, replacing some capacitors back to their previous values and swapping some out to different type of see that will affect the sound. I think this could be the configuration that I will be keeping

My updated Modification Lists:
C25 : Nichicon Muse 1000uF 35V
C5,C53,C21,C44,C34,C39: SE 100uF 16V (S.E.66)
C14: Nichicon Muse 47uF 25V [Original 10uF]
C22, C52, C45, C61: ELNA Silmic 100uF 63V
C3, C4: Nichicon Muse Bipolar 100uF 16V [Original 47uF]
C43, C46: ELNA 1uF 50V
C18, C29, C27, C36, C16, C12, C20, C58 : ELNA Starget 10uF 35V
C31, C28: ELNA Cerafine 0.47uF 50V in parallel with original 104M
C30, C32: Nichicon Fine Gold 10uF 50V
C38, C37: Nichicon Muse 47uF 50V [Original 10uF]
IC5:LME49740NA [Original MC33079]
IC7: LME49860NA [Original NE5532]
C147, C151, C51, C47: Non-elco 3900pF
C160, C155, C55, C60: Non-elco 22pF

 

[Dutchamps]

Quote:

Re. the coupling caps, Spore, he is right... I think your Bryston is DC-coupled, so no caps already on its inputs; be careful then (look for it on its user manual) and do NOT short the output caps of V-DAC.

The only thing that will hapen is that you dont get any bass out of the V-dac

If you got a nice MKP input C then the V-dac`s output C is overkill, only then.
If non input cap is in your pre-amp , you still need the output C of the DAC.

Renember people testing and trying out new or better components.

Dont espect to hear results with in minutes, the replacement for my ne5532 to the op270FZ was about 8 days befor burned in properly.
After 3 hourse running cds the opamp totaly changed the soundstage.
Also new mpk caps need to saddle down for a wile.

I renember i was shocked to hear the first sounds out of the new OP270FZ,
After a view days it was totaly changed.

Another isue i never ever turnoff the power of my devices.
search as cd/dvd player pre-amp and v-dac.
The only device my poweramp is powered off.
Stability is the most important improvement you could make

 

[Dutchamps]

To josep

My plan with the power is to solder out all the voltage regulators and on the place where the voltage regulators gives off the stable voltages on the pcb, i will make a stable 3,3 / 5 / 9 volts powersupply and connect them to the pcb where the voltage regulators where paced.
So there will be 4 wires from that powersupply on the pcb board.

not easy for me to put in the right english lauguage.

 

[spore]

After spending some time to burn in and listening thorough different types of music, some tracks just does not sound right and sound much worse in fact. I made some minor changes to optimize and fine tune the vdac sound.

I will stop my modding now as there is no end to modding. some changes, you gain some while you lose some. Now my vdac sounds balanced and good. I am satisfied. take note that results may vary depending on your equipment and preference

Take not that space is tight to fit both the capacitors at location C32 and C61. The Polypropylene capacitor are general ones which I got from my local electronics store.

My updated Modification Lists:
C25 : Nichicon Muse 1000uF 35V
C5,C53,C21,C44,C34,C39: SE 100uF 16V (S.E.66)
C14: Nichicon Muse 47uF 25V [Original 10uF]
C22, C52, C61: ELNA Silmic 100uF 63V
C45 : ELNA Silmic 100uF 63V in parallel with ELNA Cerafine 0.47uF 50V
C3, C4: Nichicon Muse Bipolar 100uF 16V [Original 47uF]
C43, C46: ELNA 1uF 50V
C18, C29, C27, C36, C16, C12, C20, C58 : ELNA Starget 10uF 35V
C31, C28: ELNA Cerafine 0.47uF 50V in parallel with original 104M
C30, C32: ELNA Simlic 10uF 50V
C38, C37: Nichicon Muse 47uF 50V [Original 10uF]
IC5:LME49740NA [Original MC33079]
IC7: LME49860NA [Original NE5532]
C147, C151, C51, C47: Non-elco 3900pF (Polypropylene)
C160, C155, C55, C60: Non-elco 22pF (Polyester)