Shigaclone MKII Black - The builders Thread

Rudy,

The board can accommodate PI or series s/pdif divider.
R60, R48, R49 are part of PI divider.
R48 and R49 are part of series divider.
R48 = 300Ω series resistance
R49 = 100Ω resistance to ground (shunt)

Modified BOM to reflect correct resistor number.

Regards,
Tibi
 
Last edited by a moderator:
The new resistors are soldered now and that was an improvement.
Series resistance:270R/ 0.4 Watt, Vishay VAR-Series "naked" Z-Foil Resistor.
Shunt resistance:100R CADDOCK / 0.75 (3/4) Watt, MK-132 Series, 1%, Radial Lead, Precision Metal Film.

And the capacitors changed from the power supply that are now 2 x 1000uF Blackgate STD.
After the diodes, 1 x black gate 4,7uF n series (under the board) parallel to Blackgate 1000uF.

Now spending a week listening to music and after a week compare if something is changed from the sound.
Do not tell exactly what the differences are with my existing Shigaclone, be patient:)

To be continued.
Regards,
Rudy
 

Attachments

  • power supply BG.jpg
    power supply BG.jpg
    403.4 KB · Views: 2,333
  • Resistor spdif out.jpg
    Resistor spdif out.jpg
    425.9 KB · Views: 2,291
The new resistors are soldered now and that was an improvement.
Series resistance:270R/ 0.4 Watt, Vishay VAR-Series "naked" Z-Foil Resistor.
Shunt resistance:100R CADDOCK / 0.75 (3/4) Watt, MK-132 Series, 1%, Radial Lead, Precision Metal Film.

270R and 100R will give an attenuation of about 11.3dB which is about a db less than the standard values used so far.

Of course how close to 0.5V your signal has to be and how big the perceived benefit is going to be, is a matter of DAC, so your sound benefit mileage may vary. The benefit of the below stuff for me was big.

My scope measurements (at the DAC's input after a 1.5m cable) were that we need about 14.2 dBs to get to the specified 0.5V peak to peak on the output

349141d1368741628-finally-affordable-cd-transport-shigaclone-story-zzzzzzzz.jpg


My choice was a T pad (50 series, 32 shunt, 50 series) for 75R input-75R output configuration (which I chose when I was wiring my own resistors. The smaller values of the Tee pad meant less wire used). With solid resistors, a Pi pad would be more suitable, because of less components in series with the signal (something like 111R-185R-111R). I have not tested the difference between L pad and symmetrical tee/pi pads directly. My choice for symmetrical impedance in the attenuator was based on the elimination of impedance change points, to minimize reflections. For the same reason I have also eliminated the RCA plug/socket, soldering the cable right to the attenuator.
 
building shiga mkII high-grade

Going further with high-grade version of Shiga MKII.

Preparing board for soldering.
Place a small drop of solder on each smd pad. So you'll solder one side of each component first.
Depending on which hand you use, place this solder drop on the right or left side of the component which was drawn on the PCB.


Regards,
Tibi
 

Attachments

  • 20140505_016.jpg
    20140505_016.jpg
    688.4 KB · Views: 2,122
  • 20140505_017.jpg
    20140505_017.jpg
    840.6 KB · Views: 2,067
Best way to solder smd capacitors, especially large or high value ones, is to have them perpendicularly mounted, rotated at 90 degrees.
This dramatically reduce piezoelectric effect due any vibrations induced by music program and mechanic resonance. Another benefit is that any parasitic capacitance between capacitor layers and PCB traces are also minimized.
Even resistors have similar ceramic substrate (mainly on thin film resistors) and these do not exhibit same magnitude of piezoelectric effects, I decided to mount few of them vertically as well (in critical places) in order to minimize any parasitic capacitance with pcb traces.

I know it is harder to solder smd this way, but worth the effort.
This type of soldering can be done only manually. Stencil will not work.

Regards,
Tibi
 

Attachments

  • 20140514_004.jpg
    20140514_004.jpg
    694.2 KB · Views: 589
  • 20140514_005.jpg
    20140514_005.jpg
    640.4 KB · Views: 476
Last edited by a moderator:
Please note that in above pictures L5 and L9 are not mounted. This need to be removed when miniregs are in place.
Also note that L8-ferrite bead, C59-10nF, R47-47ohm need to be mounted for Tentlabs XO.
Also C57 and C58 will not be mounted. These are needed only when internal LC78601R oscillator is used.


Regards,
Tibi
 
Solder IC's and smd transistor after all passive SMD parts are in place.
This is how the board should like after all smd parts are in place.
Estimated working time 3-4 h, including parts measurement and board cleaning.

Regards,
Tibi
 

Attachments

  • 20140514_012.jpg
    20140514_012.jpg
    905.3 KB · Views: 336
  • 20140514_011.jpg
    20140514_011.jpg
    819.6 KB · Views: 299
  • 20140514_010.jpg
    20140514_010.jpg
    635.3 KB · Views: 280
  • 20140514_009.jpg
    20140514_009.jpg
    737.2 KB · Views: 257
  • 20140514_008.jpg
    20140514_008.jpg
    629.6 KB · Views: 303
  • 20140514_007.jpg
    20140514_007.jpg
    718 KB · Views: 466
I must mention that following high-grade parts are my personal choice and I selected these parts after several hundred hours of listening and comparisons.
Using different parts you can get better results or better profiled on your sonic preferences.

My preferred low jitter XO
- Tentlabs XO 16,9344MHz TentlabsShop

For highly commented C8, my choice is:
- C8 - Mundorf Silver-Goild OIL 0.1uF Mundorf Film Caps EVO - Silver|Gold + Oil
Please note that lowering the value of this capacitor may lead to oscillations and instability - tracking loss. I suggest do not go under 33nF. Personally I'll stay on the safe part and use 100nF.

C10 is decoupling laser CCS. A high current low ESR capacitor is mandatory. In this regard oscon/polymer/tantalum are best here.
C43 is the capacitor after internal LA6541+2SB764 regulator. This ensure regulator stability and noise is in direct relation with the ESR of this capacitor as well.
- C10 and C43 Sanyo OSCON 270uF Capacitors_Electrolytic_Sanyo_Oscon

C19 is part of laser CCS. The quality of this cap will highly affect laser operation. Laser current is in relation with voltage across this capacitor. Use the best you can here. BG/oscon/tantalum. My choice BG PK 47uF - still available at partsconnex.
- C19 - Rubicon BlackGate PK 47uF Blackgate Electrolytic Capacitors PK Series

C20 and C23 are part of 2.5V reference. You may have noticed that on all Shiga's these capacitors are polymer in my kits.
The temperature stability of polymer caps recommend these here. The internal structure of LA9242M is in such way constructed that any leakage and internal series resistance of these capacitors will affect 2.5V reference.

My choice for R48 and R49 is Vishay naked.
IMHO the best resistors your money can buy for this application. Exceptional stability, tolerance and noise figures.
These will enhance resolution to dream level. I simply love these parts. :)
Vishay VAR-Series "naked" Z-FOIL RESISTOR

Miniregs - You have construction of these in this thread.

The icing on the cake is C11. Give it a try and be amazed.
Use a 1nF - 2n2F silver mica cap.
capacitor_film_smica.html

Regards,
Tibi
 

Attachments

  • 20140515_008.jpg
    20140515_008.jpg
    691.3 KB · Views: 494
  • 20140515_007.jpg
    20140515_007.jpg
    775.4 KB · Views: 344
  • 20140515_006.jpg
    20140515_006.jpg
    530.6 KB · Views: 348
  • 20140515_005.jpg
    20140515_005.jpg
    651.2 KB · Views: 415
  • 20140515_004.jpg
    20140515_004.jpg
    805.5 KB · Views: 491
  • 20140515_003.jpg
    20140515_003.jpg
    617 KB · Views: 483
  • 20140515_002.jpg
    20140515_002.jpg
    642.8 KB · Views: 500
  • 20140515_001.jpg
    20140515_001.jpg
    801.3 KB · Views: 563
Last edited by a moderator:
All parts mounted and board is almost ready to be connected to display and control board.
Take care on J1. A version of this connector can be mounted in two different ways. Zoom and follow below picture.

TBC.

Regards,
Tibi
 

Attachments

  • 20140515_023.jpg
    20140515_023.jpg
    841.6 KB · Views: 444
  • 20140515_020.jpg
    20140515_020.jpg
    642.7 KB · Views: 398
  • 20140515_019.jpg
    20140515_019.jpg
    493.6 KB · Views: 271
  • 20140515_018.jpg
    20140515_018.jpg
    489.5 KB · Views: 277
  • 20140515_017.jpg
    20140515_017.jpg
    763.6 KB · Views: 288
  • 20140515_016.jpg
    20140515_016.jpg
    705.6 KB · Views: 1,319
  • 20140515_015.jpg
    20140515_015.jpg
    715 KB · Views: 1,386
  • 20140515_014.jpg
    20140515_014.jpg
    750.7 KB · Views: 1,465
  • 20140515_013.jpg
    20140515_013.jpg
    533.1 KB · Views: 1,551
  • 20140515_012.jpg
    20140515_012.jpg
    663.5 KB · Views: 1,675
Last edited by a moderator:
+1 for C11
But I do not consider it icing :) I consider it a significant part of the main course. It makes a huge difference. Possibly as big as C8.
Oh and moving from a cheap 1euro noname cap to a Cornell Dubilier also had a quite audible impact.

I hope to try the miniregs very soon. I will post back...
 
I may suggest to remove that 4.7uF together with C62 100nF MKP and put a 100nF PIO.
I got degraded results paralleling/decoupling BlackGates with any capacitor, except PIO.
Also remove C66 and insert a small 10nF PIO as well.

Regards,
Tibi

Hi Tibi,

Sorry I haven't responded, I have a few days in the hospital with a throat surgery I am yesterday again come home.
I feel getting easier but it should now heal again.

Now the modifications, I still have parts at home?
Black Gate NX HI-Q 0.1 uF, I put those in the power supply C62
C19, I use here 47uF Black gate N series.
C10 and C43 Sanyo OSCON 270uF, I will order by parts
C8 - Mcap EVO - Silver / Gold & Oil 0.1uF, I will also order by parts

Also remove C66 and insert a small 10nF PIO as well.
Let me know what brand I should order?

I also have PIO Russian caps, see picture.

Regards,
Rudy
 

Attachments

  • PIO 100nf russian.jpg
    PIO 100nf russian.jpg
    250.7 KB · Views: 406
  • BG NX 0.1uF.jpg
    BG NX 0.1uF.jpg
    219.8 KB · Views: 277
  • BG NX 100nf.jpg
    BG NX 100nf.jpg
    213.9 KB · Views: 240
Last edited:
C8 value?

Tibi,
There are two caps I would like to try in C8 position. I'm building a whole new system ( speakers, amp, DAC, Transport, and cables and power cords ) and will be burning it in for 300 hours in my shop, before bringing it into the house for serious listening. If I put both .1uf caps in parallel, in the C8 position, would .2uf hurt anything in this position. This way I can burn in both caps at once, and then remove one when I bring it into the house, listen for however long I want, and then change caps and listen some more. Will it work? Thanks.:cheers: