Salas hotrodded blue DCB1 build

[AndrewT]

We are talking only about the loop through the string of 5 led's that connects the two plates of the 0.22uF capacitor right?

AndrewT,

Would twisting actually harm or improve the cancellation you are talking about?

the current through the LEDs is non-varying DC.
There is no inductance issue, because the current is non-varying.

 

[BigE]

Thank you, AndrewT.

 

[Alarudin]

LSK170 and 2SK170

Hi,

Recently ordered 1 set of 8 LSK170 B grade from DIYAUDIO Store without realizing i needed 10 of them . Now DIYAUDIO store doesnt have B grade LSK170 anymore only D grade. Will i be fine with using 4 matched LSK170 and 6 Unmatched 2SK170BL from B+D Enterprise? or does Linear System actually sells in small quantity. Kindly advice. Thank you!

 

[chungtat]

You could use 6 unmatched 2SK117B for 6 unmatched 2SK170BL. Someone asked before in this forum.

 

[Alarudin]

Thank you for your reply. My question is more of is it ok to mix the match set and the unmatch set. Like 4 x LSK170B with 6 x 2SK170BL or 2SK117BL. Is it necessary for it to be uniform or it doesn't matter. Thank you.

 

[chungtat]

You could mix. Only 4 FET for audio is restricted to 2SK170BL. The other 6 are fine with either 2SK117B or 2SK170BL.

 

[Alarudin]

Thank you. Really help me save some money and time.

 

[stajo]

What Mundorf EVO 0,33 uF cap is it that has proven good at diode decoupling? Standard, Oil or Silver/gold in oil?

 

[AndrewT]

As mentioned earlier, the current through the diodes does not change. it is non varying DC.
There can be interference applied across the voltage reference. An RC filter will attenuate interference. That is not decoupling of an active stage, that is filtering of interference.

 

[stajo]

Ok. I thought it was noice harvesting of a 5 diode vref. Mea culpa. Im interested in the subjective results anyway if someone has.

 

[Salas]

The cap across the Vref is to filter a bit of its noise. Its influential because it affects the HF impedance. It drops it evenly to counteract its inherent rise. I preferred a large cap across than RC when I needed much noise filtering for low signal level in other applications.

 

[stajo]

Noice is AC to me. Filtering, decoupling... I dont understand the biggy

 

[AndrewT]

Decoupling is a source of near instantaneous current when an active device suddenly changes it's current demand. It helps prevent the rails sagging due to supply rail impedance.

To me that seems almost unrelated to the action of an RC filter.

 

[Salas]

Noice is AC to me. Filtering, decoupling... I dont understand the biggy

Its not a decoupling cap function in this circuit, but a filter cap. The LEDs have a total series impedance with some characteristics as noise and shape vs freq.

 

[BigE]

The unit has been powered for 12 solid days. The FT-3 caps are sounding a lot better now. Will report again Sunday when nearer being on for 400 hours. At this point, I don't see an issue with the external line length. Although, I'm thinking of changing the wiring from 18 awg stranded to 20 awg solid core.

 

[Salas]

Sunday then.

 

[BigE]

Is there a sonic effect between stranded and solid core wiring?

 

[Salas]

I don't know I never used remotely located filter cap.

 

[Erlend Sæterdal]

Is there a sonic effect between stranded and solid core wiring?

Allways . I never use Stranded.

 

[BigE]

This is the only piece of stranded in the build, except for the grounding through the diode/resistor network. The input/output is solid core neotech legenburg, 24 awg equivalent. I did not use good wire for the test.

I'm going to replace the stranded wire on the caps.

Question is: 20 or 24 gauge? 20 awg is teflon/OCC silver, 24 is neotech legenburg -- rectangular OCC copper. A close coupled pair of Legenburg would be less inductive, because it is rectangular. The 20 awg would have less than 40% of the resistance of the 24 awg.

The run is about 6".

Should probably change the signal grounding wiring as well.