Zen -> Cen -> Sen, evolution of a minimalistic IV Converter

[EUVL]

Panasonic FC 470µF 35V, 10mm dia. x 20mm Length.


Patirck

 

[samoloko]

Patrick and suburra

why are you using 1.4K as a R_iv with PCM63
Isn't It a 1.5 K recommended with PCM63
do I need a buffer If I drive 20 K attenuator with SEN

 

[suburra]

Hi

Patrick, for PCM 63, recomended 1.4K for 2Vrms Out
I tried 1.4K/1.2K/2K
2K was better for my ears, and Caddock Mk132 better then Vishay TX2352

For sure 10K shunt attenuator needed a buffer
Maybe with 20K you can try without

Marco

 

[samoloko]

thank you for reply

can you describe difference with / without buffer
you said that lows are not correct ?

 

[suburra]

Hi

Without buffer the sound was constrict, and i had sensation to loosing control and energy in the bass frequency

With buffer no more these sensations
Now the input inpedance of DCB1 it's 28k and Res I/V is 2K

Hope my poor english was clear

Marco

Inviato dal mio GT-S5300 con Tapatalk 2

 

[LuisMCP]

Bias point drift?

I have been some kind of "bias drifting"
With a senoidal signal of 1kHz, output is clean an undistorted in the first minute.
After 1st minute, negative half-wave becomes clipped.
If signal is square, 1kHz, no clipping.
If signal is senoidal, 40 kHz, no clipping.

If turn-off some time, 1kHz senoidal is unclipped again 1st minute.

I have fed the circuit with batteries and a regulated psu too. Results are the same independently of what PSU you use.

I believe that the capacitors are responsible for some kind of drift when they get charged, but not sure.

 

[ryanj]

I have been some kind of "bias drifting"
With a senoidal signal of 1kHz, output is clean an undistorted in the first minute.
After 1st minute, negative half-wave becomes clipped.
If signal is square, 1kHz, no clipping.
If signal is senoidal, 40 kHz, no clipping.

If turn-off some time, 1kHz senoidal is unclipped again 1st minute.

I have fed the circuit with batteries and a regulated psu too. Results are the same independently of what PSU you use.

I believe that the capacitors are responsible for some kind of drift when they get charged, but not sure.

Your absolutely right, i had the same problem with some other russian film caps. But then i switch to Nichicon Muse caps and had almost no detectable DC leakage. - Nothing i could detect with my basic equipment anyway.

 

[LuisMCP]

Your absolutely right, i had the same problem with some other russian film caps. But then i switch to Nichicon Muse caps and had almost no detectable DC leakage. - Nothing i could detect with my basic equipment anyway.

mmmm... interesting.

Ouch! I have read BOM and saw *bipolar* caps.
Mine are polarized caps

 

[samoloko]

mmmm... interesting.

Ouch! I have read BOM and saw *bipolar* caps.
Mine are polarized caps

you can use polarized caps

 

[EUVL]

Your "drift" is very likely due to leakage in your capacitors.


Patrick

 

[LuisMCP]

Your "drift" is very likely due to leakage in your capacitors.


Patrick

I have tested today two new type of caps:

Polarized Elna Silmic
Bipolar Nitai electrolityc

Polarized cap ended with a great clipping.
Bipolar Nitai don't

I will try Nichicon Muse next days.

I have measured harmonic distortion with the Nitai's and found 2nd harmonic below 30 dB of fundamental with only 0,2Vrms of signal (18V psu)

Probably i will get some litycs of low leakage for testing

 

[TotoDIY]

>
So to be on the save side, I would not use anything less than 220uF for signal current up to 2.8mA rms (or +/-4mA).
If you want to use higher signal current, you should increase the cap value proportionately.

Thank you reply.
Is it necessary to use 100 ohm gate resistors AD1865 case?

 

[EUVL]

Yes, just to be safe.
No real down side.


Patrick

 

[suburra]

Patrick

If I wanted to put 2 PCM63 in parallel (piggyback), I have to change something in my CEN (Idss 8,5 mA_Riv 2K_ +/- 24V supply) ?
Thanks

Marco

 

[EUVL]

PCM63 has +/-2mA output. I would have used 1.5k R_iv in the first place.
But if you want to use two of those in parallel, you should half the value of the R_iv (750R), and double the bias (to say 16mA).


Patrick

 

[suburra]

Patrick,

Just as I thought...

I tried a Tx 2352 1.5 K and a Caddock 2K
With the 2K Caddock the resulting sound, for my ears, was better

Also,

I'll try Wima MKS XL 10uF and BG N 470uF instead of my current Nitai BP 220uF and will report

Thanks a lot

Marco

 

[suburra]

Patrick,

because it's impossible (for me) to find K170/J74 with 16 mA Idss (from Spencer max 9 mA), may i put another matched pair (8.5 mA) in parallel?

Or better try SEN with 18mA total Idss (each 2SK170 with 9 mA)?

Thanks

Marco

 

[EUVL]

Use 2 each K170/J74 in parallel to get 17mA.


Patrick

 

[PreSapian]

how could i get this working with tda1543? the psu for sen is not separate since i want a compact form factor. the ground is shared with 1543 so how to null dc on the chip's output?

 

[suburra]

Hi All

after long trial i finished my output stage of my Copland
It have CEN for I/V and Gherard Buffer for LPF

For Cen i choose Wima MKS 10uF for decoupling Cap and Texas Tx2352 for resistor I/V

The sound it's wonderful, grainfree and very clear, never heard CD's sound so well
So, Patrick, THANK YOU!!