[00940]

Here's for the record what I'm planning to use. To give the deserved credits, the I/V is from Cauthemoc and the amp serving as receiver is a Gilmore's dynalo. Nothing's mine

The I/V will be wired p2p ... there are only 80 bc550c/560c to solder.

It's designed with the pcm1798 in mind. The pcm1794's outputs are a bit too hot for this thing. The 4ma of the pcm1798 are easier to manage. The emitters followers are biased at 10mA. The r17/r27 trimmers have to be adjusted to read 0V at the emitters outputs.

Around the emitter and the amp, there's a 3rd order bessel analog filter. It gives around 0.7db of attenuation at 20db, and 48db at 320Khz (first alias with 8x oversampling and 44.1khz material). I'll post the sims for amplitude and group delay later on.

The output stage has a gain switch, to use it either as preamp or headphone amp (2vrms vs 6 vrms) and the volume can be controlled using R1.

I don't believe much in simulations but LTspice told me that the thd should be under 0.01% (for the I/V stage).

The values:

r16=r26= 1K (I/V resistor)

r14=r24= 1.5k
r15=r25= 2.5k
c1=c5= 2.2nF
c2=c6= 1.1nF

r2=r3= 22k
r4=r5= 10k
r6=r7= 33k
c3=c4= 90pF

r1= 10k log

 

[00940]

As promised, the amplitude and group delay graphs, at the output of the dynalo:

 

[Pars]

Yes, Christian had shown that stage to me as well. I was prototyping a slightly earlier (and less complex) stage of his, but kinda got sidetracked. I pretty much have the layout done, and thought I would etch a board and try it out..

Which brings me to my point:

Quote:

Originally Posted by 00940 /img/forum/go_quote.gif The I/V will be wired p2p ... there are only 80 bc550c/560c to solder.

That should make for a fun layout! Any reason you picked bc550/560 over 2sa970/2sc2240? Keep us informed...

 

[amb]

Someone must have a current mirror fetish... lol

 

[Cauhtemoc]

Quote:

Originally Posted by 00940 Here's for the record what I'm planning to use. To give the deserved credits, the I/V is from Cauthemoc and the amp serving as receiver is a Gilmore's dynalo.

I don't deserve full credit for the I/V. I ripped some parts from Wadia, which in turn (if I remember correctly), ripped some parts from Jocko Homo.

Looks nice anyway. Good job.

 

[00940]

Pars: you know, if you divide the I/V into 4 parts, those can be quite small. And if you're using protoboard with strips, actually quite simple. See pic below. The board is 3cm/7cm, everything on a 0.1inch grid.

The BC550-560 isn't set in stone. It's mainly a cost/ease of getting them thing.


A question though: I now wonder if the third pole of the bessel shouldn't be a cap in parallel with the I/V resistor... The filter response is now altered (not much but still) by the varying R1 pot.

 

[barmanekm]

Hi, could you repost schmetics? The picture links from the first posts are gone.
Thank you.