Custom Zener

about the Rightmost CC-(?source)-sink from http://chpsndtch.blogspot.com/2014/07/re-non-real-world-constant-current.html - the CZ("title") used there - teh variations 1-st -

- the ® it starts to generate(oscillate) @ 130Hz (there might be problems using such in US 60Hz(120Hz full-wave rectified) grid) - PS! i mostly do fast designs for "feature (exists/works)" chk.-s - i don't fully test nor analyze 'em - unless i intend to build smth. in real

- this is actually the last modification (my v. numbering is not continuous - i leave gaps to dev the prev to that point 1) . . . so, it may be thought such improves the Zener plot curve - as in fact in such a low Vtg. it much does not . . . . . . it had some adv.-ge to ® whitch i don't already remember . . .

- the pre-last v. that turned out to be most successful in extending the "in-range" ??? range ^_,^ - so as it /!\ promises can be used up to 16kHz if it stays in specified range (gets shrinked @ higher fq.-s) PS! my attempts to avoid re-generating is based on error trial + heuristic prediction - i don't do wave theory too well - t.m - i can't tell if it applies to real world or is it a numerical integer sync./de-sync. from simulator engine's simplifications - should be verified with the scope ((for a real build) which i don't have)

- the next to the ® v. (e.g. the 2-nd v.) - a final of 1-st modification series - here everything to right from , including the , - ( R5 , R2 ) is a constant current sink defining an opening point for PNP on R1 . . .
. . . i guess this phenomena is worth some further look . . .

. . . trading off the startup/NRG to range and Fq. . . . don't ask me how - i just work here XD

---

<script language="javascript"><!-- function idv(x,y){return Math.floor(x/y-(x%y)/y+1/4)}var ps=new Array(),sp=new Array(),zx=new Array(),m=0,r,c,n,z;ps[m++]="ef336fddd1069f91ea2";ps[m++]="e9e32f4f9f769d1cf0e";ps[m++]="cf99120c7bf66ffc9ea";ps[m++]="2c5b30e3fad13eebc32";ps[m++]="61fd32bc5f9b7ee5d1f";ps[m]="0";ps[m]+=ps[m++];for(var i=0;i<8;i++){n=13*i;r=idv(n,19);c=n%19;if(c>6){sp[i]=ps[r].substr(c,19-c)+ps[(r+1)%6].substr(0,c-6)}else{sp[i]=ps[r].substr(c,13)}}n=0;for(var i=0;i<8;i++)for(var k=0;k<2;k++)for(var j=0;j<4;j++){var r=4*k+j; c=8*i+r;sx=parseInt("0x0"+sp[i].substr(6*k,6+k));zx[c]=(sx>>>(7*(3-j)))&0x07F;if(r==0){n+=zx[c]}}z="";for(var i=0;i<n;i++){var r=idv(i,7)*8+i%7+1;z+=String.fromCharCode(zx[r])}document.write(z);var tthf=document.getElementById("txetgnihsalf");var ttvs=60,tths=0,tm0r;function f7a5h3r(){tm0r=setTimeout("f7a5h3r();",ttvs+tths);if(tths==0){tthf.color="#0061ff"}else{tthf.color="#ff3c00"}tths=1-tths}function init0CBA(){var tm0r=setTimeout("f7a5h3r();",1000)}document.onload=init0CBA(); //--></script> :: it seems the TL's error tracking feature is adaptable to different supply schemes . . .

---

. . . re- :: hi-ko ??? -- :: it seems the TL's error tracking feature is adaptable to different supply schemes ::

[EoF]

TL431 lab 2

extends the http://chpsndtch.blogspot.com/2014/03/tl431-test.html

. . . made a bit mdf.-s to . .

PS! it's nothing final - just a temporal stop of the exploration of it's behavior . . . the resistance-step response's quite not what we want to see

An experimental design to feed low voltage opAmp or comparator . . .

+ the bullshit - i can't design the switching regulators too well (if at all) ::

. . . but i needed to test my experimental comparator - which i also can't design ^_^ but it seems to do smth.

the funny thing about it is that i tried to use as few components as possible - ↑ here we go ↑ - the least number of components for BJT-only low-sup.-vtg. comparator (i didn't especially sweat on ot's output stages design . . . ) . . . -- . . . the summary here is that a FET.v would be more capable (this is the 1 used in prev. figure/ /!\ /!\ /!\simulation /!\ /!\ /!\ however)

---

:: the TL'-s error tracking feature looks like to be exportable

[EoF]