Monday, December 28, 2015

Virtual Ground tracking circuit

? I 1-st thought it´s impossible . . . but it like does something

. . . it sums the wrong trace !!! voila !

here´s a modified v.

and that´s more like it F;Y


Sunday, December 27, 2015

a flip-flop experiment

a ? "T-trigger" from two NAND gates ::

►► the idea behind it was to use a half of 7400 for one "pulse driven RS-JK-trigger" . . . which might be the case if this thing can be opt-/minmized to a lesser component count (total design/build-cost) and made a lot more noise immune (can be operated in realistic env.) - it will start to oscillate if it´s shoulders components won´t match too well (which means almost precisely) -- here i just found the reasonable upper frequency limit for (about 44MHz for inp.-clk.)

the Q0 , Q1 are inter-coupled to get high-fq. readout - with lesser frequency there´s no need to couple them - and originally they were added to confirm the valid readout can be had from K and/or M gates -- so , the fast ver. of this trigger requires the full 7400

. . . just had that question


Saturday, December 26, 2015

Adjustable duty/frequency OpAmp Oscillator

1-ce you get customized with these things (OpAmp-s) - they're nifty to use for all sort of tricks -- the below design was chosen due it's adjustability × lesser number of components compared to it´s TTL or CMOS variant

 the interesting point to note here is that the КР140УД1408 : LM308 specify their lowermost supply range at ±5V(+10V) : ±2V(+4V) respectivly -- while the voltage buffer/-follower configuration still works with +5V for КР140УД1408 without any fq. compensation ??? here´s the component level model of neither* УД14 nor LM308 OpAmp-s exactly (*also coz "we" don´t know the exact transistors used in the actual thing -- nor does the circuit exactly match neither -- about What? = it's close enough to simulate both (only some 10% of Spice simulations hit the real life - the small alterations to OpAmp model won´t significantly change that figure))


yet another LDO 5V regulator

the Regulator :: /̲!̲\̲ this thing seems to perform better than LE50A (STMicroelectronics) , LM78L05ACM (National Semiconductor Corporation) , MC78L05AC (Motorola, Inc.) /̲!̲\̲ when powered by 9V 6LR61 alkaline -- in a sense of IOUT/ΔUOUT . . . . . . . . . history of :: the goal : steady supply for OpAmps powered from Lo-Power battery ►► everything is good as long as you have to implement the current limit . . . so this one has 1 unintentionally "built in" (adjustable by transistor-switch current source) ::

the actual transistor used for switch is ancient 2SD1252A (Panasonic)the custom zeners are random Red-LED + Д9К (apx. 2V + 320mV drop) other BJT-s are 2N3906(-04) but you can use a better ones such as fixed beta 2SA1015 , 2SC1815 ...

!note  :: altough theLE50 is labelled as "Very Low Drop" it actually performs worse than 1 of those 78L05-s and the all three b_a_r_e_l_y manage the 150Ω load with reasonable output (4.8 dn2 4.7V) . . .
 . . . when the brought regulator does 100Ω with about 10mV output-drop and acts similar to above 3 at about 26Ω with 175mV drop e.g. 4.83V output -- though /̲!̲\̲ it does not have thermal protection and it's current limiting is poorly studied /̲!̲\̲


Thursday, December 17, 2015

theoretical wall ~AC line detector

since there is reason to believe the ®´s j-fet input stage rather attenuates than amplifies
the E - field variations i set up a simple band-pass based 1 ::

this might still require extra input stage (for deep buried wires)
+ experiment with the layout , sensor , 3-sholds´ sensitivities before we can say if it´s for any good

updated an old Sq.-wave oscillator to a voltage controlled one -
to be mixed with prev. ~50Hz output ::

it is good to pre-test and choose + adj. the optos to the same transfer ratio ,
everything else shouldn´t be too much trouble ?


Tuesday, December 15, 2015

Single v. Double Op-Amp "Rectifier"

Intuitive design ::

Modified from the book ::

/̲!̲\̲ "Ideal DIODE" /̲!̲\̲ ::

e.g. below 24k ÷ 2 = 12k Hz (Dbl v.) OR 2k ÷ 2 = 1k Hz (Sng v.)