SMPS -- (bad) design attempts

1. -- Testing out a random circuit from www

2. -- Testing loosely pre-calculated design (not fully conceptually correct)

---

[Eop]

Comparing Different 6LF22 to +5V supplies

Circuits ::

---

Cross Ref. ::

Param.	100	110	120	122	124	128
Description	Pos. collector-load linear-reg.	Pos. collector-load linear-reg. with fold-back	Pos. experimental Op-Amp regulated collector-load linear-reg.	Neg. j-Fet - emitter-load linear-reg. with fold-back	Pos. emitter-load linear-reg. with current lim.	Pos. VF-PWM buck conv.
Average Efficiency	51%	53%	57%	38%	28%	83%
Short-Circuit Current	120 mA	1.7 mA	NOT Ltd.	2.6 mA	* 57 mA	NOT Ltd.
Voltage Regulation	± 600 µV	± 250 µV	± 300 µV	± 15 µV	(± 550µV)	± 54 mV
Output Current : Nominal (Max.)	80 (85) mA	80 (145) mA	80 (150) mA	50 (105) mA	* 50 (54) mA	200 (>400) mA

---

Source Circuits ::

↑↑ src :: ISBN 5-440-00345-2

↑↑ src :: GE Transistor Manual 1964 ↑↑

↑↑ found from :: LED Torch Circuits ↑↑

---

[Eop]

about using BjT-s as capacitors --and-- diodes

bugs found -- see update below

a past experiment to verify the speed of different transistor's diode configurations ...

---

! Update ::

---

[Eop]

The КР140УД1208 (µA776) variants' Comparison

as by control current range -- i can telll -- i still don't have the Model 4 the Op Amp in question ...

anyway , here's the variations in between what i've come up with ::

↑ and this is better ↑ than nothing -- because it reveals the SR being dependent on the supply voltage and the control current . . . ← but also the Outp.Zero or Balance changes along . . . so the uA776 is a head-ake . . . -- to be set up with the floating supply --and/or-- with the wide-varying signal amplitude/-frequency range

► a small signal pulse response test -- it necessarily won't have the best performer being the best match2® dependency ...

► a small signal sine test -- reveals (as and brought below) the offset and phase/SR differences
-- the unit is -- (µ-"ns") per (µA)² -- the delay is revealed indirect from ::
dt(for 776) = dt(776.Outp , Fn.Gen.Outp) – dt(LT1013.Outp , Fn.Gen.Outp)
near the ~AC.zero --or-- the SG (Signal Ground) , e.g. -- near/@ ±0V in the circuit below

---

[Eop]

9V-to-5V Step-down concep design

the thing

switching wave-forms at max.-min. load

---

[Eop]

? Inductorless Flash-light energy efficient ?

about ::

src. :: Single-Cell Programmable Constant Current Boost LED Driver

src. :: Питание Для Светодиодов (Started by vanPelt, April 26, 2012)

---

Principal functional analog (or better)

 -- about the 500mΩ being enough to regulate the current

---

[Eop]

revised the yy78Lxx 5V variants

tests so far ::
PS! -- although the following setups perform well the real tings may not coz the Spice models use highly accurate fixed current sources - - - while there's no indication the real 78Lxx-s had any implemented //// the last time i tested my real 78Lxx-s showed an ulta poor performance where the voltage started to drop already at !!!above 30-/40 mA drain and significantly differed from the nominal at 50mA (while i didn't notice them getting hot -- so that a possible thermal prot. feature could have had engaged ???)

partial apx. match to d/s ::

---

[Eop]

Improved the К554СА3 model

some tests ::

PS! : because the low CE voltage drop transistors used i cannot guarantee the device will work down to that it does in simulations but i guess it does significantly below the specified ±13.5V

PS#2 ! : the LM311 model is obviously an over idealized one . . . is just included for the comparison

---

! Update :: "Unlimited" frequency range oscillator

---

[Eop]

Improving LM339 + "tunable" osc variant

► the tunable osc. as by selecting the RC net-wk

► the tests that lead to (as by what i've tested so far) "improved" v. of LMx93

► 2MHz - TEST (driven from realistic src. 74HC04 -- 1 gate has just about the power required!?)

---

[Eop]

Couple of osc.-s based on LM339 planned for 2x AA voltage converter

the bad thing with the next is the threshold levels're floating with the supply voltage and specific set of components (makes it difficult to wick the reliable PWM out of it) . . . the concept design → ▼

INPUT below Negative rail - oscillator ▼

---

[Eop]

YX8018 ~about "guessing the operation"

→NB←! the following has pretty much nothing to do with the actual chip !!! ←
i just matched some known or guessed data into random simulation substitute

from what we can conclude ::

1. the device operates best at certain fixed supply voltage (there might be one built in to YX8018)
2. the best efficiency is occurring below the certain output power = you should not replace the inductor to lesser inductance one without making sure :
   a) that there's enough power to drive the inductor (somewhat aided by the 1µF and 5µH caps inductors at the schematic) ,
   b) that the inductor won't be driven into saturation
   ( ↑↑ ← both are primarily indicated by a notable drop in efficiency compared to initial -- however the lesser inductance does, by the idea, give a* higher output power and usually there is some efficiency drop for "trade" to such* )
3. the onboard inductor in my particular solar lantern has 4Ω series resistance (it might be useful to protect the mosfet from peak power ? from accidental out of cycle re-trigger . . . or provide some voltage regulation for more up to charge battery) , otherwise there's a plenty of room to install a better inductor (such might also cause the over-voltage at the switch drain -- it must be tested out at max input power)

---

[Eop]

Predicate research for using the LED as a photodiode

in prior of doing the following i measured the target Blue LED that did output some millivolts but no nanoamps , then a high efficiency Blue LED that i used for below experiments (← the series array of theses gave pretty much alike data than the single one) . . . then the iR LED parallel array from the (used for . . . ?? pre-clearing the image in) laser printer , it output over 1V and i guess several microamps . . .

. . . all that illuminated from the narrow beam 1W Power LED

the conflicting data (with web) is that when illuminating the LED it outputs fw.-polarity voltage -- e.g. [+] from anode and [-] from cathode * . . . all 3-types of the above without the exception ???

Note* ::  no conflict (http://www.ques10.com/p/9330/explain-with-neat-diagram-construction-and-worki-1/) , conflict ( . . . there was a site that actually textually claimed that ? intentionally or not . . . can't re-find it at this time . . . and it was not the only one such . . .)

---

first couple of web found setups (slightly modified)

and my own guesses of j-Fet setups

---

[Eop]