close-circle
Close

"experimentorture" circuit

A project log for enhancement to depletion

DIY method to lower gate threshold voltage of regular MOSFETs, ultimately reversing it.

DeepSOIC 10/28/2016 at 00:265 Comments

This is the new circuit for torturing mosfets. The switches are actuated out of phase. As I explained earlier, the switches are now solid state (pairs of MOSFETs driven through photovoltaic optoisolator).

Discussions

Rene Jams wrote 06/19/2017 at 14:24 point

spice ::

https://goo.gl/photos/gixberTgQDARswfr5

it seems surprisingly there's no major flaws but PEAK GATE SPIKES

  Are you sure? yes | no

Rene Jams wrote 06/18/2017 at 21:14 point

just found out something "interesting" - as i was creating (read: designing / simulating) the near "zero current" (relatively) fast "voltage followers" -- an op.amp. 1-s

skip to asterixes (****** . . .) below ////

after 3 complex concepts/setups -- with LT1057(J-Fet input , Vdd-Vss max. 40V) LTC6240(CMOS input , Vs.tot max. 12V) -- i "added supply" to rails of measured circuit (Vs.tot 9V) so it took ±12V for J-Fet OA for it to be able to follow a high ohmic signal in 9V circuit and (!!!over the range) ±9V for CMOS . . . e.g

( 1 ) J-Fet :: ±16.5V to measure ±4.5 source . . . (!!! it ((the signal inside 9V circuit)) actually went another +9V over it's own pos. rail - so - correcting) ±13.5V source -- (((?? Vdd - 3V))) -- ?? d/s : "VOUT , Output Voltage Swing , RL = 2k , min. ±12 typical ±13 for ±15 Vs" . . . ****

( 2 ) CMOS :: ±13.5V to measure !!! ±13.5V source -- (((?? Vdd - 0V))) -- ?? d/s : "The LTC®6240 / 6241 / LTC6242 are single, dual and quad low noise, low offset,  r a i l - t o - r a i l _ o u t p u t, unity gain stable CMOS op amps that feature 1pA of input bias current. Input bias current is guaranteed to be 1pA max on the single LTC6240." . . .  damn! , a " late binding " / catch-up **** . . . 

. . . so i guess my 3-complex circuits are much more interesting than that ****

. . .

? suppose it's trivial for folks who have a "closer relation" with op.-amp.-s ? -- that the out of supply rails can't be followed simply and fast ??? . . .

. . .

!!!!!!!!!!!!! ok -- it's not complete catch up coz i have failed to follow the fast narrow pulses with J-Fet OpAmp before at lower than 9V Vs circuit . . . !!!!!!!!!!!!!

**************************************
not going more into depths of this -- we can summarize

**************************************
to follow/meter a fast and high impedance src. takes :
( 1 ) keeping the src. near (the OpAmp's virtual GND) / the supply median
( 2 ) keeping the Vs near what's specified on d/s

**************************************
( 3 ) . . . i just were concerned about the data-log.-r's capability combined with comm.'g 10M through MOS-gate . . . while i saw ►
+24V , 5k , MOS-gate , 10M , + undefined V

◄ -- e.g. your FET shunts it's gate through 5k to "minus sense" (another long story short :: ) to meter mosfet gate threshold takes a tricky biasing as at "down-left" on 

R1,R3,R5 (near "Vp") actually have narrow limits (as absolute , as relative) -- those limits are also MOSFET's type/channel-polarity dependent . . . i donno . . .

. . . maybe your circuit works fine but the 1uF and shunting +24V through TL431 while jumping around with mosfet gate . . . rises many questions for me F;i

**************************************

shortly -- this experiment likely can be improved

  Are you sure? yes | no

Rene Jams wrote 01/13/2017 at 14:35 point

another issue may rise from the OFF channel resistance or

from the 

dynamic OFF channel resistance of the SSR in hand --

 namly SSR in series with 10MΩ resistor - the ? time for internal to SSR transitions to form a next discreet state (might be up to several milli-seconds here -- if it does go completely off at all ...)

  Are you sure? yes | no

Rene Jams wrote 01/13/2017 at 14:32 point

a corrective note: in prev. the term AN(anode) states for the (*in circuit) positive terminal of the device (the TL431) the datasheet lists the opposite names according to the reverse protective diode in it's construction -- so to comlpy with the DS the prev. comment should read "... shows std. 10nF ((from Cathode to Adj.)) neg. capacitive feedback ..."

. . . however knowing the device (TL431) internals -- it acts more like the NPN in it's output -- i prefer(-red in a prev. post) calling the pos.*terminal of device-X (TL431) anode -- as without any advance knowledge here 1 presumably gets it right at 1-st that way . . .

  Are you sure? yes | no

Rene Jams wrote 01/12/2017 at 16:38 point
Is it 1µF capacitor?

Consider shunting that with TL431 with no limiting resistor . . . (there are few cases where such actually may have desired effect but usually it's less-equal to 100nF then)

(Please!) "recompute" "stability" here!

(the datasheet shows std. 10nF ((from Anode to Adj.)) neg. capacitive feedback for "431" the LT Spice likes values as ~100pF, 2n2 , 22n , .1u , 1++uF -- that's from AN→Adj. or from the voltage being regulated to Adj.INP ...)

  Are you sure? yes | no