Author Topic: Zero current voltage regulator?  (Read 1885 times)

0 Members and 1 Guest are viewing this topic.

Offline fonographTopic starter

  • Frequent Contributor
  • **
  • Posts: 369
  • Country: at
Zero current voltage regulator?
« on: August 03, 2018, 12:22:20 pm »
Imagine a two plate capacitor,one plate number 1 is connected to our device X,its stationary and other plate number 2 is sometimes moving closer towards the first plate,then returns back.When it gets closer,as it moves towards the plate 1,that will induce voltage and current in the stationary plate 1.

I am looking for electronic device X that will sense the increased voltage and create its own voltage of equal amplitude but opposite polarity so it cancels the increasing voltage caused by plate 2 moving closer.I dont know how to call the device X,I guess its some kind of voltage regulator comparator balancer feedback thing... I dont know,but I think such thing is different from typical voltage regulator that regulates flowing DC current,my device X should prevent any current flow by applying voltage of opposite polarity,kind of like static voltage regulator.
« Last Edit: August 03, 2018, 12:23:58 pm by fonograph »
 

Offline dzseki

  • Frequent Contributor
  • **
  • Posts: 524
  • Country: hu
Re: Zero current voltage regulator?
« Reply #1 on: August 03, 2018, 01:54:12 pm »
Not sure if I understand perfectly, but it looks like you need some sort of P(ID) controller built around your capacitor where the plate voltage is your input parameter. You need some fine FET input opamp for the voltage sensing. INA116 would be a good example for this with only 3fA input bias (typ.) it has almost near zero input current...
HP 1720A scope with HP 1120A probe, EMG 12563 pulse generator, EMG 1257 function generator, EMG 1172B signal generator, MEV TR-1660C bench multimeter
 
The following users thanked this post: fonograph

Offline jcw0752

  • Contributor
  • Posts: 35
  • Country: us
Re: Zero current voltage regulator?
« Reply #2 on: August 06, 2018, 05:26:12 am »
Hi fonograph,

I am not sure I agree with your physics of the moving plates. Just moving plates relative to one another will not induce current or increase voltage. If a voltage is externally being applied to the plates there will be a current flow as the plates are brought closer as the decrease in distance creates an increase in capacitance. Capacitors hold charge because of the interaction of the imbalance of electrons on the adjacent plates. If you have a capacitor charged to a specific level and then you add more capacitance the voltage will drop. In fact, if the capacitance is doubled the voltage will drop by one half.

John
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Zero current voltage regulator?
« Reply #3 on: August 06, 2018, 02:07:14 pm »
That is an application for a charge amplifier.  Typically a low input current operational amplifier configured as an integrator is used.
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 8275
  • Country: ca
    • LinkedIn
Re: Zero current voltage regulator?
« Reply #4 on: August 06, 2018, 05:30:03 pm »
I use a simple N channel J-Fet with the gate connected to GND.  I used to use J110, J111, J112, J113, each having a different v-gs giving a different voltage output.  From Jfet-Jfet, the voltage varies a bit, but, each Jfet will act as a linear regulator and hold their voltage.
(I once went through all the available J-Fets at Digikey, I had a selection of output voltages from 2.1v to 5.4v.  Sorry, I no longer have that table list of part numbers VS output voltage.)


The input is the drain and the output is the source.  The maximum input voltage is what the Jfet can handle.  The minimum input voltage is the actual output voltage, (in other words, 0v dropout regulator), otherwise the output voltage = the input voltage anywhere below that point.

The quiescent current to GND is in the low Fento-Amp range, in other words, nothing.  Perfect for projects needing regulation with small watch battery lithium cells.  Even the smallest batteries would last over a decade with no drain at all.  Note that if you want low Fento-Amp currents, the quality of flux residue cleaning on you PCB now counts, you'll waste current if it is not done really good!

If you need to supply more than 25ma, feed the output of this regulator into an NPN transistor's base in an emitter-follower configuration, or, Vgs match a few Jfets and tie them all in parallel.


Now for your application.  Based on your description, 'I think', what you mean to do is use the same Jfet as above, but have a cap with bias correcting battery in series to the gate.  Your source load would be on the other side of that cap.  However, this will have no gain, and there may be a fixed offser error on the output, but, the in to out would have close to nothing in current.  And depending on the cap, ony offset charge will hold very long.  However, any build of charge in that cap beyond the Jfet's capabilities may destroy the Jfet, so you have a narrow operating range.  (The capacitor is your 2 plates)  Use the output of the Jfet to feed your opposite plate directly, or, take that output and feed a standard op-amp to invert that signal to make that repelling voltage.

Your other choice is to look for Fento-amp-input op-amps and use 1 as a voltage follower, then feed another op-amp in an inverter configuration which then feeds your capacitive plate.
« Last Edit: August 06, 2018, 05:52:29 pm by BrianHG »
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 8275
  • Country: ca
    • LinkedIn
Re: Zero current voltage regulator?
« Reply #5 on: August 06, 2018, 06:46:08 pm »
Note that if you need a high voltage range beyond +/-5v, and stay within the Fento-amp range, your only solution may be a vacuum tube amp.  You can expect a +/-50 to +/-100v range here depending on design and you would be fairly immune to static discharge.

 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 8275
  • Country: ca
    • LinkedIn
Re: Zero current voltage regulator?
« Reply #6 on: August 06, 2018, 07:32:10 pm »
The highest voltage Jfet currently is 55v.  It is possible to make a 24v range (+/-12v) inverting and non inverting voltage following amp with one of these.  Though, you might need to deal with the voltage bias depending on your application.  You can also make a single positive voltage following amp with 50v range (+/-25v) and feed that into a large voltage op-amp to invert and offset correct the signal to get you opposite following voltage output.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Zero current voltage regulator?
« Reply #7 on: August 07, 2018, 02:08:05 am »
Note that if you need a high voltage range beyond +/-5v, and stay within the Fento-amp range, your only solution may be a vacuum tube amp.  You can expect a +/-50 to +/-100v range here depending on design and you would be fairly immune to static discharge.

Very special vacuum tubes were used for electrometer applications.  The better more typical vacuum tubes had input bias currents in the nanoamp range.

The highest voltage Jfet currently is 55v.  It is possible to make a 24v range (+/-12v) inverting and non inverting voltage following amp with one of these.  Though, you might need to deal with the voltage bias depending on your application.  You can also make a single positive voltage following amp with 50v range (+/-25v) and feed that into a large voltage op-amp to invert and offset correct the signal to get you opposite following voltage output.

JFET input bias current increases at high drain voltage so their operating range for low input bias current is limited.  What can be done though is to bootstrap the drain voltage so that it follows the source.  Solid state electrometers do this to achieve hundreds of volts of input range at 10s to 100s of femtoamps.

Linear Systems makes small signal MOSFETs which could be used also but I do not know that they would be any better.  They are tested to the same gate leakage specification as low leakage JFETs.  One could also use an LMC6001 operational amplifier.
 
The following users thanked this post: BrianHG


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf