Increase a voltage value every hour.

Thread Starter

girafa

Joined Oct 5, 2018
27
Hi,
I want to increase a voltage value every hour.
After one hour the voltage value goes from 0v to 1v. After an hour passes to 2v and always like this.

I'm not sure how to do it.

What will be the best way to do it with analog electronics. Using Operational Amplifiers?
 

Bernard

Joined Aug 7, 2008
5,257
Maybe a 4060 counter & 4017 feeding a V divider ? How many hours are required?
 

spinnaker

Joined Oct 29, 2009
7,815
Hi,
I want to increase a voltage value every hour.
After one hour the voltage value goes from 0v to 1v. After an hour passes to 2v and always like this.

I'm not sure how to do it.

What will be the best way to do it with analog electronics. Using Operational Amplifiers?

Lookup voltage ladder.
 

Thread Starter

girafa

Joined Oct 5, 2018
27
Okay, so it's like this
Voltage increases every hour through a pulse of 0-1 to 12 hours.

The current in the load will be too low to enter a voltage follower.

I can not use digital circuits.
 

danadak

Joined Mar 10, 2018
3,573
Something like this -



Staircase generator, lots of variations on this. Think
integrator for one polarity. If you want to reset you
would add mosfet across C2.

Leakage is your enemy here, bias current of OpAmp,
diode leakage, capacitor leakage. Choose components
wisely.

From a guru -

https://www.electronicdesign.com/analog/whats-all-soakage-stuff-anyhow

https://www.electronicdesign.com/test-amp-measurement/whats-all-teflon-stuff-anyhow

https://www.ti.com/ww/en/bobpease/assets/www-national-com_rap.pdf This
has the above articles + more on these topics.


Regards, Dana.
 
Last edited:

Thread Starter

girafa

Joined Oct 5, 2018
27
Something like this -



Staircase generator, lots of variations on this. Think
integrator for one polarity. If you want to reset you
would add mosfet across C2.

Leakage is your enemy here, bias current of OpAmp,
diode leakage, capacitor leakage. Choose components
wisely.

From a guru -

https://www.electronicdesign.com/analog/whats-all-soakage-stuff-anyhow

https://www.electronicdesign.com/test-amp-measurement/whats-all-teflon-stuff-anyhow

https://www.ti.com/ww/en/bobpease/assets/www-national-com_rap.pdf This
has the above articles + more on these topics.


Regards, Dana.

Yes it's anything like that, but the problem is the leakage currents. That is why I want another solution but I do not know which one. We are talking about very long times.
 

danadak

Joined Mar 10, 2018
3,573
Today's opamps will give Femtoamp performance. Diodes single digit pA performance.
Makes me wonder if a jfet used as a diode would be possible.

http://www.ti.com/lit/an/sboa058/sboa058.pdf


https://www.ti.com/ww/en/bobpease/assets/www-national-com_rap.pdf

Go to page 4 (not labeled) for a list of topics, page 45 (not labeled) on Femtoamp
discussion.

http://yawp.com/ref-library/articles/unread/Design-femtoampere-circuits-with-low-leakage---Part-2--Component-selection.pdf

Also look at Pease piece on teflon..


Regards, Dana,
 
Last edited:

Bernard

Joined Aug 7, 2008
5,257
Take a clock movement , attach a magnet to hour hand. Arrange 12 reed switches so that each one will clos
every hour. Series connect 12 1k resistors & connect one end to - other end to +12 v. Connect each R junction
to one end of a reed SW in a consecutive arrangement with other ends all tied together forming the output.
Does this qualify as analog?
 

ebp

Joined Feb 8, 2018
2,332
If a linear ramp is acceptable, 1 nA for 43 200 S for an final voltage of 10 V would require only a 4.32 µF capacitor.

A 10 µF polypropylene capacitor rated at perhaps 100 V would probably have leakage low enough to keep the error quite small. A 2 nA current source wouldn't be especially easy, but probably isn't extremely difficult. A CMOS op amp with the non-inverting input pin soldered directly to the capacitor pin (not via a PCB) would probably be OK at room temperature to buffer the voltage. Input bias current in the range of 3 or 4 pA at room temperature is a reasonable expectation, for an error of well under 1% with a 2 nA current source.

I'd probably look at specs for PP capacitors and perhaps select higher capacitance. They are big, but not really huge. 22 µF might be a reasonable choice and would get the current up to a little over 5 nA.

If a stepped waveform were required, I'd use a much higher charge current for a perhaps a second once every 10 or 15 minutes (or 60 if that were an absolute requirement). I'd consider a reed switch with separate coil (so I could be sure the glass tube was very clean) to connect the charger. A CMOS 555 would probably work OK - but the 555 has a digital output and a relay is a digital device, so they may be prohibited. As Dana suggested, a JFET makes a good low-leakage diode if the right type is selected. Charge could be injected via such a diode.
 

Sensacell

Joined Jun 19, 2012
2,437
Use a small gear motor to turn a potentiometer.

No drift, leakage or non-linearity!

And no digital stuff either!
 

ArakelTheDragon

Joined Nov 18, 2016
1,265
Sounds like this is homework.
Why can't you use digital circuits?
It is certainly homework or a coursework, but the problem with these things is that normally they have the purpose of teaching you something. Just finding the circuit might not be the best decision.

Otherwise it can be done in many ways, the easiest one is a capacitor with a resistor that gets charged with "1VDC" per hour, until it reaches the 24 hours you want, but it will not get reset and start from "0" again unless you put something to discharge the capacitor fast. It can be made with a regulator with a capacitor and a resistor in the feedback of an OPAMP and the load in the output.
 
Top