0
\$\begingroup\$

I have a signal which can have a different offset voltage based on the requirements. However, I need the voltage to start at zero.

normal signal

Since the offset is variable, I was hoping if there are any techniques/ideas how to design a analog circuit which makes a variable offset in such a way that is ensured that the voltage is always positive (so starting at zero).

I am aware that I might be asking something which is impossible but I hope someone designed something similar.

\$\endgroup\$
3
  • \$\begingroup\$ not entirely the same, the problem in my application is that I have a easy way to make a trapezoidal signal, however I want to have the flexibility to have a larger fall time than rise time, as a result the signal starts to have negative dc off set, which i need to compensate. \$\endgroup\$
    – Navaro
    Sep 24, 2018 at 23:56
  • 1
    \$\begingroup\$ How fast can the "variable offset" change? It's important to know. \$\endgroup\$
    – jonk
    Sep 25, 2018 at 1:54
  • \$\begingroup\$ Is this signal part of a complex waveform, such as NTSC video where the SYNCTIP defined the "zero" voltage? \$\endgroup\$ Sep 25, 2018 at 3:47

1 Answer 1

1
\$\begingroup\$

This has been done for over half a century on TV sync pulses that are AC coupled and then DC clamped to the ground on the back porch to set the black level.

The way it works is the leading edge to your desired 0V level is differentiated to trigger a very short pulse and then AC coupled for this "active clamp circuit." It shorts the input signal to ground with an open collector or open drain and charges up the cap with the DC offset voltage. (T=RdsON*C) to the achieve 0V out. then the FET switch is opened until the next leading edge. Obviously, any load current must be minimized by a low bias current, low input offset Buffer to prevent decay and reclamp charge current.

\$\endgroup\$
11
  • 1
    \$\begingroup\$ do you have any reference which I could use to get a feel of the idea. And will this distort my linear trapezoidal signal? \$\endgroup\$
    – Navaro
    Sep 25, 2018 at 0:07
  • 1
    \$\begingroup\$ not if done right, define distort 1% glitch? did you not yet search for active clamp cct? If not, get it done. \$\endgroup\$ Sep 25, 2018 at 0:45
  • 1
    \$\begingroup\$ @Navaro Perhaps here? Wiki clamp. Or did I misunderstand things? \$\endgroup\$
    – jonk
    Sep 25, 2018 at 1:49
  • 1
    \$\begingroup\$ @Navaro THey even have an IC for this diodes.com/assets/Datasheets/ZXFV4089.pdf \$\endgroup\$ Sep 25, 2018 at 2:33
  • 1
    \$\begingroup\$ Thanks for all the comments....the circuit in the link provided by @jonk seems to work. \$\endgroup\$
    – Navaro
    Sep 25, 2018 at 19:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.