|
 |
Analysis help for full wave active rectifier - RF Cafe Forums
|
Mr.Whatever
Post subject: Analysis help for full wave active rectifier
Posted: Thu Oct 19, 2006 12:32 pm
Captain
Joined:
Fri Apr 07, 2006 6:10 pm
Posts: 18
Hi all! I'm trying
to see if I've done my analysis correctly.
For the first opamp,
I have the following, assuming positive input:
VA1 = VIN * (R2
/ (R1 + R2))
VA1 = VA2 = VOA (Voltage follower)
Therefore,
VB1 = VOA = VB2.
Since I know VB2, I can calculate the current
going thru the resistors.
-I3 + I4 + I5
I3 = (VS-VB2)/R3
I4 = (VB2 / R4)
I5 = (VB2 - VOB)/R5
Then, I combine all this
and get a VOB / VIN relationship:
VOB = VIN * ( (R2/(R1+R2) +
R2*R5/(R4*(R1+R2)) - R2*R5/(R3*(R1+R2)) - R5/R3)
However, when
I try plugging in my resistor values, VOB is always equal to zero.
What am I doing wrong? My simulation clearly shows an active fullwave
rectifier.
Thanks!
Top
fred47
Post subject:
Active RectifierPosted: Thu Oct 19, 2006 3:40 pm
General
Joined: Wed Feb 22, 2006 3:51 pm
Posts: 104
Hi!
It's
hazardous to rely on generic models, rather than specific SPICE models
of specific circuits - especially when you're depending on nonlinear
behavior of linear ICs (like OpAmps). Most opamp input active ranges
are less than Vcc to Vee, likewise the outputs. The few for which that's
not true are specially denoted by the term "Rail-to-Rail", for input
or output or both. Even there, the outputs have some restrictions.
When you say, "when I try plugging in my resistor values", do you
mean in a simulation or in a breadboard? It sounds like you mean in
a breadboard, and that the physical results don't match the simulation.
Reality wins over simulation every time.
My general rule for
simulations (SPICE or other) is that if they tell you a circuit won't
work, they're probably correct. If they tell you that a circuit will
work, it doesn't mean there aren't a lot of hidden problems that might
keep it from working.
Good Luck!
Fred
Top
Mr.Whatever
Post subject: Posted: Thu Oct 19, 2006 4:41 pm
Captain
Joined: Fri Apr 07, 2006 6:10 pm
Posts:
18
Thanks, fred47!
Actually, this all refers to a simulation
environment. I just wanted to see how the configuration actually rectifies
the input. I re-did my work and realized that I had a problem with one
of my signs (e.g. treated it as a positive instead of a negative).
By the way, regarding your comments on SPICE, do convergence problems
in the simulation account for the actual circuit not working?
I've been getting a lot of convergence errors. From what I found
out so far, it's possible that it may be due to some settings in the
simulation environment.
Top
fred47
Post subject:
Convergence ErrorsPosted: Thu Oct 19, 2006 5:47 pm
General
Joined: Wed Feb 22, 2006 3:51 pm
Posts: 104
Hi!
Lots
of things can cause convergence errors. One of the most common is positive
feedback; you're right in thinking that sometimes it's simulation parameters
or settings.
Active full-wave rectifiers usually require matched
resistors.
Good Luck!
Fred
Posted 11/12/2012
|
|
|
|
