Simple circuit to convert Bi-polar Square wave to Unipolar

R

rylmp2k1

Guest
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
 
rylmp2k1 wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
Click to expand...
How about an isolation transformer?
 
On Sat, 24 Oct 2009 08:28:30 -0700 (PDT), rylmp2k1
<royalmp2001@gmail.com> wrote:

Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
Click to expand...
This is a classic "DC restore" circuit where the baseline is set by a
DC power supply:


in---------cap--------+----------------out
|
gnd | gnd
k diode
a
|
|
|
+ dc supply
-
|
|
|
|
gnd


Set the DC supply to roughly +1 volt to get a small positive baseline
on the output. You'll need a big cap to get to 1 Hz into most loads.

You could actually derive the DC supply from the sig gen output. That
would be cute.

John
 
On Oct 24, 11:51 am, "Greg Neill" <gneil...@MOVEsympatico.ca> wrote:
rylmp2k1 wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks

How about an isolation transformer?
Click to expand...
I need to avoid passing the signal through a transformer or DC
blocking cap as it screws the square waveform up.
I need a good square waveform all the way from 1Hz to 2MHz - but
thanks for taking the trouble of replying.
 
rylmp2k1 wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
If you don't mind inverting the signal..
Click to expand...

Drive a common emitter inverting circuit via a bias
R so not to over drive the NPN transistor..

Something in the line of a 4.7k sounds good.

Put a SI diode from base to common so that the
(-)input gets redirected to common and not cause the
transistor to go into zener mode.

Select a sufficient load R for the collector from a fixed
DC source some where..

The collector will be your output.

When the transistor goes into saturation state via a
(+10) through the bias R, you'll get ~ .25 volts on most
commonly used switching transistors. Maybe less.

When input goes (-10), it'll get sucked up in the SI diode
and force the NPN to be off, which will allow your load R to
give you what you need.

You'll get a .2 to What ever DC source you select as your square wave.

If you don't want it inverted, just pass it through another inverter
stage..
 
rylmp2k1 wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
Click to expand...
A clamp? Simply a diode? You can cut off the negative part because it is a
square wave. It will only reduce the peak to peak amplitude.

You can use a diode to simply clip the negative half. If you don't like that
use a diode clamp or an active diode.
 
Jon Slaughter wrote:
rylmp2k1 wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a
diode, the anode connects to the -ve of a small value electrolytic
capacitor and the +ve goes to the ground output. My resulting
signal is taken either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of
the cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks

A clamp? Simply a diode? You can cut off the negative part because it
is a square wave. It will only reduce the peak to peak amplitude.

You can use a diode to simply clip the negative half. If you don't
like that use a diode clamp or an active diode.
Click to expand...
Another method might be an optoisolator(the bidir kind)...
 
On Oct 24, 11:28 am, rylmp2k1 <royalmp2...@gmail.com> wrote:
Hi everyone.
  I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output.  My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset.  I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive.  If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
Click to expand...
You state the output is variable 0 to 10v. Can't you adjust it as so?
Does it have an offset control? Care to share more details that you
are not telling us yet?
regards,
al
 
On 2009-10-24, rylmp2k1 <royalmp2001@gmail.com> wrote:
Hi everyone.
I want to convert my square wave signal generator to produce only a
positive going waveform without just chopping off the -ve signal.

The signal generator has variable output from 0-10V (20Vpeaktopeak or
-10V to +10V)).

I came up with a simple circuit that nearly works:
The hot output of the signal generator goes to the cathode of a diode,
the anode connects to the -ve of a small value electrolytic capacitor
and the +ve goes to the ground output. My resulting signal is taken
either side of the diode.

Works well, but I get the signal going a fraction of a volt negative.
If I use a shottky diode the negative offset is reduced but still
there.

It is important that the output does not go negative at all - in fact
I need it about a quarter of a volt positive offset. I have power
supply rails that I can access from the signal generator, but
connecting a "pull-up" resistor does not work - I think because of the
cap.

Any ideas - keeping it simple?
I know I could achieve this with an op-amp but want to avoid that and
try and keep it relatively passive. If I have to I could use one of
two transistors.

Is what I ask possible? Thanks
Click to expand...
a resistive divider?

try this:

.-------------------------------------------------------------.
| This is an ascii schematic, if the diagram appears garbled |
| try switching to a fixed-pitch font (courier works well) |
| pasting it into notepad works well on ms-windows. |
| or in google groups "show original" (in "more options") |
`-------------------------------------------------------------'

+ ---[1000]-----.
10V |
supply |
- ---. |
| |
gnd |
|
-------[1100]---+----
sig
gen out
--. +----
| |
gnd gnd
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top