Limiting Amplitude of 555 Oscillator Output The Cheap N Easy Way

[/dev/phaeton]

Hello!

I have constructed a 555 based oscillator. The schematic is here:

http://www.uoguelph.ca/~antoon/circ/555test.gif

but as you can see it is a 'standard practice' type of circuit. I made a
few modifications to alter and adjust the frequency of oscillation
(replaced a resistor with a pot and shrunk the size of the cap). I will
be using this to inject a signal into an audio amplifier for diagnostics.

Nonetheless, when powered by the convenient 9V battery, the amplitude of
the output is several volts (as expected). I would like to
1) limit this to under 1V maximum, and
2) block any DC from entering the amplifier's input.

Is it easy enough to simply put a capacitor inline with the "Optional
Pulse Output" in the pic to block the DC, and then use two anti-parallel
diodes connected from there to ground to 'cut down' the
signal? AFAICT the diodes will safely reduce the signal down to ~0.6V for
your garden variety silicon signal diodes (1n914, 1n4148, etc) and ~0.3V
for the oldschool GE types. Sound plausible? Is there a better way to go
about this?

Thanks for any feedback,

-phaeton

 

[John Popelish]

Hello!

I have constructed a 555 based oscillator. The schematic is here:

http://www.uoguelph.ca/~antoon/circ/555test.gif (snip)
Nonetheless, when powered by the convenient 9V battery, the amplitude of
the output is several volts (as expected). I would like to
1) limit this to under 1V maximum, and
2) block any DC from entering the amplifier's input.

Is it easy enough to simply put a capacitor inline with the "Optional
Pulse Output" in the pic to block the DC, and then use two anti-parallel
diodes connected from there to ground to 'cut down' the
signal? AFAICT the diodes will safely reduce the signal down to ~0.6V for
your garden variety silicon signal diodes (1n914, 1n4148, etc) and ~0.3V
for the oldschool GE types. Sound plausible? Is there a better way to go
about this?

The two diode signal clipper is fine, if you add a series
resistor (that can be upstream of the capacitor or
downstream). Without this resistor, there is a large
current spike on each edge that will heat the 555 and bounce
the battery voltage around.

You can also connect the ends of a pot between the DC
blocking capacitor and ground and take the signal from the
wiper, so you can adjust the amplitude anywhere between zero
and full output. A 10k pot should work fine for most signal
injection applications. If you want to protect the wiper
contact and 555 from a short to ground at full amplitude,
you can either add a series resistor to the capacitor end of
the pot or in series with output from the the wiper.

 

[/dev/phaeton]

On Sun, 14 Oct 2007 20:35:19 -0400, John Popelish wrote:

Thanks for the reply, John.

What value resistor would be a good starting point?

I thought about the potentiometer to add a 'volume' control, but without
really knowing the output amplitude (my DVOM doesn't respond fast enough
to read the voltage) I figured I'd just fix it at 0.6V so that I could
always know it was 'safe'.

Most amplifiers that I work with (guitar, bass, etc) expect something on
the order of a 100mV signal directly from the instrument. They are often
used with a boost or distortion effect that (on average) has a gain of
about 100, so a 1V signal should be ok. I don't know if the 7-8V signal
directly from the 555 would damage anything or not, so I decided to take
the safe route and cut it way down.

-phaeton

 

[John Popelish]

On Sun, 14 Oct 2007 20:35:19 -0400, John Popelish wrote:

Thanks for the reply, John.

What value resistor would be a good starting point?

The 555 has a pretty high peak current rating, but to keep
the battery voltage from dipping each time a positive swing
takes place, I think limiting the current to a few milliamps
is good enough. A 1k would work.

I thought about the potentiometer to add a 'volume' control, but without
really knowing the output amplitude (my DVOM doesn't respond fast enough
to read the voltage) I figured I'd just fix it at 0.6V so that I could
always know it was 'safe'.

Most amplifiers that I work with (guitar, bass, etc) expect something on
the order of a 100mV signal directly from the instrument. They are often
used with a boost or distortion effect that (on average) has a gain of
about 100, so a 1V signal should be ok. I don't know if the 7-8V signal
directly from the 555 would damage anything or not, so I decided to take
the safe route and cut it way down.

Sounds like a plan. Small Schottky signal diodes also make
good low voltage clamps similar to germanium diodes, if you
don't have those to spare.

 

[/dev/phaeton]

Sounds like a plan. Small Schottky signal diodes also make
good low voltage clamps similar to germanium diodes, if you
don't have those to spare.

Excellent! Thanks for the tips. I have a bunch of Ge diodes, but they're
a little costlier than some others. That, and i don't want to use
them frivolously since someday they will all be gone. :-(

So this technique with the diodes is called 'clamping'?

-phaeton

 

[John Popelish]

Excellent! Thanks for the tips. I have a bunch of Ge diodes, but they're
a little costlier than some others. That, and i don't want to use
them frivolously since someday they will all be gone. :-(

So this technique with the diodes is called 'clamping'?

Clipping or clamping, depending on who you ask.

If you can handle a little surface mount device, these have
two diodes in one package for about $0.13 each from Digikey:
http://www.fairchildsemi.com/ds/BA/BAT54.pdf

 

[Jamie]

Excellent! Thanks for the tips. I have a bunch of Ge diodes, but they're
a little costlier than some others. That, and i don't want to use
them frivolously since someday they will all be gone. :-(

So this technique with the diodes is called 'clamping'?

-phaeton

You may want to consider using a voltage divider through a
padded RC network other wise, compressing the signal is going
to place a load on the circuit which will drain a battery faster
than expected. I guess you could pass a series resistor from the
output of high value before the clamps. In either case how ever,
you'll still get square wave as a results. I think you did say
something about injecting it into an amp? If that being the case,
you'll want to use a simple RC network that can shape the wave
form and reduce level drive (attenuate).

 

[Rich Grise]

Clipping or clamping, depending on who you ask.

Just to be devil's advocate, the way I learned it, they're two
different things. A clipper limits the output excursion - a
clamp changes the DC bias such that one signal excursion never
exceeds a certain voltage, more like a level shifter:

Input signal, 4Vpp
__ __ __
+2V | | | | | |
| | | | | |
0V | | | | | | etc.
| | | | | |
-2V _| |__| |__| |__


Clipped to 2Vpp:
__ __ __
+1V | | | | | |
0V | | | | | | etc.
-1V _| |__| |__| |__


Clamped at 0V:
__ __ __
+4V | | | | | |
| | | | | |
| | | | | | etc.
| | | | | |
0V _| |__| |__| |__

The clamp voltage can be anything you want, and can be on the
top or the bottom of the waveform. They do this to TV video,
(clamp the sync pulses to some reference) but call it a
"DC restorer".

Cheers!
Rich

 

[John Popelish]

Just to be devil's advocate, the way I learned it, they're two
different things. A clipper limits the output excursion - a
clamp changes the DC bias such that one signal excursion never
exceeds a certain voltage, more like a level shifter:

Input signal, 4Vpp
__ __ __
+2V | | | | | |
| | | | | |
0V | | | | | | etc.
| | | | | |
-2V _| |__| |__| |__


Clipped to 2Vpp:
__ __ __
+1V | | | | | |
0V | | | | | | etc.
-1V _| |__| |__| |__


Clamped at 0V:
__ __ __
+4V | | | | | |
| | | | | |
| | | | | | etc.
| | | | | |
0V _| |__| |__| |__

The clamp voltage can be anything you want, and can be on the
top or the bottom of the waveform. They do this to TV video,
(clamp the sync pulses to some reference) but call it a
"DC restorer".

I agree that clipper or limiter are the better choices for
reducing the peak-to-peak swing.