50 Per Cent Duty Cycle From The TLC555?
Ron Hubbard
possible to get a real 50 per cent duty cycle using just one
resistor and one cap, but it doesn't bother to actually say
how.
Does anyone know how to get a real square wave out of the
TLC555?
Thanks
Michael Black
ground. Then connect your timing resistor from that junction to
pin 3, the output.
Since the output is symmetrical, the capacitor will charge and
discharge at a symmetrical rate, giving you the 50 percent duty
cycle.
These scheme is almost thirty years old; I saw it in "Electronics"
magazine when someone sent in little bit about it.
Michael
Uwe Zimmermann
Is it this datasheet?
http://www-s.ti.com/sc/ds/tlc555.pdf
If it is, just a hint:
p.10 output waveform duty cycle = Rb/(Ra+2Rb)
with Ra=0 this gives duty cycle = 0.5
Uwe.
Uwe Zimmermann
off-datasheet:
http://www.e-online.de/public/schaerer/bilder/ast555_2.gif
http://www.westminster.org.uk/intranet/departments/electronics/555.htm
http://www.intersil.com/
Their ICM7555-datasheet contains a bit more information than TI's...
Uwe.
Ron Hubbard
sheet that comes in the Radio Shack package is just a small
sheet that like so many things from Radio Shack, is more
useless than helpful. :-)
Uwe Zimmermann wrote in message
<3DCC2924...@imit.kth.se>...
Phil Allison
"Michael Black" <et...@FreeNet.Carleton.CA> wrote in message
news:aqh82b$aae$1...@freenet9.carleton.ca...
> "Ron Hubbard" (nolo...@hotmail.com) writes:
> > The data sheet that comes with Radio Shack's TLC555 says it's
> > possible to get a real 50 per cent duty cycle using just one
> > resistor and one cap, but it doesn't bother to actually say
> > how.
> >
> > Does anyone know how to get a real square wave out of the
> > TLC555?
> >
> > Thanks
> >
> >
> Connect pins 6 and 2 together, putting your capacitor from there to
> ground. Then connect your timing resistor from that junction to
> pin 3, the output.
>
> Since the output is symmetrical, the capacitor will charge and
> discharge at a symmetrical rate, giving you the 50 percent duty
> cycle.
** Unfortunately that arrangement is not stable in frequency with
changes in supply voltage.
................... Phil
John Larkin
wrote:
Why not? Everything is ratiometric on Vcc.
John
Phil Allison
"John Larkin" <jjla...@highSNIPlandTHIStechPLEASEnology.etc> wrote in
message news:0ssosuggm4fiffqh6...@4ax.com...
** Is it ?? My experience is with the NE555, maybe the TLC version
is better.
The output swing as a percentage of the supply varies with supply
voltage, so the time to charge and discharge the cap between 1/3 and 2/3 of
the supply varies. It gets longer at low supply voltages.
.......... Phil
Uwe Zimmermann
>
> Thanks, Uwe, but don't forget that paper still exists. The
> sheet that comes in the Radio Shack package is just a small
> sheet that like so many things from Radio Shack, is more
> useless than helpful. :-)
Since we don't have RadioShack here in Europe, I'm not aware of what
they package as "datasheet" and I assumed it might just have been a
copy of the datasheet...
Uwe.
John Larkin
wrote:
On the CMOS versions (like TLC555) the output will swing r-r, so the
whole thing is ratiometric. But you're right, if there's junction
drops in the pullup direction, like in the classic bipolar '555,
frequency will change some with supply voltage.
John
Phil Allison
"John Larkin" <jjla...@highSNIPlandTHIStechPLEASEnology.etc> wrote in
Phil wrote:
> >
> > ** Is it ?? My experience is with the NE555, maybe the TLC
version
> >is better.
> >
> > The output swing as a percentage of the supply varies with supply
> >voltage, so the time to charge and discharge the cap between 1/3 and 2/3
of
> >the supply varies. It gets longer at low supply voltages.
> >
> >
> > .......... Phil
> >
> >
>
>
> On the CMOS versions (like TLC555) the output will swing r-r, so the
> whole thing is ratiometric. But you're right, if there's junction
> drops in the pullup direction, like in the classic bipolar '555,
> frequency will change some with supply voltage.
** The TLC555 is not rail to rail under all conditions. At supply
volts under 5 volts there is a significant loss of voltage when driving even
a few milliamps of load. I took a peek at the data sheet.
................ Phil
John Larkin
wrote:
>
Geez, Phil, it's just a 555. If you want a precision square wave, use
a crystal oscillator and a flipflop. Or a caesium standard.
John
Phil Allison
"John Larkin" <jjla...@highSNIPlandTHIStechPLEASEnology.etc> wrote in
** Hang on a minute - you challenged my statement John, seems on flimsy
grounds, so now you want to make smart remarks at my expense ??
.................. Phil
dB
site
www.davidbridgen.com
If you use a semi-decent supply you shouldn't need to worry about it
causing timing variations.
Phil Allison
"dB" <dmb0...@yahoo.com> wrote in message
news:1757808.02110...@posting.google.com...
** I notice you did not give a formula for the frequency.
............. Phil
John Larkin
wrote:
Oh, no expense. The smart remarks are all free.
** John
Phil Allison
"John Larkin" <jjla...@highSNIPlandTHIStechPLEASEnology.etc> wrote in
>
> Oh, no expense. The smart remarks are all free.
>
> ** John
>
** Free of liability I see too.
............ Phil
Rob Paisley
http://home.cogeco.ca/~rpaisley4/555.html
Circuits # 14 and 23
Rob.
"Ron Hubbard" <nolo...@hotmail.com> wrote in message news:<aqh69i$a22rb$1...@ID-43450.news.dfncis.de>...
onestone
since Ra = 0
fo = 1.44/2RbC
Al
Phil Allison
"onestone" <ones...@chariot.net.au> wrote in message
news:3DD19251...@chariot.net.au...
> fo = 1.44/((Ra+2Rb)C)
>
> since Ra = 0
>
> fo = 1.44/2RbC
** Where is the supply voltage factor ?? Or is any formula better than
none ?
............... Phil
onestone
Now now, sarcasm doesn't become you Phil. This is the formula quoted by
both Ti and National, who both state that the frequency is independant
of supply voltage. However, the TLC555 data sheet does go into slightly
more depth to suggest that propagation delays from TRIG and THRESH
inputs to DISCH may cause slight variation between calculated and actual
period, and that the internal on state resistance (ron) during discharge
adds to Rb to provide another source of timing error when Rb is very low
or ron is very high.
Of course if you wanted that much precision you'd not be using a 555 in
the first place.
Al
Michael Black
since it's in Walter Jung's "IC Timer Cookbook". (The original circuit
was apparently in Electronics for June 21, 1973, which would be when
I remember it from: "IC Timer Plus Resistor Can Produce Square Waves"
by S. A. Orrel.)
From Jung: "Ideally the output voltage would be V+ in the high state;
however, the high-state saturation limit of the 555 is nearly 2 volts
below V+. This causes some time assymmetry in the waveform, especially
at low supply voltages near 5volts. The timing period would be simply
2 X 0.693RC (or 1.386RC) were it not for this error. In light of this
error, the time equation given is approximately 1.4RC and this will
vary with differing supply-voltage levels. However, even in view of
these limitations, the circuit is still quite useful due to its inherent
simplicity and low cost".
Which of course is why I've used the circuit all these decades when
I needed a simple signal source. No sense fussing with the extra
resistor of the "normal" 555 configuration, or worry about the waveform
getting too narrow over a wide frequency range. I wouldn't use it
for precision use, but then like someone already posted, you'd be
using something other than the 555 for precision use.
Of course, the trick for fixing the problem is one I'd even forgotten
because the problem wasn't my concern. The next section of the book
mentions using the same configuration and adding a 1K resistor between the
output (pin 3) and the V+ line. "Here R1 is an added pull-up resistor,
which forces the pin 3 voltage to rise to V+ in the high state. This removes
the time assymmetry error, and the waveform becomes square in shape.
The time (and frequency) expressions then become more precise, as
shown. This assumes that the loading at pin 3 for the high state is
low." T=1.386RC
Michael
Phil Allison
> Phil Allison wrote:
> >
> > "onestone" <ones...@chariot.net.au> wrote in message
> > news:3DD19251...@chariot.net.au...
> > > fo = 1.44/((Ra+2Rb)C)
> > >
> > > since Ra = 0
> > >
> > > fo = 1.44/2RbC
> >
> > ** Where is the supply voltage factor ?? Or is any formula better
than
> > none ?
> >
> > ............... Phil
>
> Now now, sarcasm doesn't become you Phil.
** Quite the opposite - I am very good at it.
This is the formula quoted by
> both Ti and National, who both state that the frequency is independant
> of supply voltage.
** They post no such formula and make no such claim for using pin 3 in
an astable.
However, the TLC555 data sheet does go into slightly
> more depth to suggest that propagation delays from TRIG and THRESH
> inputs to DISCH may cause slight variation between calculated and actual
> period, and that the internal on state resistance (ron) during discharge
> adds to Rb to provide another source of timing error when Rb is very low
> or ron is very high.
** You are OT. The cct is the one that involves the use of pin 3.
>
> Of course if you wanted that much precision you'd not be using a 555 in
> the first place.
** The LM555 is speced as having a 0.15 % per volt frequency drift with
supply - used with a standard astable cct. Also a 90 ppm per degree C drift
with temp - which is better than the passives likely to be used to set the
frequency.
Maybe you would care to check Walter Jung's IC Cookbooks for more on
the topic.
................ Phil