# How can I design an lm555 timer 50% duty cycle for 10 Khz?

#### asd asd 1

Joined Mar 27, 2017
3
Hi I'm trying to design astable circuit with %50 duty cycle for 10 Khz ( tlow=50us and thigh=50us) by using lm555 timer. There are lots of configurations for %50 duty cycle timer circuits but none of them is not working properly for me. Can anybody help me?
Thanks a lot.

#### EM Fields

Joined Jun 8, 2016
583
Hi I'm trying to design astable circuit with %50 duty cycle for 10 Khz ( tlow=50us and thigh=50us) by using lm555 timer. There are lots of configurations for %50 duty cycle timer circuits but none of them is not working properly for me. Can anybody help me?
Thanks a lot.
If you need precisely 50%, run the astable at 20kHz and feed its output into a divide-by-two circuit, like this:

Last edited:

#### dl324

Joined Mar 30, 2015
12,911
Welcome to AAC!

The easiest way is to double the timer frequency and use a flip flop.

#### Alec_t

Joined Sep 17, 2013
12,148
The post #4 circuit should give a ~50% duty cycle if the 555 is a CMOS type but, for the more common bipolar type, the fact that pin 3 can only get within a volt or so of the +ve rail makes the duty cycle significantly different from 50%. The deviation is supply-voltage dependant.

#### ian field

Joined Oct 27, 2012
6,539
Hi I'm trying to design astable circuit with %50 duty cycle for 10 Khz ( tlow=50us and thigh=50us) by using lm555 timer. There are lots of configurations for %50 duty cycle timer circuits but none of them is not working properly for me. Can anybody help me?
Thanks a lot.
There is a small duty cycle error because the timing capacitor charges through 2 resistors but only discharges through 1 of them - the discharge transistor must have a current limiting resistor so it doesn't short out the supply during the discharge period, The easiest way to get 50/50 is to drive the CR chain with the output and just ignore the discharge pin.

You can also have separate charge and discharge resistors and a steering diode in series with each - say for example; you attach a preset pot wiper to the top of the timing cap, you can connect the charge and discharge chains to each end of the pot and trim the mark space ratio to exactly what you want.

#### crutschow

Joined Mar 14, 2008
27,389
Depends upon how precisely you need it to be 50%.

#### ian field

Joined Oct 27, 2012
6,539
Depends upon how precisely you need it to be 50%.
Doubling the frequency and dividing by 2 is hard to beat for symetry, but 20kHz is somewhere near where a bipolar starts to get flaky from the frequency drift point of view.

Using the output to drive the CR network is pretty good - its mainly any difference in the output rise and fall times that could skew the symmetry very slightly.

#### asd asd 1

Joined Mar 27, 2017
3
If you need precisely 50%, run the astable at 20kHz and feed its output into a divide-by-two circuit, like this:
View attachment 123474

What about the control voltage? Why is it left as not connected?

#### absf

Joined Dec 29, 2010
1,949
Just connect a 10nF capacitor from pin 5 to GND.

Allen

#### absf

Joined Dec 29, 2010
1,949

Get the R and C from 555 calculator

Output Freq from 555 is about right

Yellow trace if the o/p of 555 and blue trace is the o/p of 74HC74

Allen

#### EM Fields

Joined Jun 8, 2016
583
What about the control voltage? Why is it left as not connected?
Unless you're in a really noisy environment and jitter on the output is a problem, letting it float is OK.
If you want / need to decouple it, though, 10nF or so between pin 5 and GND will work.

#### GopherT

Joined Nov 23, 2012
8,012
@asd asd 1
The smallest recommended resistor from positive supply to pin7 is 1k.
With that, a standard astable 555 can be built with a 50.3% duty cycle. So, with rounding errors, it depends on your definition of "50%".

Resistor from pin7 to pin6 = 67k ohm
Cap from pin6 to ground = 1 nF

Replace the 67k with a 47k in series with a 50k multi-turn trimmer to dial in 10k HZ exactly.

#### hp1729

Joined Nov 23, 2015
2,304
The post #4 circuit should give a ~50% duty cycle if the 555 is a CMOS type but, for the more common bipolar type, the fact that pin 3 can only get within a volt or so of the +ve rail makes the duty cycle significantly different from 50%. The deviation is supply-voltage dependant.
Is that output voltage relevant since the threshold and trigger levels are set at 2/3 and 1/3 of VCC?

#### Alec_t

Joined Sep 17, 2013
12,148
Yes, it is relevant when pin 3 provides the current path for both charging and discharging the timing capacitor (unlike in the usual textbook circuit where the positive rail is the charging source and pin 7 is the discharge sink).

#### crutschow

Joined Mar 14, 2008
27,389
Here's a 555 circuit that can be adjusted to 50% duty-cycle:

#### ian field

Joined Oct 27, 2012
6,539
Just connect a 10nF capacitor from pin 5 to GND.

Allen
That's what most application notes recommend - a lot of people just don't bother.

#### EM Fields

Joined Jun 8, 2016
583
That's what most application notes recommend - a lot of people just don't bother.
And with good reason, since the jitter caused by letting the pin float is almost always negligible and the addition of a decoupling cap would be wasteful and might change the timing formula.

Last edited:

#### asd asd 1

Joined Mar 27, 2017
3
Unless you're in a really noisy environment and jitter on the output is a problem, letting it float is OK.
If you want / need to decouple it, though, 10nF or so between pin 5 and GND will work.
Thank you so much

#### dl324

Joined Mar 30, 2015
12,911
Doubling the frequency and dividing by 2 is hard to beat for symetry, but 20kHz is somewhere near where a bipolar starts to get flaky from the frequency drift point of view.
This is from the Signetics Analog Applications manual for NE/SE555:

The Nat Semi datasheet implies that 100kHz is typical: