My 555 timer in astable mode is generating about 0,1Hz square wave.
Following is the circuit that I am using.

The problem is that it creates a longer pulse when just turned on. I understand that the first pulse in a NE555 astable circuit is longer because the timing capacitor starts at 0V and must charge up to 2/3 Vcc.

It is critical because the circuit is connected to other circuits to control the motors of a crane at my railway model club. So the angle of rotation is not the same.

I have tried also with a TMC 555 CMOS version but it is the same problem.

Any idea ?

Thank you !

 

Try "precharging" C1 just below approx. 1/3 VCC, point "B", like this.

 

hi @Philexium
A little 'messy' but it works.
E

 

Another approach is to let the oscillator free run and gate it's output. Depending on the specifics of the problem, this could be very simple or require a bit of effort.

 

eetech00's post #2 works fine but may need to increase R4 to 1.5K to compensate for the diode drop with Vcc at 5 volts.

 

A simple mod is to connect the timing capacitor to V+ instead of ground.
There will be a delay on startup (if that is tolerable), but the first pulse will be the same width as the rest (LTspice sim below).

Edit: This only works if the circuit is started by applying power.
It won't work if the circuit is controlled by the Reset input.

 

Another solution is to use a different circuit design.
You can run the 555-timer at a much higher frequency and then feed that into a divide-by-n counter.

 

Is the pulse duty-cycle (on vs. off time) or just the on-time important?

 

Try "precharging" C1 just below approx. 1/3 VCC, point "B", like this.

View attachment 365169

Hi @eetech00
Thank you for this clever idea;
I have slightly reduced the length of the first longer pulse

 

hi @Philexium
A little 'messy' but it works.
E
View attachment 365172

Hi Eric !
Two many components I don't have a lot of room
Best regards

 

Hi @eetech00
Thank you for this clever idea;
I have slightly reduced the length of the first longer pulse

Try changing R4 to 1.5K to reduce the length more.

 

Another technique that needs more room and parts is to increase the freq of the 555 and run the output through a counter. For example, if you increase the 555 freq by 128x and run the output through a 8-bit counter, the extra width of the first half-cycle will be reduced by 99%.

You can combine the oscillator and divider into a single part such as the CD4060, but you have to add some extra gating to get the output you want.

EXAR used to make parts that were a 555 plus a divider in a single package, the XR2240 / 2242 / 2243. I think they are long gone by now. Too bad; the XR2242 and 2243 were great.

ak

 

Another technique that needs more room and parts is to increase the freq of the 555 and run the output through a counter. For example, if you increase the 555 freq by 128x and run the output through a 8-bit counter, the extra width of the first half-cycle will be reduced by 99%.

That works as long as the delay for the first pulse output is not a problem.

One solution for that, would be to add a XOR gate with the inputs being the 555 enable signal, and the counter output.
That way the output will go high with the enable signal, and then go low when the counter goes high at the end of its count.
LTspice of example circuit below using a 555 and counter with the XOR gate: