Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Making a timer on delay and off delay without microcontroller

fadlianturu6

New Member
I am new and want to make a compact circuit without a microcontroller. The idea is when you push the button/switch, it will delay the timer for at least 6 hours. After that, it will activate the motor for 20 seconds and loop forever. I am thinking of using a 555 timer with a 1000uF capacitor here, but it fails miserably. Any ideas?
Screenshot_227.png
 
So, I've made some progress on my project. Since 6 hours is too long, I divided it into 5 segments, resulting in a 72-minute delay with a 4-second activation time. I also removed the 555 timer and only used the CD4060BC because it's the more simple and why i BC because it is the only available.

I set the frequency to 0.25 Hz because I need the output from Q4.
1708612468778.png

There is no Q1 and Q2
this is my calculation

1708612648716.png


so my circuit look like this
1708615391755.png

(this is the real one, but i make the shorter to see anything wrong)
1708613776680.png

However, I encountered an issue with the resistor calculation. I expected a 2-second delay from Q3 (18 kOhm), but I got 2.3 seconds. After some trial and error in the software, I found that using a 16 kOhm resistor gave me a 2.05-second delay.

I'm wondering if this discrepancy will happen in real life or if it's just an issue with the Proteus simulation.
1708615034751.png
or
Blue one is 18k and yellow is 16k

Additionally, I'm facing a problem where the oscillator doesn't loop when I split the output to some load.
1708615540462.png

Got stuck like that, idk why is it grey
 

Attachments

  • 1708613760046.png
    1708613760046.png
    27.6 KB · Views: 40
What clock / timing accuracy do you want ?

Connecting an LED directly to a logic output not a good idea. Your logic
level into and gate most likely never meeting a "1", see spec. Use a R in
series with LED to limit its current and allow the min logic "1" to be achieved.
Freq accuracy for this device is HORRIBLE due to device and C changes,
see timing variation :

1708621123912.png


Regards, Dana.
 
Last edited:
In a real circuit, you'll have to tweak the RC values. 4060 timing cap should be non-polar type, so if a large cap, use two polar caps connected in series. Using RC timing components will be a compromise on accuracy.

Edit: Deleted Circuit (not good)
 
Last edited:
Beware of tying large caps to CMOS inputs :


Page 7 -



Regards, Dana.


Good info links for many.

Fortunately the R-C-R "T" has equivalent current limiting to the proposed oscillator C-R-R connections so the ESD diodes can prevent shoot-thru from transient power ON-OFF-ON. I believe the CMOS inputs use two stages of 50kohms to diodes so 500 pF is suggested ma.x when terminated to AC equiv. short ( gnd or voltage source.) But that is not done here.
 
First pass concept schematic for 4 seconds on, 72 minutes off, one chip, and no 555.

The ratio of 4 seconds to 72 minutes is 1080:1, while the maximum output period ratio for a CD4060 is between Q4 and Q14, which is only 1024:1. Thus, the outputs of one 4060 can be decoded to both intervals only if the output ratio spec is reduced slightly. One option is 4 seconds on and 68 minutes 12 seconds off.

Even with that, the output gating is a bit messy. Adding a simple pulse-former to the 4060 Q14 output separates the output pulse width from the overall period.

In this schematic, all component values are approximate.

C1-R1 set the relay on-time.

C2-R2 set the overall cycle time.

C4-R4 perform a power-on reset.

D2-R5 are an optional heartbeat indicator to show that the timer is running.

The circuit pulses the output once at power-on, then enters the long off period.

ak

4060-Cycle-Timer-1-c.gif
 
Last edited:
I have some little time delay relays with long delays.
They do not use a 555 for long delays because the oscillator is not stable at very low frequencies.
They use CD4541 or CD4521 IC which has an oscillator and 24 stage counter.

Circuit copied from the internet. I do not know about it. 3-switches to set how long the counter is and a pot to fine adjustments.
1708627904698.png
 
The 4541 is great for extra-long delay periods, but it has no advantage in this application other than allowing for a timing capacitor that is 4x smaller than one for a 4060.

The 4521 can use an even smaller timing cap, but still does not have enough counter stages brought out to decode a 4 second pulse in a 72 (or 68) minute period.

ak
 
Good info links for many.

Fortunately the R-C-R "T" has equivalent current limiting to the proposed oscillator C-R-R connections so the ESD diodes can prevent shoot-thru from transient power ON-OFF-ON. I believe the CMOS inputs use two stages of 50kohms to diodes so 500 pF is suggested ma.x when terminated to AC equiv. short ( gnd or voltage source.) But that is not done here.


Where do you find 50K, two stages ? Shoot thru not only issue with ESD diodes. Depending on way
supply is removed determines C discharge current, which with large C's can cause localized hot
spots in die, blow out bond wires, damage the ESD diode.....The Toshiba ap note covers basics, this
TI ap note more info :



Regards, Dana.
 
Where do you find 50K, two stages ? Shoot thru not only issue with ESD diodes. Depending on way
supply is removed determines C discharge current, which with large C's can cause localized hot
spots in die, blow out bond wires, damage the ESD diode.....The Toshiba ap note covers basics, this
TI ap note more info :



Regards, Dana.
That's my memory when I was a Test Engineering Mgr in the early 80's implementing a plant-wide ESD/EOS implementation and awareness.

Let me see if I can simulate it for you as it all about R ratio attenuation ratios with diode Rs from ESD Vpk for ultra fast Sch diodes. Rs is inversely related to power or max. input diode current which is provided in all datasheets the 5 or 10 mA depending on family.

The current limit in datasheets in order to attenuate stored Q at max voltage and attenuate to 0.2V as the Vf = 0.2V is threshold limit outside rail using DC.

I might have to choose diodes with these current limits to understand. They MAY or MAY NOT have improved designs now to obtain single stage input protection but that's how it started in CMOS. BUT I DO NOT SEE HOW IT IS POSSIBLE.

I know about hot spots are from low impedance C loads and unequal current sharing.
 
Last edited:
Where do you find 50K, two stages ? Shoot thru not only issue with ESD diodes. Depending on way
supply is removed determines C discharge current, which with large C's can cause localized hot
spots in die, blow out bond wires, damage the ESD diode.....The Toshiba ap note covers basics, this
TI ap note more info :

Note scales on each plot.
1708640698219.png


https://tinyurl.com/232r4mmw

Plot A is what I recall was used. As Logic Family increases in speed, ESD diodes must be faster for the Rs C = Tau parameters have improve but Rs is still high and current limit was only improved from 5 to 10mA for most CMOS at DC outside rails.
The conceptual internal CMOS ESD diode design for 0.2V max. based on recommended max Q of 500 pF @ 5V
 
First pass concept schematic for 4 seconds on, 72 minutes off, one chip, and no 555.

The ratio of 4 seconds to 72 minutes is 1080:1, while the maximum output period ratio for a CD4060 is between Q4 and Q14, which is only 1024:1. Thus, the outputs of one 4060 can be decoded to both intervals only if the output ratio spec is reduced slightly. One option is 4 seconds on and 68 minutes 12 seconds off.

Even with that, the output gating is a bit messy. Adding a simple pulse-former to the 4060 Q14 output separates the output pulse width from the overall period.

In this schematic, all component values are approximate.

C1-R1 set the relay on-time.

C2-R2 set the overall cycle time.

C4-R4 perform a power-on reset.

D2-R5 are an optional heartbeat indicator to show that the timer is running.

The circuit pulses the output once at power-on, then enters the long off period.

ak

View attachment 144619
How do you generate a pulse using a timer? I tried connecting all the VCC (positive supply voltage) terminals to a single battery, but the outcome was unsuccessful. The current didn't increase or decrease; it remained constant. Additionally, when I attempted to connect a motor, LED, or relay, there was no response. How can I calculate the pulse using an R-C (resistor-capacitor) circuit?
is it
tau = R*C
R = Tau/C
R = 10^-5/4
R = 400k Ohm
if that right then my calculation should work
Or is it error because different type of mosfet ?

1708659007589.png

1708660485798.png

tried it with a 555 timer, but the outcome is almost the same; it didn't even reset the counting.
1708659500416.png
 
Last edited:
In the schematics in post #34, the Q1 drain and source are reversed. Also, R13 is very large for an LED current-limiting resistor.

Beyond that, I don't understand things. Are these supposed to be my schematic re-drawn into a simulator? If so, there are several errors. Why is the Q4 output connected directly to the 4060 Reset input? Why is it being used at all? Why is there a 455K resistor in the oscillator?

In your calculation you indicate that the pulse width you are trying to calculate the components for is 10^-5, or 10 microseconds. I thought you wanted an output pulse of 4 seconds.

ak
 
How do you generate a pulse using a timer? I tried connecting all the VCC (positive supply voltage) terminals to a single battery, but the outcome was unsuccessful. The current didn't increase or decrease; it remained constant. Additionally, when I attempted to connect a motor, LED, or relay, there was no response. How can I calculate the pulse using an R-C (resistor-capacitor) circuit?
is it
tau = R*C
R = Tau/C
R = 10^-5/4
R = 400k Ohm
if that right then my calculation should work
Or is it error because different type of mosfet ?

View attachment 144631
View attachment 144633
tried it with a 555 timer, but the outcome is almost the same; it didn't even reset the counting.
View attachment 144632

If you use RC timing components at the gate of the mosfet like that, then the timing is heavily dependent on the mosfet vgs(th), so timeout can vary depending on the mosfet batch/manufacturer.
 
In your calculation you indicate that the pulse width you are trying to calculate the components for is 10^-5, or 10 microseconds. I thought you wanted an output pulse of 4 seconds.
sorry wrong number
R = tau/C
R = 4/(10*10^-6) (that why i write 10^-5)
R = 400k ohm
and this is your design in proteus
1708668851738.png

And this is your design in Proteus. I have a question about how the reset will function in a setup like this.
Why is the Q4 output connected directly to the 4060 Reset input? Why is it being used at all? Why is there a 455K resistor in the oscillator?
I connected it because I thought the reset should occur when the output from Q14 is on.

So, in that circuit, I want to check the output from (R6, C5, and Q1) in your design (C1, R1, and Q1), but it still doesn't activate for 4 seconds.
 

Latest threads

New Articles From Microcontroller Tips

Back
Top