A one time timer

[iamashwin99]

I need to switch on a motor for 5 minutes once it receives a 5v input and stop after 5 minutes. And this should cycle again once restart it.

I want to do this without a micro controller. I think it should be possible with a 555 but im a newbie here and i dont know how to use it to get this output.

Can some one suggest me a way.

[larsdenmark]

This should help:
https://electronics.stackexchange.com/questions/180716/555-timer-one-shot-trigger

[Benta]

Use a CD4538 or MC14538. A much better choice for this.

[Brumby]

You could do this with a 555.  However 5 minutes is a bit of a long period and you may get some variation on that time, so it depends on how precise that 5 minutes needs to be.

[iamashwin99]

I dont need it to be precise, anything from 5-8min is acceptable.
But even then how do accomplish it with a 555.
If the question isnt clear, the input and the output waveforms are as in the attachment, the input can be from a button press(which can last a few sec) and the output has to be high for 5-8 minutes.

[Wimberleytech]

https://www.allaboutcircuits.com/tools/555-timer-monostable-circuit/

[iamashwin99]

https://www.allaboutcircuits.com/tools/555-timer-monostable-circuit/

here the output is set high when a falling edge is detected, is it possible to set it high when the input rises instead.

[mikerj]

https://www.allaboutcircuits.com/tools/555-timer-monostable-circuit/

here the output is set high when a falling edge is detected, is it possible to set it high when the input rises instead.

Add a single transistor stage to invert the edge.

[iamashwin99]

https://www.allaboutcircuits.com/tools/555-timer-monostable-circuit/

here the output is set high when a falling edge is detected, is it possible to set it high when the input rises instead.

Add a single transistor stage to invert the edge.

Ok thanks ill try that out.

[Benta]

Read my reply #2 again. A 4538 will give you inverted input, non-inverted input, inverted output, non-inverted output, retriggering, non-retriggering and is probably cheaper. Why are you f*cking around with a 555?

Sorry, I had to get that out of my system, but it's a mystery to me why all designs are centered on a 555.

[Zero999]

Why use a 555?

Real engineers use a discrete monostable. If the trigger pulse is always shorter than the delay, then R4 and C2 are not needed: R4 should be open and C2 shorted.



EDIT:
I omitted R5, which is required to limit the base current through Q1, when C1 charges. Without it, nearly the full load current will pass through the base, potentially destroying it!

[alsetalokin4017]

Real Engineers may use a discrete monostable, but many ordinary people just use what they have available in their junqueboxes.

[Brumby]

I have no qualms about using any solution when it is capable of doing the job.  If a 555 can do it - then why not?

Besides ... the 555 has a very well deserved place in history and getting to implement one in a circuit can certainly be considered a right of passage.

[james_s]

The 555 is a great part, but that doesn't mean it's the right tool for every job. When talking of delays of several minutes or more, there are better choices.

[Brumby]

There may be "better" - but better how?  Are they going to be better for the criteria:

I dont need it to be precise, anything from 5-8min is acceptable.

The thing is - not to get too wound up with the latest and greatest or even some super-duper chip when an entirely satisfactory solution can be had with a 555.

JMHO

[larsdenmark]

@Benta - You are probably right about the CD4538. However, the world is short on tutorials on how to use the CD4538 or why one should use it.
Perhaps you could share a few (more) words about how and when to use the CD4538?

A quick search finds that the 555 is a lot cheaper than the CD4538 and the 555 is a lot easier to get.

[Zero999]

Real Engineers may use a discrete monostable, but many ordinary people just use what they have available in their junqueboxes.

What? You don't have a couple of transistors handy? A transistor would most be required to drive a motor anyway, as unless it's a very small motor, the 555 won't be able to output enough current.

My previous post was a bit tongue in cheek: of course, use whatever is available.

And no one spotted my mistake? With high current loads, a resistor is needed to limit the base current to Q1. In the usual configuration it's not needed, since the load on the opposite transistor is normally small, so won't exceed the base current rating of the transistor for long enough to case any damage, but in this case the motor will allow a huge current to flow. R5 limits it to around 44mA, which may be a little over the BC549's rating, but the surge won't last for long enough to cause any damage.

[tautech]

Read my reply #2 again. A 4538 will give you inverted input, non-inverted input, inverted output, non-inverted output, retriggering, non-retriggering and is probably cheaper. Why are you f*cking around with a 555?

Sorry, I had to get that out of my system, but it's a mystery to me why all designs are centered on a 555.

Agreed.
There are other worthy alternatives too 1/2 of a 4013 or 4047.
Sure none have the drive capability of a 555 but for power switching a silicon switch is needed anyway. Logic level MOSFETs are a good match for 4000 IC's.

[hydrolisk1792]

Build a 555 one shot "monostable" throw a general purpose transistor on the output, even put an led there to see of the circuit is on from the board.  Then throw a relay on that transistor, don't forget the flyback diode. Then hook the motor to the relay, and that is that. Bob's your uncle!
Hope this helps.

[Benta]

@Benta - You are probably right about the CD4538. However, the world is short on tutorials on how to use the CD4538 or why one should use it.
Perhaps you could share a few (more) words about how and when to use the CD4538?

A quick search finds that the 555 is a lot cheaper than the CD4538 and the 555 is a lot easier to get.

The 4538 or 74HC4538 is a good choice whenever you need an edge triggered monostable multivibrator (MMV). Note the qualification: if you want a level triggered input (eg., when adding a certain extra time to a variable length input pulse) the 555 is better suited.

There are no tutorials on the 4538, simply because it is super-simple to use and the datasheet gives all the needed information. Choose a resistor and a capacitor for the desired pulse length and you're done.
As a bonus, you get two MMVs in the same package.

Concerning "a lot cheaper": Mouser shows 40 cents for the MC14538 and 30 cents for the NE555. Considering the monstrous capacitor you'll need for making a 5-minute delay with the 555, I'm certain that a 4538 solution will be much cheaper in the end.


Edit: the 7555 (CMOS version of the 555) costs 45 cents. And at such long time delays you're forced to use that version. So much for "cheaper".

[CopperCone]

If you dont mind paying more but you want easy and neat look at timerblox chips

[Wimberleytech]

The 4538 or 74HC4538 is a good choice whenever you need an edge triggered monostable multivibrator (MMV). Note the qualification: if you want a level triggered input (eg., when adding a certain extra time to a variable length input pulse) the 555 is better suited.

There are no tutorials on the 4538, simply because it is super-simple to use and the datasheet gives all the needed information. Choose a resistor and a capacitor for the desired pulse length and you're done.
As a bonus, you get two MMVs in the same package.

Cool, I have never used this part...learn something new every day here.  I noticed that the datasheet does warn about using large capacitors that can discharge into the chip on power down.  It tells how to solve this with an extra diode.

[larsdenmark]

@Benta - You are probably right about the CD4538. However, the world is short on tutorials on how to use the CD4538 or why one should use it.
Perhaps you could share a few (more) words about how and when to use the CD4538?

A quick search finds that the 555 is a lot cheaper than the CD4538 and the 555 is a lot easier to get.

The 4538 or 74HC4538 is a good choice whenever you need an edge triggered monostable multivibrator (MMV). Note the qualification: if you want a level triggered input (eg., when adding a certain extra time to a variable length input pulse) the 555 is better suited.

There are no tutorials on the 4538, simply because it is super-simple to use and the datasheet gives all the needed information. Choose a resistor and a capacitor for the desired pulse length and you're done.
As a bonus, you get two MMVs in the same package.

Concerning "a lot cheaper": Mouser shows 40 cents for the MC14538 and 30 cents for the NE555. Considering the monstrous capacitor you'll need for making a 5-minute delay with the 555, I'm certain that a 4538 solution will be much cheaper in the end.

Edit: the 7555 (CMOS version of the 555) costs 45 cents. And at such long time delays you're forced to use that version. So much for "cheaper".

Thank you for this information. I'm always willing to learn something about a new group of ICs. The datasheet for the ON Semiconductor version of the MC14538B has a nice little graph that I've attached. It shows that the MC14538B is only good for a pulse up to 10 s whereas  the MC14541B can do 5 minutes and the MC14536B can do 23 hours(1), which I guess is a lot more than the 555 can handle.

The datasheet for the NE555 says: "Timing From Microseconds to Hours" so it should do pulses of 5 minutes. Since I have a NE555 I made a small test with a 100 uF cap and a 2.7 MOhms resistor and it gives a pulse of slightly more than 5 minutes. I can buy the NE555 at TaydaElectronics for $0.13.

[Benta]

The reason for the long times of the MC14541 and MC14536 is, that they're counters with built-in oscillator. Different story altogether.

The MC14538 will do 5 min. comfortably, but due to cap leakage currents not as precisely as up to 10 sec. But anyway, if you want precise timing, the oscillator/counter is the way to go.

The 555 claim of timing to "hours" is hard to believe. Finding a big cap with so little leakage current will be a major and expensive task and it will probably be the size of a whisky bottle, reducing the semiconductor cost to peanuts.

Who is Tayda Electronices? I only buy from authorized sources, which costs me a bit more, but saves me from a lot of grief.

[larsdenmark]

It may be fun to try using a MC14536 (but a MCU may be easier to work with).

From what I can read both the 555 and MC14538  doesn't say anything about accuracy beyond 10 s. For a 1 s pulse the 555 has an accuracy of 1% whereas the MC14538  has an accuracy of 5%. Based on the original problem of just being able to have a pulse of 5 minutes I see no reason to not use a 555. However, there may be other very good reasons to use a MC14538 since it seems to support (as you write) a lot more options on triggering and input/putput handling.

Tayda is www.taydaelectronics.com . They are based in Thailand. I have used them for years and they have never disappointed me. Shipping to Denmark is 7 days.

[Zero999]

It may be fun to try using a MC14536 (but a MCU may be easier to work with).

From what I can read both the 555 and MC14538  doesn't say anything about accuracy beyond 10 s. For a 1 s pulse the 555 has an accuracy of 1% whereas the MC14538  has an accuracy of 5%. Based on the original problem of just being able to have a pulse of 5 minutes I see no reason to not use a 555. However, there may be other very good reasons to use a MC14538 since it seems to support (as you write) a lot more options on triggering and input/putput handling.

Tayda is www.taydaelectronics.com . They are based in Thailand. I have used them for years and they have never disappointed me. Shipping to Denmark is 7 days.

For long delays, the semiconductor element, be it an IC or discrete components, is not the limiting factor which determines the accuracy, but the tolerance of the capacitor. For short delays, small capacitors can be used, <10nF which can be found down to 1% tolerance. Longer delays will require an electrolytic capacitor, which will have a tolerance of 20% or poorer.

For example, the discrete circuit I posted previously will have a much poorer accuracy than an IC such as the '555 or '4538 for ms delays, because the base-emitter voltage is not perfectly stable, but for 10 second delays or longer, the accuracy will be similar, because the 20% tolerance of the electrolytic capacitor will swamp and variances is base-emitter voltage.
https://www.eevblog.com/forum/beginners/a-one-time-timer/msg1574389/#msg1574389

[james_s]

I've encountered the MC14541 in a few things, it's a neat part. Being a counter driven by an oscillator you can get quite long delays without using outlandish component values.