Creating a Simple Timer Circuit without an IC

[Xyius]

So I started really getting into circuits about a year ago and it has quickly become my new favorite hobby. (A little background into my education, I am a senior physics undergraduate.) Since I started doing circuits I have wanted to learn everything that I can about them. I don't like using IC's without having at least some knowledge of how they work first. That being said I want to make a simple timer circuit. How would I be able to do this without using an IC? In my eyes, using an IC doesn't teach me anything. At this point, all I want to do is learn.

My idea that I had was to use a chaser circuit. Something like this..

http://img708.imageshack.us/img708/9961/simplechaser.png

Now this circuit will have the LED's blink in succession, one right after the other. Having a lower voltage as the source causes it to blink slower. So my thinking is, if I just remove all the LED's except the last one, I can use that as a timer and just adjust the voltage to set the time it takes to turn on. The problem with this is that it only turns on as long as the capacitor is discharging. As soon as the cap is finished discharging it turns off and the cycle continues.

Any better ways to do this? (probably)
Thanks! :D!

 

[Averagesupernova]

You need to change the attitude that using ICs doesn't teach you anything. I have found that there is a lot of advice out there (including here on PF) that involves using a microcontroller to do something which seems incredibly simple. Using a microcontroller does in fact teach you how to use it which is a valuable skill, but I believe in starting at the basics. A better approach for your project in my opinion would be to look at the block diagram of a 555 timer IC. You could build a variation of it with discreet parts if you are set on doing so. Once you understand how the 555 works you will probably change your approach. What you have now is an oscillator that I believe would not be very stable over the temperature range, etc.

 

[Xyius]

You need to change the attitude that using ICs doesn't teach you anything. I have found that there is a lot of advice out there (including here on PF) that involves using a microcontroller to do something which seems incredibly simple. Using a microcontroller does in fact teach you how to use it which is a valuable skill, but I believe in starting at the basics. A better approach for your project in my opinion would be to look at the block diagram of a 555 timer IC. You could build a variation of it with discreet parts if you are set on doing so. Once you understand how the 555 works you will probably change your approach. What you have now is an oscillator that I believe would not be very stable over the temperature range, etc.

Yeah I didn't mean to say it doesn't teach you anything. I just don't like using something that I have no idea how it works. All I would need to do is make the circuit once then I would use the IC.

 

[Averagesupernova]

Personally I think you have it backwards. Use the IC first, then take the pertinent measurements while looking at the data sheet to understand it. An IC is something someone else put a lot of thought into and there is no reason for you to reinvent the wheel. Learn from what the developers of the IC learned while they designed the device. Of course with the majority of ICs out there it is impossible to do what they have done with discreet components. However, there are some older ICs that could be built up or at least variations of them. Good luck getting something like an op-amp with an open loop gain of 100,000 (for instance) to work with discreet components though. You may get some form or function of operation but it won't work over the range or bandwidth that an IC would work.

 

[vk6kro]

You could try some simple monostable circuits like this:

http://www.play-hookey.com/digital/experiments/images/rtl_monostable_sch.gif

The transistors suggested were 2N3904, so almost any NPN would probably be OK.

You can read about it here:

http://www.play-hookey.com/digital/experiments/rtl_monostable.html



Great advice about using ICs though. You can learn about them in the data sheets freely available on Internet, if you want to, but mostly you need to learn about what the pins are for and what to expect from the outputs.
Good free datasheet search engine: http://www.datasheetpro.com/

Electronics isn't all about learning. Sometimes you need to do something useful and ICs are great at getting a job done effectively. Fortunately for us, they are mostly pretty cheap, too.

 

[Averagesupernova]

You could try some simple monostable circuits like this:

http://www.play-hookey.com/digital/experiments/images/rtl_monostable_sch.gif

The transistors suggested were 2N3904, so almost any NPN would probably be OK.

You can read about it here:

http://www.play-hookey.com/digital/experiments/rtl_monostable.html



Great advice about using ICs though. You can learn about them in the data sheets freely available on Internet, if you want to, but mostly you need to learn about what the pins are for and what to expect from the outputs.
Good free datasheet search engine: http://www.datasheetpro.com/

Electronics isn't all about learning. Sometimes you need to do something useful and ICs are great at getting a job done effectively. Fortunately for us, they are mostly pretty cheap, too.

My italics/bold. And it is great advice. Getting a useful end result psychologically provides the feedback which helps motivate learning.

 

[es1]

It's a little bit funny because this is exactly the technique used in some ICs to make clocks (a.k.a. a timer)
http://en.wikipedia.org/wiki/Ring_oscillator
Notice especially the three inverter oscillator image on the right of the page.

 

[DragonPetter]

It's a little bit funny because this is exactly the technique used in some ICs to make clocks (a.k.a. a timer)
http://en.wikipedia.org/wiki/Ring_oscillator
Notice especially the three inverter oscillator image on the right of the page.

I'm curious, in the wikipedia article it says this is used commonly in VCO implementations. How would a ring oscillator's operating frequency be varied with this? 3 identical varactor diodes?

 

[es1]

Xyius's circuit shows it pretty well. The Vbe of the NPNs stay the same. The R1/C2, and other, time constants all stay the same too. As V1 lowers it will take more multiples of R1/C2 (or more time constants) to get to Vbe. This slows the clock. Increasing V1 means it takes less time to get to Vbe on C2, et. al, which means the clock runs faster. Works very similarly for FETs, etc.

 

[skeptic2]

That being said I want to make a simple timer circuit. How would I be able to do this without using an IC? In my eyes, using an IC doesn't teach me anything. At this point, all I want to do is learn.

My idea that I had was to use a chaser circuit. Something like this..

Now this circuit will have the LED's blink in succession, one right after the other. Having a lower voltage as the source causes it to blink slower. So my thinking is, if I just remove all the LED's except the last one, I can use that as a timer and just adjust the voltage to set the time it takes to turn on. The problem with this is that it only turns on as long as the capacitor is discharging. As soon as the cap is finished discharging it turns off and the cycle continues.

Any better ways to do this? (probably)
Thanks! :D!

The best thing to do before you begin a project like this is write down everything you want the circuit to do. What is the maximum time you need the timer to time? How much accuracy and repeatability do you need? Does it have to be adjustable?

Once you have the specifications thought out, and with a little experience, it will become apparent that some circuits are better than others.

Yes you can learn a lot from ICs. You should study at least the block diagram of the 555. It charges and discharges between 1/3 and 2/3 of Vcc. That makes its timing interval very repeatable even using different voltages.

 

[DragonPetter]

Xyius's circuit shows it pretty well. The Vbe of the NPNs stay the same. The R1/C2, and other, time constants all stay the same too. As V1 lowers it will take more multiples of R1/C2 (or more time constants) to get to Vbe. This slows the clock. Increasing V1 means it takes less time to get to Vbe on C2, et. al, which means the clock runs faster. Works very similarly for FETs, etc.

Thanks for the explanation :)