Maker Pro

Beginner circuit question - 555 timer

Is Dave an idiot?


  • Total voters
    1

dbarak

Apr 25, 2010
16
Joined
Apr 25, 2010
Messages
16
Hello,

I've got what I'm guessing is a pretty simple question for most of you. I want to blink three LEDs that are in series, with a LOOONG duration on and off, about five minutes at a 50%-ish duty cycle. I found a calculator online that let me determine component values based on the time I wanted, and I found a calculator that let me figure the best resistor value for LEDs in series. One diagram I found shows a 10μf capacitor between pin 5 and ground, but another shows a a circuit that's identical except that pin 5 isn't connected to anything. For now I'm going with the circuit that DOES have the capacitor, but it can obviously be taken out of the circuit easily if it's not needed or if it actually would cause problems.

So, would anyone be willing to give my circuit diagram a quick review? I mapped this out based on the actual pin locations of a 555, rather than the more common diagram showing the pins in the most expedient locations (4,7,6,2,1,5,3,8).

I'm starting to figure out some of this stuff on my own, so I'm looking for a second, much more experienced set of eyes to make sure I'm not messing this up. Thanks all!

Dave
 

Attachments

  • flasher-circuit.jpg
    flasher-circuit.jpg
    17.6 KB · Views: 108

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
10uF seems large to me.
Are you sure it's that size as mostly it would be maybe 0.01uF, put there to eliminate noise much like a cap on the supply rails to other chips.
If you do a search you will find there are also recommended maximum resistor/cap combos to use with the 555.
Long timing is usually achieved in a different manner.

Try here..... http://www.talkingelectronics.com/projects/50 - 555 Circuits/50 - 555 Circuits.html
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Hi Dave and welcome to Electronics Point :)

You have the right idea, but as Bluejets pointed out, five minutes is too long for a resistor-capacitor delay. There are other ways to do it though. A common one is to use the CD4060 (see http://www.digikey.com/product-detail/en/CD4060BE/296-2060-5-ND/67311).

Download the data sheet http://www.ti.com/lit/ds/symlink/cd4060b.pdf and have a look at these links to learn about how the 4060 works.

http://www.brighthubengineering.com...imer-oscillator-circuits-using-a-single-chip/
http://www.zen22142.zen.co.uk/ronj/24h.html
http://www.doctronics.co.uk/4060.htm
http://www.coolcircuit.com/circuit/timer_4060/
http://homemadecircuitsandschematics.blogspot.co.nz/2012/01/how-to-make-simple-versatile-timer.html

Those tutorials are all pretty similar but you may find that some of them explain things better than others.

Some of them have a diode connected from the output back to the oscillator. This is used when you want a "one-shot" timer that waits a certain amount of time, activates (or deactivates) its output, then stops. You want a continuous oscillating timer, so don't use the diode.

Steve has written a tutorial here at https://www.electronicspoint.com/resources/long-duration-timers-notes-for-beginners.4/ that you may find useful.

The CD4060 can operate from any supply voltage from 3V to 15V but 12V is a good choice.

You need to look into your LED drive as well. With a 9V power supply and three LEDs with forward voltages of 3V, there is no voltage available for the current limiting resistor, and the LED current will be very poorly controlled. Also you need to switch the LEDs using a transistor or a MOSFET, which will have a consistent and fairly low voltage drop.

Have a look at Steve's LED driving tutorial at https://www.electronicspoint.com/resources/got-a-question-about-driving-leds.5/
 

dbarak

Apr 25, 2010
16
Joined
Apr 25, 2010
Messages
16
Thanks Bluejets! The circuit I found that listed 10μf is here: http://www.ohmslawcalculator.com/555_astable.php - but now that I think about it, since it's hand-drawn, could it be 10nf? (Nano-Farads I guess?)

That link you provided could keep me busy for months, if not years! :) I really like that the circuits are explained and I don't have to go bouncing all over the web to find the basic information in one spot.

10uF seems large to me.
Are you sure it's that size as mostly it would be maybe 0.01uF, put there to eliminate noise much like a cap on the supply rails to other chips.
If you do a search you will find there are also recommended maximum resistor/cap combos to use with the 555.
Long timing is usually achieved in a different manner.

Try here..... http://www.talkingelectronics.com/projects/50 - 555 Circuits/50 - 555 Circuits.html
 

dbarak

Apr 25, 2010
16
Joined
Apr 25, 2010
Messages
16
Hi KrisBlueNZ,

Thanks for the assistance! I found an online calculator that verified for me (finally) how the LED vf thing works in regards to current limiting resistors. Essentially, assuming everything operated perfectly, three 3vf LEDs would use a 9v supply with no resistor. If I were using 3.2vf LEDs I assume that would work but they'd be a little dimmer.

As you can tell, my knowledge of this stuff is very basic, not much beyond understanding polarity, seriously. :rolleyes: That 4060 chip sounds interesting, and apparently, judging from some of the circuits I checked out, I can run up to four LEDs at different on/off cycles.

I've started checking around online to see what maximum capacitor/resistor combinations I can use with a 555 (not for this project now, just for my education). I'm having a bit of a hard time, but can you advise what the approximate longest times achievable would be, whether high, low, or 50% duty cycle? I figure that at worst, if I can't find the maximum component combination, I could plug the maximum time(s) into one of the online calculators to see approximately what the components would be.

Dave

Hi Dave and welcome to Electronics Point :)

You have the right idea, but as Bluejets pointed out, five minutes is too long for a resistor-capacitor delay. There are other ways to do it though. A common one is to use the CD4060 (see http://www.digikey.com/product-detail/en/CD4060BE/296-2060-5-ND/67311).

Download the data sheet http://www.ti.com/lit/ds/symlink/cd4060b.pdf and have a look at these links to learn about how the 4060 works.

http://www.brighthubengineering.com...imer-oscillator-circuits-using-a-single-chip/
http://www.zen22142.zen.co.uk/ronj/24h.html
http://www.doctronics.co.uk/4060.htm
http://www.coolcircuit.com/circuit/timer_4060/
http://homemadecircuitsandschematics.blogspot.co.nz/2012/01/how-to-make-simple-versatile-timer.html

Those tutorials are all pretty similar but you may find that some of them explain things better than others.

Some of them have a diode connected from the output back to the oscillator. This is used when you want a "one-shot" timer that waits a certain amount of time, activates (or deactivates) its output, then stops. You want a continuous oscillating timer, so don't use the diode.

Steve has written a tutorial here at https://www.electronicspoint.com/resources/long-duration-timers-notes-for-beginners.4/ that you may find useful.

The CD4060 can operate from any supply voltage from 3V to 15V but 12V is a good choice.

You need to look into your LED drive as well. With a 9V power supply and three LEDs with forward voltages of 3V, there is no voltage available for the current limiting resistor, and the LED current will be very poorly controlled. Also you need to switch the LEDs using a transistor or a MOSFET, which will have a consistent and fairly low voltage drop.

Have a look at Steve's LED driving tutorial at https://www.electronicspoint.com/resources/got-a-question-about-driving-leds.5/
 

dbarak

Apr 25, 2010
16
Joined
Apr 25, 2010
Messages
16
By the way Bluejets and KrisBlueNZ,

I should let you know what I'm trying to build - it may help you get me on a more direct path to achieving my end goal. I'm working on a short film, and we're building a spacecraft set, which will include panels with caution lights as found in aircraft (and spacecraft) cockpits (see example image). I've got 30 lights, each using three LEDs to create a bright, wide, even spread of backlighting, and one double size light that will use two banks of three LEDs. I've already got the LEDs on hand - 20ma and about 3.0 to 3.2 vf (I've been figuring using 3vf for safety). Ideally, I'd have these lights going off randomly at very slow cycles - on for several seconds, off for long if possible, but a 50% duty cycle okay if necessary. Typical warning lights don't flash, although the larger one, the Master Caution light, normally would flash in order to attract attention, prompting closer examination of the panel.

I actually built an Arduino-controlled slow-spinning motor controller that spins a disk with embedded magnets, something like six concentric circles of randomly-spaced magnets. Each circle has a magnetic switch positioned above it, and that would in turn control the timing of LEDs. Although it works, I just don't trust the mechanical aspect - I'm at least as bad with mechanical things as I am with electronics. Put Photoshop in front of me and I'm fine - anything else and I embarrass myself. :)

To give you a little bit of what the whole thing is about, you can visit the film's website - http://milestogo-movie.com/wp/. It's all unpaid work, as is most of the filmmaking going on. Of course the goal is to eventually turn this into something bigger, but the odds of that are pretty low. And if you want credits in the final short film, I'd be happy to oblige!

Dave

PS: I'm not offering the credit in the film for design services, just for the help you've already provided. :)
 

Attachments

  • caution.jpg
    caution.jpg
    278.9 KB · Views: 121
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Essentially, assuming everything operated perfectly, three 3vf LEDs would use a 9v supply with no resistor.
Yes, but VF for any particular current (IF) is not tightly controlled. It varies from one batch of LEDs to another; it varies somewhat between LEDs in the same batch; it varies with temperature; it varies with age. The simple fact is that driving an LED with a fixed voltage is bad juju. It can cause the LED to go into thermal runaway, where it gets hotter, so it conducts more, so it gets hotter, so it conducts more, and so on, until it's more of a HED than an LED!
If I were using 3.2vf LEDs I assume that would work but they'd be a little dimmer.
Yes. The nominal VF of an LED is specified at a specific "standard operating" current. The relationship between voltage and current is given in the data sheet in the typical characteristics. (These graphs are just indicative; they're not guaranteed in any way.) Here's a VF vs. IF graph from the typical characteristics section of the data sheet for a common green LED:

Typical LED VF vs. IF.png
With an LED that matches that typical graph, 10 mA forward current will flow when there is about 1.9V across the LED. So for a 10 mA operating current, its VF is 1.9V. But another LED with the same part number could have a VF of 2.0V at the same current - 0.1V higher. So for that second LED, the curve on the graph would be shifted to the right (actually that's a simplification, but let's go with it), and applying 1.9V to that second LED would cause only about 4 mA to flow.

The LED's brightness is roughly proportional to the current. The voltage only relates to the current, according to the graph.
As you can tell, my knowledge of this stuff is very basic, not much beyond understanding polarity, seriously. :rolleyes:
Perhaps that explanation will help :)
That 4060 chip sounds interesting, and apparently, judging from some of the circuits I checked out, I can run up to four LEDs at different on/off cycles.
Actually it has ten outputs. They are named Q4, Q5, Q6, Q7, Q8, Q9, Q10, Q12, Q13 and Q14. Q4 is a square wave signal at 1/16th the frequency of the oscillator. Q5 is half the frequency (twice the duration) of Q4, and so on - each Qn output toggles at half the rate of the previous Qn output. For some unknown reason, the designers decided to omit Q11. It's there inside the chip, but you can't get to it. But you can use any outputs you want.
I've started checking around online to see what maximum capacitor/resistor combinations I can use with a 555 (not for this project now, just for my education). I'm having a bit of a hard time, but can you advise what the approximate longest times achievable would be, whether high, low, or 50% duty cycle? I figure that at worst, if I can't find the maximum component combination, I could plug the maximum time(s) into one of the online calculators to see approximately what the components would be.
There isn't a hard-and-fast rule. The limiting factor is the leakage current in the capacitor. Ideally, the capacitor charges and discharges through the resistor(s) and this can happen at any frequency, but to get a low frequency, you need a large capacitance and a high resistance.

For the high capacitance, you need to use an electrolytic capacitor, and these have a leakage current that's equivalent to a high-value resistor (a very weak path for a small amount of current flow) in parallel with them. When you combine that with a high-value timing resistor in the circuit, which only produces a very small charging current, the leakage current in the electrolytic can be so high that the capacitor doesn't actually charge up, and the 555 gets stuck with its output high, with nothing happening.

The inputs of the 555 itself also have leakage currents that can affect the oscillator's operation when very high resistor values are used, especially if you use the original 555 rather than one of the improved variants.

There are some capacitor families that are specifically designed for low leakage current (this can be important in some other applications as well) but generally you're limited to oscillator periods in the low minutes range.
 
Last edited:

dbarak

Apr 25, 2010
16
Joined
Apr 25, 2010
Messages
16
I actually understand what you posted. Some of it confirms what I suspected, mainly how the capacitor and resistor team up to manage the timing. I also get how the leakage factor for long durations would be unworkable.

I've ordered a few of the 4060s - one to kill by accident via static electricity, one to kill by accident from bad circuit design, and a few extras to actually use. :)

Thanks for all the help!
 
Top