Hi!
First of all, it's my first approach to electronic so please be nice with my ignorance!
I'm trying to make this with leds in order to simulate a plane anti-collision lights:

With a lot of hours of work and learning (Really I never touched anything in my life) I could managed myself to draw this in crocodile clips, adding an example, and another one, and another one. I know maybe it's very wrong but well, I'm learning.

With this I've two problems.

1) I don't know why the three "collision simple" strobs are so low.
Anyway, the post is about the second and most important question:

2) How to made that collision simple in double strobe like the blue line in figure 1?
I tried adding an example about "uneven clicks" but I couldn't manage how to replace the speaker by the led.

Thanks in advance and sorry for my stupid question!

 

I figured myself the first one, just writing it! The resistance of 1k on the output is heavy. I put down then to 500 both!

 

You're on the right track by defining the timing with a timing diagram. I counted 17 steps of 50ms which could be built with a 555 and two 4017s. Take a look at this old thread and look for others using a 555 with a 4017.

http://www.edaboard.com/thread218710.html

I just noticed that this is posted in "Homework Help". Is this an assignment or project for school?

 

Thanks a lot for your reply! I'll check that link!
Actually I'm C++ programmer, making my first illuminated model Airbus 1/144. Just for hobby and learn! But I don't find another thread to post this :S There is one?

 

The general electronics chat forum would be appropriate. I'll ask a moderator to move the thread there for you.

Since you are a programmer, you may find it useful to do this project with a microcontroller such as an Arduino or PIC.

 

KJ6EAD, your simulation is perfect! Thanks a lot! I saw that forum but I never saw your applet!
I'll try to simulate it!

I saw something about Arduino, the thing is that my knowdledge of electronic is really basic so I think (I don't know really) but I think several 555 is more easy to understand to me. I will try to understand the 4017, I took a lot of time to understand the 555

 

Like KJ6EAD, I'd use a 555 and CD4017 to do this. The CD4017 can't provide enough current to drive the LEDs directly, so an NPN transistor would be needed as a driver.

 

KJ6EAD, your simulation is perfect! Thanks a lot! I saw that forum but I never saw your applet!
I'll try to simulate it!

I saw something about Arduino, the thing is that my knowdledge of electronic is really basic so I think (I don't know really) but I think several 555 is more easy to understand to me. I will try to understand the 4017, I took a lot of time to understand the 555

The 4040 might be a better choice than the 4017 - you get a choice of pulse widths on successive outputs, then you just need some basic gate "glue logic" to combine those pulse widths into the pulse train you require.

Some of the PIC, AVR etc are available in low pin count packages, if you only need one pin to be an output - I think one of them is available in what looks like a 5 or 6 pin SMD transistor.

 

If you use a CD4060, you can eliminate the 555 timer(s). Add some logic to derive the pulse sequences you want.

It looks like there's a mistake on the red waveform. The first pulse isn't centered in the 1 second low on the blue waveform; the second is.

Lastly, you shouldn't wire LEDs in parallel; each should have it's own current limiting resistor.

 

The "some logic to derive the pulse sequences (he) want(s)" far outweighs the separate 555 oscillator, and the 555 freq is much more stable over voltage and temperature. While his blue line shows the mark as twice the space, videos of other models and of the actual plane indicate a mark/space of 1:1. At the 4017, diode-OR outputs 0 and 2 to the collision lights, output 5 drives the beacon, tie output 9 to Reset and it's done.

ak

 

Thanks for all replis. Sadly I don't know too much about electronic so many answers are really hard to understand to me, but I'll try testing what you response. Thanks again!

 

I was thinking along the same lines as AK, post #10.

 

If you use a CD4060, you can eliminate the 555 timer(s). Add some logic to derive the pulse sequences you want.

.

The 4060 has the advantage over the 4040 that I suggested, of having an oscillator front end that you just add a CR timing network and not much else. The downside of the 4060 is less divide stages - and I think, less of the Q outputs are brought out to pins, but I'm not certain of that so the TS should compare the data sheets before ordering.

 

Something like this:

 

The downside of the 4060 is less divide stages - and I think, less of the Q outputs are brought out to pins

The 4060 is a binary counter whose outputs need external decoding logic. The 4017 is a decade counter with already-decoded outputs.

 

Add my vote to the CD4017 solution.

 

The 4060 is a binary counter whose outputs need external decoding logic. The 4017 is a decade counter with already-decoded outputs.

The 4017 sequentially pulses each of the 10 outputs 1/10 the width of the clock pulse - since you only have one pulse at any given point in time, its impossible to use combinatorial logic to generate gated sequences of pulses.

With the 4040 and 4060 etc you can use the slower Q outputs to gate the faster ones on and off to generate a variety of custom pulse patterns.

 

thanks folks! Thanks ak, I'm trying to figure out how can I transform that to a battery flow but it really upgraded my crocodile simulation

 

The 4017 sequentially pulses each of the 10 outputs 1/10 the width of the clock pulse - since you only have one pulse at any given point in time, its impossible to use combinatorial logic to generate gated sequences of pulses.

Actually, the 4017 pulses each of the 10 outputs the full width of one clock period. In the above schematic the 555 is running at approx. 2 Hz, and each 4017 output is high for 0.5 s. Also, note the two diodes forming an OR gate function to combine two outputs to produce a gated sequence of pulses.

While a 4040/60 can produce much more complex gated sequences (with much more complex gating), that is not required by the original question.

Edit: The 555 timing resistor should be 160K for 2 Hz output frequency.
ak

 

Actually, the 4017 pulses each of the 10 outputs the full width of one clock period.

It was getting late when I typed that - what actually happens is, its rather like each clock pulse gets routed in turn to one of the 10 outputs in sequence.