Here is a quick GIF of what I'm hoping the interior part of the design (not the outer ring) will look like. The timing is a bit messy (given it was windows movie maker). I would plan on having 2 555's to control the clock signal for the bottom square wave and the top bar (so they finish at the same time, as in the gif they do not).

You guys/gals have all been a great help! I can't thank you enough,
-Dan

EDIT*** The LEDs that illuminate throughout the square wave are correct, but the ones on the bar above them are incorrect (after looking over that timing diagram, I realized I had it wrong). I will add a second 555 to control the top bar which will be easy enough. Now that that's established, this project has just become along easier. Now its the Comet Trail Ring comprise of as many LEDs as possible. I will also have to play wit the frequency to ensure that the square wave, top bar, and circle around it all finish their loops at the same time; but that should just be a matter of picking the right potentiometer, no?

 

For the circle of leds, i would use a Lm3914 chip in Bargraph mode, and cascade them, this would involve a ramp generator, sweeping the voltage at the input.

 

Here is an outline of snake driving the bar. Inputs are high as might be derived from 4017 bar graph.
All LED anode's go to + V supply. Operation matches new video.
For low level inputs, like from a shift register, might use 4082 ,4 in AND, but need a low level input driver.
EDIT: Shift register, SR, like 74HC164 , output determined by loaded input, it was the decoders that have active low outputs. My pick would be the SR as easiest , just load ones, when full, reset.

 

Some assumptions: ring, LED circumference 30 in., snake clock = 7 Hz., travel time 5 sec.+-, 3 X 7 Hz= 21
Hz., @ .295 in/ LED, 30/ .295 = about 105 LEDs in outer ring ? Other way is to find common denomenator
in clocks & use phase locked loop , lost word. Must run.

 

Here is a quick GIF of what I'm hoping the interior part of the design (not the outer ring) will look like. The timing is a bit messy (given it was windows movie maker). I would plan on having 2 555's to control the clock signal for the bottom square wave and the top bar (so they finish at the same time, as in the gif they do not).

You guys/gals have all been a great help! I can't thank you enough,
-Dan

EDIT*** The LEDs that illuminate throughout the square wave are correct, but the ones on the bar above them are incorrect (after looking over that timing diagram, I realized I had it wrong). I will add a second 555 to control the top bar which will be easy enough. Now that that's established, this project has just become along easier. Now its the Comet Trail Ring comprise of as many LEDs as possible. I will also have to play wit the frequency to ensure that the square wave, top bar, and circle around it all finish their loops at the same time; but that should just be a matter of picking the right potentiometer, no?

If the X-axis is supposed to represent Time, why is time pausing for the LEDs to clime the verticals of your square wave? I think they should just pop all the verticals in at once, like a real square wave.

 

I will add a second 555 to control the top bar which will be easy enough.

You should use one master clock for the whole circuit. For the square wave and bar in my simulation, I used the same counters for both sets of LEDs. To get the pause on the vertical edges, I just used a 4 input OR gate.

If you make the number of LEDs in the ring a multiple of the square wave, you won't have to worry about keeping the LEDs in sync.

 

If the X-axis is supposed to represent Time, why is time pausing for the LEDs to clime the verticals of your square wave? I think they should just pop all the verticals in at once, like a real square wave.

You're right, I'm a power engineer who deals with kilovolts all day... low power digital logic is something I'm still learning thanks to you guys!

-Dan

 

Here is an outline of snake driving the bar. Inputs are high as might be derived from 4017 bar graph.
All LED anode's go to + V supply. Operation matches new video.View attachment 123487
For low level inputs, like from a shift register, might use 4082 ,4 in AND, but need a low level input driver.
EDIT: Shift register, SR, like 74HC164 , output determined by loaded input, it was the decoders that have active low outputs. My pick would be the SR as easiest , just load ones, when full, reset.

Thanks Bernard!!! You seem like you have a cornucopia of digital electronics up your sleep. I'm relatively fresh out of college so I have no shame admitting that I know nothing... yet

-dan

 

I should have more to show for it being out of college for 67 years & still do not do PICs.
Here is an outline of a shift register, like in the video. High level output current stinks, around 400 uA
but still illuminates the LEDs. Would use ULN2803 drivers.
Back to comet tail from bar graph? Might use a transistor, MPSA13, for each LED + R & a C. Could insert a comet every second or so ?
How many of these plaques are required?

 

I should have more to show for it being out of college for 67 years & still do not do PICs.

Really! Dont tell me that you're 88 now....

I thought I'll have to stop messing with electronics when I'm 70. My eyesight and my shaky hands wouldnt allow me to do electronics anymore. Besides, I'll be too busy playing with grand-children.

Allen

 

I'm really just 90, retired in 1986. Bowled a 160 yesterday- or was that two weeks ago?
Here is my version of comet tail ring. Just one of the LED decay ckts. shown. The decay is set by Rx X Cx.
Approx. values might be R = 1 meg. & C = 4.7 uF for about 3 sec. Assuming that on power-up all are reset,
all outputs are low, U2-1&2 pins are high. On first clock a high will be loaded into the first stage, input goes low.
With following clocks the high will travel to the last stage +1 where the output is inverted to clear all stages. Clear is active low.
Be back later.

 

I should have more to show for it being out of college for 67 years & still do not do PICs.
Here is an outline of a shift register, like in the video. High level output current stinks, around 400 uA
but still illuminates the LEDs. Would use ULN2003 drivers.
Back to comet tail from bar graph? Might use a transistor, MPSA13, for each LED + R & a C. Could insert a comet every second or so ?
How many of these plaques are required?View attachment 123504

I'm not sure yet, the image I showed you of the square wave with the circle around it is our logo, so I'm just trying to use our free credit with our PCB fab house to do something nice for the engineers I work since they've helped me out so much and I hope to continue to build our culture with this logo (we already have made t shirts, ping pong balls and paddles, stickers, etc) that builds morale among young engineers at a smaller startup firm.

 

If I were making just one plaque, it would be dead bug style with SRs mounted upside down & spaced near ring
one SR for every 8 LEDs, but as there might be more, maybe a PCB same size as logo board, component side facing back of logo. If logo board is no more than 1.6 mm thick then there would be about 5 mm for component height. Solder all comp. except for LEDs. Drill LED holes slightly loose & possibly tapered. Hopefully the logo bd. will fit nicely over PCB with just the domes of the LED s protruding. This is assuming 5 mm LEDs. Boards held together with threaded studs glued to back of logo bd. A third board covers the back side with a rimed hoop ring covering the edges. ???

 

That's a good idea, I'll have Figure out the number of plaques first...

You guys are so helpful I can't believe it. I'm going to do some more learning on a protoboard before I make a prototype on a pcb... but be sure, Bernard--i William definitely ship you one of the final products!

-dan

 

An after thought: If more space is needed between PCB & logo bd. & LED leads do not have flats, then # 2 glass bugle beads make nice 4.7 mm spacers.

 

WE Aremissing a A SIMPLE circuitT hERE. SHIFT REGISTERS,aTYPdFLIPFLOP ARRAY ;IT IS SIMPLE AND CHEAP. I willdeaw some ideaas up when I get a reply to a PM.

 

I miss labeled IC on Dan's Plaque as ULN2003A, only 7 stages, should be ULN2803A, with 8 stages. Drawing is one + stages of snake, here after will be SQW. Lettered outputs go to bar decoding, Dan's Plaque.
Clock is the same as before. If SQW has 36 LED's , that requires 5 - 74HC164's, with reset taken from 5th stage of last SR.

 

Wendy, we have started a design using shift registers, 74HC164's, but there is always room for design improvements and better components.

 

So I generated a few .GIFs of how I'd like the final design to look, and I continued to realize that the top bar was no syncopated (it did not light up at a steady rate). I wasn't thinking that this is because I have more LEDs (36) in the square wave than I do on the top bar (18). While this makes sense as it is half of bulk of the LEDs, but I was thinking it would be much simpler if I make the top bar have the same amount of LEDs as the square wave (i.e., the square wave has 36 LEDs, as does the top bar). This is kind of like a knight rider circuit I made where instead of on set of LEDs scrolling from right to left (or vice versa), it was a knight-rider scheme that started in the middle and worked its way outward. The schematic for the knight rider schematic is uploaded.

The 2 GIFs of the final design are getting closer to where I'd like to end up, but I think the syncopation is wrong due to the number of LEDs (whic hI think one of you referenced earlier--if there were the same number, I could cascade 4017's similar to the schematic image I posted). I think I'm getting my point across; either way you guys have been a great help to me.

-Dan

 

I'm really just 90,

You just graduated from awesome to scary.

ak