Author Options:

Update - Circuit problem: Can anyone figure out what may be causing this project to only work half way? Answered

Ok, I have a recently built prototype of a circuit I was going to publish, but it is not working exactly as planned. It is not that it is not working at all, but it is acting strangely.
What is happening: pairs are blinking alternately, when they should be blinking in sequence. Can anyone see a fundimental flaw in my drawing, or is it a possible soldering error. Opinions and insights welcome.

PS: the included video is the first one I have even edited too (I forgot that the camera picked up sounds, and I had grunted, and verbally expressed surprise and it sounded inappropriate, so I edited it out.)

PPS: I believe I have to declare FAILure on this one, at least for now. Sorry....



9 years ago

Here's a guess as to the behavior... There are four stages, all "identical." Their tendency is to work in pairs; each wants to the inverse of the stage before. There is no stable condition here--once powered up, they want to either charge or discharge those capacitors, and to immediately restart the cycle. Since there are small differences between each component, the oscillation begins at a random stage, and then each pair starts to do it's thing. Try disconnecting (or grounding) the base of the first stage will create a stable state, then connect to initiate... Each stage still wants to be the opposite of the preceding stage, but maybe...

Thanks. Yeah, the person that showed me a similar schematic said that jumping the first capacitor momentarily would stabilize the blinking into sequence, but that appears to do nothing, which led me to believe in a possible cap failure. I will try what you suggest as nothing else has worked so far.

Notice that the pairs alternate (1 and 3 fire, then 2 and 4).

It altered the flash sequence for a moment or so, but then things went back to normal alternating flashes, as soon as the jump was removed. It didn't seem to matter how long or short I kept it there, either.

There's a circuit schematic that's almost identical here, which describes operation. Possibly a higher

However, it uses 5V filament lamps which draw 200mA. Assuming your components all check out OK, 100 ohms for a base resistor would seem too small for a 20mA load (ballpark LED load.)

Maybe that tends to bias ON. Once it's running, alternating behavior definitely looks like a stable condition.

It also describes starting the circuit by "jumping" one of the caps...

Regardless, a multi-LED chaser is pretty cool, too.

Oh, so that is where this is from originally....thanks. So, do suggest I increase the base resistance?

It's a guess. A smaller load (1/10) with a bipolar transistor calls for a higher value base resistor...

I will have to try that then, maybe a 1 k resistor to start with?

Sounds like a plan...

I never got it to work properly.....too much futzing with the parts even after installing pots.

Into the box of "partially unrealized projects..."

Yeah, it has to do with the load and the transistors I used somehow, but I don't have that much understanding of the process (nor the patience) to figure it out

Ignore the sentence fragm

gmoon has the answer:)

Have I represented the thing correctly here? If you right-click to "save as" you can edit it and post amended. Could you tell us exactly what components you're using? L


I hope this helps..... PS: I am not ABSOLUTELY sure about the diodes, but I think that is their part number.

Blinking Problem 003.jpgBlinking Problem 001.jpgBlinking Problem 002.jpgSequencer LED blink revsd.bmp

Yes, that happened to me (again) but I noticed and deleted them. So I got it right then? I think I might build this. (wouldn't know why it doesn't do what you want but someone else might? L

Yes, yours was correct (I was trying to make that drawing using Eagle, but I haven't learned how to use it yet, so it was taking too long :-) Like I said above though, I had the one I actually built, set up for 4 LEDs, and not just the three in the drawing I have to get some time to make sure I have no solder bridges across any of the caps.....I think that might be the problem, and if not, one or more of the caps may be bad (they are recycled from another board). The circuit looked so simple, and yet, without using a dedicated IC like 2 556's, or 74LS90 or whatever, it looked like it could sequence, with the next LED waiting until the former one turned off, before lighting. That was the effect I was after.

Um, I don't know how all those OTHER pics got attached....It was not my doing

I am going to have to see if I can get a better picture of that, or draw it up (all my tools are gone from reinstalling the OS, so I have to find and download and install them again.....suggestions for schematic drawing software?

I think a lot of people use Eagle. or just take a whack at it with Mspaint. I dunno, do you have to have a separate ground to isolate each transistor. or do you need a capacitor/diode to cascade the firing of the LEDs. or is leaking voltage from the other legs causing a problem?

That's one....now I remember.....Eagle....thanks.

I am wondering if there isn't a small leak across the the second or third capacitor in (btw, the schematic only shows three stages, when I put together 4). Would that cause this? The fellow that showed me this told me that, if it didn't start sequencing on power on, short the first capacitor for a short period of time, and it will work. Well, this is doing an alternate flashing at 9v dc, but doesn't at all at the prescribed 5v *shrug*

Oh sorry, to answer your questions, no the transistors Emitters and the diode are all grounded on the same plane. It is the cap and diode that help with the cascading, as I understand the circuit.