555 timers MUST have:
1) An 0.1uF (100nF) capacitor across the Vcc and GND terminals.
2) A 1uF or larger cap across the Vcc and GND terminals.

These are not optional. If they are omitted, you will have problems.
I see that you have a 100nF cap, but no 1uF or larger cap.

 

Can you post a picture of your board, both sides.

John

The actual board itself hasn't yet been drilled so i've made do with stripboard for the time being. I won't have access to a drill press until Wednesday at the earliest so I got a bit impatient.

But i'm absolutely positive i've got the wiring done correctly on the stripboard. The motor/LED section of the circuit is working fine as expected but the servo section is still as problematic as it was on the breadboard. I spent three hours meticulously laying out the stripboard and ensuring that everything was wired up correctly but still no dice. I'm starting to think that there's something inherently wrong with the design (Particularly resistor values). Either that or i'm missing something exceptionally obvious.

 

555 timers MUST have:
1) An 0.1uF (100nF) capacitor across the Vcc and GND terminals.
2) A 1uF or larger cap across the Vcc and GND terminals.

These are not optional. If they are omitted, you will have problems.
I see that you have a 100nF cap, but no 1uF or larger cap.

Decoupling caps?

 

@abc27
I have built and tested that circuit at least three times. It works. The 0.1 cap is needed and is shown on the PCB I drew for you. I did not have a 1 ufd in any of the designs I used. I use an HP/Agilent 6330A power supply. Besides, absence of the 1 ufd cap, which is suggested by SgtWookie, would likely not lead to the symptoms you are seeing.

At this point, you need to review your construction and provide pictures. The circuit works!

John

 

http://forum.allaboutcircuits.com/attachment.php?attachmentid=27665&stc=1&d=1298162246


There's the board. I've annotated strip breaks with a red cross and strip bridges with a green line.

All solder joints are perfect and there is no unintentional bridging of any points. I took a photo of the solder side of the board but it's simply unusable thanks to the light reflecting from the solder.

 

Is that topmost capacitor a 222J or 223J? The wire is covering the marking.

John

 

Is that topmost capacitor a 222J or 223J? The wire is covering the marking.

John

Neither. It's marked 22nJ100

 

The caps certainly are a loooong way from the 555 timer.

The 100nF cap should be practically on top of the timer, with leads as short as you can make them and still connect to the Vcc/GND leads.

 

I am still concerned about that 22 nF capacitor. None of my capacitors are marked like that, nor could I find examples in common places like educypedia (http://www.educypedia.be/electronics/datacomponent.htm ). Do you have a capacitor marked "223" or variations of that like 223J and 223K, how about 0.022, etc? Do you have any way to measure it?

John

 

Oriental "green caps" are marked 223. European Philips or Siemens (or whatever their combined company is called today EPCOS?) metalized poly film capacitors are marked 22n. I have some 100nF ones that are marked "u100".

Here is a photo of some EPCOS capacitors with many marked "10nJ63" which is 0.01uF (or 10nF), 5%, 63V.

 

Interesting information. Thanks. I haven't bought TH caps in quite awhile. Most of mine came form DigiKey and are the regular chicklet style.

John

 

Would moving the capacitor nearer to the 555 have any tangible effect? On the breadboard it was very close to the 555 and I still had the same issue.

I've just done some calculations using the formula for an astable 555 timer and according to that my duty cycle is approximately 50% (As it should be) and the length of the pulse is 1677-1762 seconds. Surely that can't be right?

 

I've just done some calculations using the formula for an astable 555 timer and according to that my duty cycle is approximately 50% (As it should be) and the length of the pulse is 1677-1762 seconds. Surely that can't be right?

Where are you going with this? The schematic you posted in #56 doesn't give values anywhere near what you have just calculated.

It works for controlling a servo(at 5V).

If you want something to do, clean up your breadboard construction. Get rid of the rat's nest of wires, so we and you can see what's attached to what.

I built that circuit three times, including only a minor modification that would not affect the pulse with or frequency. On my breadboard, a decoupling capacitor, 1 ufd capacitor, nor capacitor to pin 5 was necessary. It ran without any of those additions.

The symptom you describe is not due to a little bit of noise or jitter. Your timing is most likely screwed up by an error in construction.

1) What voltage are you running at? You have made various statements related to that. Either 4.8V (4 cells) or 6V (5 cells) will work. Do not use 9V or you will likely fry the servo.
2) Have you made the PCB I designed for you? At this point, that may be easier than untangling the mess of wires you have.
3) Untangle the wires. Use short lengths. Use a similar layout to what I did for the PCB.
4) Finally, although I think the most likely problem is a wiring error, we cannot rule out that you have permanently damaged either the servo or the 555 chip. Do you have any way to test the servo? Do you have any new 555's? Before throwing new parts after old, get your circuit cleaned up.

John

 

Where are you going with this? The schematic you posted in #56 doesn't give values anywhere near what you have just calculated.

I just used a formula for calculating the values. I dunno, I just can't find out where i'm making the error.

I built that circuit three times, including only a minor modification that would not affect the pulse with or frequency. On my breadboard, a decoupling capacitor, 1 ufd capacitor, nor capacitor to pin 5 was necessary. It ran without any of those additions.

I don't doubt the circuit. The problem is that i've tried many different layouts and I have had no success with any of them. Surely I couldn't have made the exact same error so many times.

The symptom you describe is not due to a little bit of noise or jitter. Your timing is most likely screwed up by an error in construction.

1) What voltage are you running at? You have made various statements related to that. Either 4.8V (4 cells) or 6V (5 cells) will work. Do not use 9V or you will likely fry the servo.
2) Have you made the PCB I designed for you? At this point, that may be easier than untangling the mess of wires you have.
3) Untangle the wires. Use short lengths. Use a similar layout to what I did for the PCB.
4) Finally, although I think the most likely problem is a wiring error, we cannot rule out that you have permanently damaged either the servo or the 555 chip. Do you have any way to test the servo? Do you have any new 555's? Before throwing new parts after old, get your circuit cleaned up.

John

1) 6V. That's what i've been using since the start.
2) I have and it has been etched. The problem is that I won't have access to a suitable drill press for a good few days.
3) I can try but I have an injured thumb that's preventing me from using extremely short cable lengths. I can try cutting the very long wires through the middle and resoldering them to shorten them.
4) I've tried two different servos and they both behave in exactly the same way. I've also tried many different 555 chips so i'm pretty sure my parts are ok.

Thanks,
Michael

 

You have a solderless breadboard. Use that and 22 awg solid wire to make connections. There are not that many. Make it neat. If you don't have 22 awg insulated wire, try excess leads from resistors. Save the resistors for your pcb.

John

 

I'll try again with the breadboard I suppose. I'm working on shortening wire lengths on the stripboard now though. Maybe it will have a positive effect.

 

You don't show how you've connected up your potentiometer in your actual circuit.

In your original schematic, you connected the "ends" of the pot between the 56k resistor R1 and the three 3.3 MEG resistors, which would give a fixed 100k resistance rather than a variable 0k-100k; which would keep your output PWM duty cycle the same all the time.

You should post an image of your potentiometer and connections.

If the pot has three pins in a row, it's typical that the "wiper" is the middle connection. You'll want to use the wiper and one of the ends. If you have a multimeter, measure the resistance between the pins; if it stays fixed while you rotate the pot shaft, then you are across the ends of the pot rather than the wiper and an end.

 

Don't worry. I am not that bad to wire up a potentiometer wrong. I've one end soldered to the end pin and another to the wiper.

As for the stripboard i've shortened all the wires. Somehow i've introduced a short circuit as the circuit is completely dead at the moment.

Edit: Scratch that, it just appears that the battery's flat.

 

Well i've finished the PCB today. I managed to drill the holes by using a sewing needle and a cordless drill. Surprisingly good results and nowhere near as flimsy as a 1mm drill bit.


That's all the good news though. It's completely non-functional, not even the LEDs are lighting up. The servo just twitches when I connect the battery but nothing more than that, it doesn't even go fully clockwise like it used to.

I've spent the past three hours going over the board and I can't find any error that would be causing something like this. Would a blown LED cause the entire circuit to fail like this for example? Even the DC motor isn't working.

Edit: I tested everything with freshly charged batteries that I confirmed as being working with another DC motor off-PCB.

 

This is what the board from the top should look like:

You have the diode band pointing to pin 8 and the servo pins on the left, among other things. In the layout I sent you, the diode band is near pin 5 and the servo pinouts are on the right.

If you used that layout, you have a mirroring problem, so my first guess is that the board and probably the 555 are toast.

Remember, the board image is looking down on the board from the top. I have no idea what you did for the image transfer, so I can't advise you on whether to use the mirror print function or not.

John