How I solved a 555 problem with LED continuously on

[richard1991]

Hi. Today I made a 555 circuit and connected it to work in astable mode.  But, the LED was on continuously.  I could not find a fault in the circuit.  Eventually I thought I would lower the supply voltage form 9v to 6v. And guess what? the LED now flashes.  I would not have expected to face a problem using a 9v supply.  Anyway, just thought I'd mention my experience.  Rich

P.S. This was not in fact the solution. See end of thread. The problem was to do with a tantalum bead timing capacitor.

[xrunner]

Hi. Today I made a 555 circuit and connected it to work in astable mode.  But, the LED was on continuously.

What was the expected frequency of this circuit?

[richard1991]

The expected frequency was about 1.8Hz. It flashes at about the speed now. I've haven't a clue why it works on 6v but not 9v. Unless at 9v oscillation occurs. But I sort of doubt that because I think the circuit is essentially a relaxation oscillator, not an LC type. Rich

[richard1991]

Ohh darn it, after flashing 10 minutes the LED is now continuously on again.  I removed the 6v and then connected the battery again (6v) and the LED now does not flash. Weird. I wonder if I have a connection issue.

Possibly an undervolt issue now. I'll check.

[phliar]

You should post a schematic. I've run 555s up to 15V with no problems.

Which 555 are you using?

[StillTrying]

Go around the 555 and check every pin goes to where it should.
A diagram would be useful.

[richard1991]

I can ramp the voltage up from 0v right up to 15v with a PSU.  But now it's continuously on again, no matter what the voltage when it  is above about 3.6v.  It's now set at 6v. But when I first went from a 9v battery to a 6v one, the LED was flashing.  So, the circuit must be okay. I'm using a 10uF tantalum capacitor for C1 and R1 (rail and pin 7) is 1K and R2  (pin 7 and pins 6 & 2)is 33K. R3 is the LED ballast resistor. Pin 5 has a 10nF capacitor to ground (C2).

Oh, the voltage was 3.6v and just now when I looked at the LED it's flashing.  I ramped up to 6v and it went continuous.  Now, I've gone down to 3.6v and it's flashing again.

Anyone have an idea what is going on? Thanks. It does not always flash at 3.6v.

[Zero999]

I can ramp the voltage up from 0v right up to 15v with a PSU.  But now it's continuously on again, no matter what the voltage when it  is above about 3.6v.  It's now set at 6v. But when I first went from a 9v battery to a 6v one, the LED was flashing.  So, the circuit must be okay. I'm using a 10uF tantalum capacitor for C1 and R1 (rail and pin 7) is 1K and R2  (pin 7 and pins 6 & 2)is 33K. R3 is the LED ballast resistor. Pin 5 has a 10nF capacitor to ground (C2).

Oh, the voltage was 3.6v and just now when I looked at the LED it's flashing.  I ramped up to 6v and it went continuous.  Now, I've gone down to 3.6v and it's flashing again.

Anyone have an idea what is going on? Thanks. It does not always flash at 3.6v.

How is the schematic constructed? Solderless breadboard by any chance?

I suspect it's got nothing to do with the supply voltage. It's probably a loose connection on the capacitor or the actual capacitor itself has gone bad by developing a poor connection internally. Ensure that everything is properly connected.

[StillTrying]

I think the polarity of the tantalum timing cap being the wrong way around might cause that.

If you've got decoupling on the CV pin but not across the supply pins that often makes it less stable.

[richard1991]

Hi. It's the tantalum bead capacitor. It's new, rated at 25v, and was correctly inserted for polarity, but gave unreliable results. I put in a regular electrolytic, and the circuit works fine.

The astable circuit using the 555 has just not worked using a tantalum bead capacitor.

Maybe the voltage across the tantalum capacitor was insufficient, or leakage is to great for the circuit.

Anyway, reducing the supply voltage was NOT the solution. :-)

[StillTrying]

Interesting, at 9V supply the voltage on the tant would have only been 6V. If it's that bad you should be able to detect it's leakage with a DVM on the ohms ranges.

[Zero999]

Hi. It's the tantalum bead capacitor. It's new, rated at 35v, and was correctly inserted for polarity, but gave unreliable results. I put in a regular electrolytic, and the circuit works fine.

The astable circuit using the 555 has just not worked using a tantalum bead capacitor.

Maybe the voltage across the tantalum capacitor was insufficient, or leakage is to great for the circuit.

Anyway, reducing the supply voltage was NOT the solution. :-)

Are you sure the tantalum capacitor was inserted correctly for polarity?

I hope you're aware that the stripe on a tantalum capacitor denotes the positive lead, rather than the negative, as is the case for an aluminium electrolytic capacitor

[alsetalokin4017]

Please post the exact schematic that you are using, including all component values. Not just a verbal description of the circuit! 

[richard1991]

I used the typical circuit for astable mode as shown here. However, please note that in my circuit, C1 is 10uF, C2 is 10nF, R1 is 1k, R2 is 39K, R3 not shown would be LED limiting resistor connected to pin 3, and there was D1, the LED also not shown. As I say, the circuit did not work with a tantalum bead capacitor for C1. And C1 was connected correctly, the positive lead is clearly marked with a + sign. When I inserted an electrolytic for C1, the circuit worked.

The actual capacitor used is as per the picture: 10uF, 25V,  Plus sign near one lead. Where the reflection is, is a circle with what looks to be a 2 or z.  I presume this capacitor has too much leakage for the circuit. I replaced with another of same type, LED flashed about 5 times then on continuously.

[Zero999]

I used the typical circuit for astable mode as shown here. However, please note that in my circuit, C1 is 10uF, C2 is 10nF, R1 is 1k, R2 is 39K, R3 not shown would be LED limiting resistor connected to pin 3, and there was D1, the LED also not shown. As I say, the circuit did not work with a tantalum bead capacitor for C1. And C1 was connected correctly, the positive lead is clearly marked with a + sign. When I inserted an electrolytic for C1, the circuit worked.

The actual capacitor used is as per the picture: 10uF, 25V,  Plus sign near one lead. Where the reflection is, is a circle with what looks to be a 2 or z.  I presume this capacitor has too much leakage for the circuit. I replaced with another of same type, LED flashed about 5 times then on continuously.

Tantalum capacitors generally have a lower leakage current than aluminium electrolytic capacitors.

If the circuit won't work with those tantalum capacitors but works with ordinary aluminium capacitors, then you must have a bad batch of tantalum capacitors. Where did you get them from? I suggest you discard them, as they could cause problems if you used them in other circuits. Tantalum capacitors can fail catastrophically by going short circuit, melting down and burning up which can damage other components and result in fire. In my opinion, it's not worth holding on to a bad batch of tantalum capacitors.

[richard1991]

Of the 20 capacitors I bought 5 work 15 do not. It must be a bad batch. Just noticed that the seller wrote "New Old Stock".

[richard1991]

I put the non working capacitors across a 9v battery and seemed to have reformed them. They now do work in the circuit.

[Zero999]

I put the non working capacitors across a 9v battery and seemed to have reformed them. They now do work in the circuit.

That's interesting. I've heard that aluminium electrolytic capacitors often need to be reformed but didn't know the same can be true for solid tantalum capacitors.

[David Chamberlain]

Just wanted to add a tail of caution regarding tantalum caps. If your starting out in electronics I would recommend avoiding them completely, plenty of other alternatives are available in larger packages per Farad.

The first answer here gives a good overview of the issues.
https://electronics.stackexchange.com/questions/99320/are-tantalum-capacitors-safe-for-use-in-new-designs

and also reflects my own professional fails where, in one instance, we needed a fire extinguisher (no joke) to get the situation under control. First time this happened was at a trade-show where we were showing off a prototype product, very embarrassing. The competition did not hesitate in taking plenty of photographs and posting them online. A hard Lesson learnt.

Here is an example datasheet.
http://www.kemet.com/Lists/ProductCatalog/Attachments/254/KEM_T2005_T491.pdf

Note the "Maximum Allowable Ripple Current" exceed that and sooner or later a runaway thermal reaction of the FIRE kind will ensue. Not cool in your home.

This will be somewhat surprising as most components that short out will blow a fuse, burn out a trace and so on. But these things create a self sustaining chemical reaction akin to thermite that will just not stop.

Am I being dramatic enough? good.

[richard1991]

I'm not exactly a beginner, however I have not dabbled in electronics for a long time. It would be nice if there was a capacitor alternative from say 0.47Uf to 100uF which did not rely on electrolytic action.  Do such components exist? Thanks.

[Zero999]

I know many warn against using tantalum capacitors, I find them very reliable, much more so than aluminium capacitors. I've only recently seen some boards with failed tantalum capacitors on. Very little was damaged, the board still worked, as they were only used for decoupling (often more decoupling capacitance is added than necessary) and were very easy to replace.

It depends on what you want the capacitors for. Multilayer ceramics are commonly available up to 10µF but beware of that the capacitance declines sharply, as the bias voltage increases and film capacitors also exist up to 100µF but are very bulky and expensive. For decoupling and 555 timer circuits, aluminium and tantalum electrolytic capacitors are really your best bet. Another thing you can look at are solid polymer electrolytic capacitors, which are essentially aluminium capacitors with the electrolyte replaced with a solid, conductive polymer, which can't boil off or dry out.