0

Apologies in advance for this probably-grade-school-level question, but, of all the silly things, I want to make a light-up card that turns off after about 30 to 45 seconds. Fading would be a bonus but auto-off is the main goal. It's for a card (using paper cardstock), so the idea is for it to not be left on for too long...

I have looked all over the internet; I found example circuits using 9V and even 12V, neither of which can be used for a card. The thing is that I don't know enough to figure out how to convert the (v-battery-example to 3V, or even whether the example can be converted to a 3V power source.

I'll try to paste the 9V example below; otherwise, the closest I've come is LMC555 Fading led problem

Again, I know this is probably simple to the point of being stupid for any EE, but for all of the things I've studied and done well in, electronics (and math) were never among those...

Thank You! ...this is the 9V example I was hoping to convert to 3V: enter image description here

BirdsInMyBrain
  • 23
  • 5
  • 5
    30 s starting when? – Reinderien Oct 25 '23 at 23:50
  • 1
    run the LED from a capacitor through a resistor, hook the switch/button directly to the cap. You'll probably need a red LED to have any amount of on-time with 3v. – dandavis Oct 26 '23 at 00:47
  • 1
    that seems to be intended as seems to be some sort of led fader blinking circuit. – Jasen Слава Україні Oct 26 '23 at 03:08
  • simple as in cheap? If so you can get that with a single extra part (a tiny microcontroller). If you want to do it analog, you need at least an RC for timing and a transistor – tobalt Oct 26 '23 at 05:08
  • Does this answer your question? Is the NE555 the IC I need, and if not, what do I replace it with? – Marcus Müller Oct 26 '23 at 20:19
  • Hello, Reinderien - 30 sec starting from a button press. I've seen cards that light using very small buttons, and turn off after a few seconds. – BirdsInMyBrain Oct 27 '23 at 00:42
  • Hello tobalt - Well, for something this small, the parts are already pretty cheap ;) I also have most of the items, and the LMC555 (or similar "C" prefix 555s) and so on are not expensive. So I've looked into the pricing and it's not a problem. – BirdsInMyBrain Oct 27 '23 at 00:48
  • Hello, Marcus Müller - I have to say both yes and no ad to its applicability. I'm looking at the new chips that use very little power. All of the examples I've found online are older and deal with 9V battery voltage sources, or an unidentified "5V" source.

    I know that newer timers use much less power, but I don't have the skill to re-calculate the example so I can get the correct parts...

     – BirdsInMyBrain Oct 27 '23 at 01:05

5 Answers5

7

schematic

simulate this circuit – Schematic created using CircuitLab

Maybe something like this:

It needs a dual-throw switch but should use very little power in the off state.

winny
  • 14,707
  • 6
  • 46
  • 66
Jasen Слава Україні
  • 33,738
  • 1
  • 33
  • 70
4

You basically already have the answer.
The only problem is the old style LM555, forget about that.

Use a CMOS TLC555, it is spec'd to work down to 2 V.

Better yet, get a TLC551, that version works at 1 V.

The older post with the LMC555 claims to have had problems. But there seems to have been other issues with that. You may not be able to do a whole lot with only 3 V, especially if it is just a small coin battery.

The output of the TLC555 and TLC551 claim to be capable of 10/100 mA source/sink, so you may not need the extra transistor on the output. You may be able to drive the LED with as low as 2 mA if you use certain low current LEDs.

The TLC551 datasheet also says that the higher input resistances allow for the use of smaller value timing caps. Including physically large caps inside a thin greeting card would not be so practical.

winny
  • 14,707
  • 6
  • 46
  • 66
Nedd
  • 10,391
  • 17
  • 24
  • Hello, Nedd, Yes, the "C" 555 ones are what I was looking at. I had missed the TLC551, so thank you for that reference; the low-power LEDs are also interesting for future reference. Thanks! – BirdsInMyBrain Oct 27 '23 at 02:16
2

schematic

simulate this circuit – Schematic created using CircuitLab C1 will fade out the LED according to its capacitance. The diode D4 prevents reverse current through the led. SW3 will either connect or discharge C1.

EDITUM

This second circuit is more efficient. I tested with 10uF. the first circuit gave +- 3 seconds. The second circuit 6 or 7 seconds. Theorically, 100uF should give something between 30 seconds and one minute, including fading. The 100KOhms resistor is also important but requires a voltage of at least 5V. Higher voltage gives more light but not a lot of more time. The LED is a regular 3V SMD green LED, 0603 footprint. It's actuated with a single momentary switch.

schematic

simulate this circuit

SECOND EDIT This third circuit is compatible with 3V supply, which can be inserted in a card. Ideally it should be a 1.5V to use a single button battery. 3V requires two batteries. 1.5V won't be able to turn on the Mosfet reliably.

schematic

simulate this circuit

Fredled
  • 2,310
  • 1
  • 10
  • 19
  • seems like a good start but the circuit will need to be rearranged make it work more than once.. – Jasen Слава Україні Oct 26 '23 at 23:54
  • Hello Fredled - That looks almost too good to be true! ;)
    Is the switch a "momentary switch"...?
    Thanks!     – BirdsInMyBrain Oct 27 '23 at 02:23
  • And you know what they say about things that seem too good to be true. For example to get 30 seconds from this circuit you'd need a huge cap value, (and a discharge resistor.) – Nedd Oct 27 '23 at 12:43
  • 1
    Of course it's a start and a discharge resistor may be needed. It still has to be tested. The switch connects either one or the other terminal. It can be permanent or momentary. In both cases, the contact should be closed long enough to charge and discharge the capacitor. I don't think the cap has to be that huge if you are using a 3V SMD LED, preferably green, If your goal is to turn on an indicator LED. If you need to turn on power LED it's another story. – Fredled Oct 27 '23 at 17:20
  • 1
    As with all ideas there is always something useful within. However in this case the OP wanted a 30 second LED blink. Anyone care to calculate the cap size to accomplish that from the above simplified circuit? Oh, and the battery is a 3v type that can fit in a card. But there is a way to make use of the basic idea, let me try a little magic, more later. – Nedd Oct 27 '23 at 17:59
  • Thanks Nedd and Fredled! - You've given me some new ideas for composing searches to try to find an example circuit. Thanks!

    I know it's do-able because I actually had a couple items some years ago that used multiple LEDs flashing in sequence before turning off, and one that only stayed on for some number of seconds. I unfortunately no longer have them to take apart and copy.

    Since I lack the skills to know how to recalculate the values in the examples I've seen, I'll just go with a basic multi-LED setup and rely on the recipient to turn the it off.

    Thank you for trying!

     – BirdsInMyBrain Oct 28 '23 at 04:46
  • What is the operating current of that LED in the second circuit? – Nedd Oct 29 '23 at 23:08
  • Nedd It peaks at 0.12mA! The circuit has the advantage of being extremely low consumption as well as extremely simple. (Emphasis on the word "extreme"). During the time I considered ON, the LED gave enough light to be clearly seen in normal indoor condition. If you rise the voltage to 24V or 36V, the LED will be brighter during the same time. But will dim quickly after that time. – Fredled Oct 30 '23 at 16:55
  • @Fredled - Your circuit 3 looks quite good provided you can find a reliable MOSFET that works well in the 2-3 V range. For a further improvement try C1=10uf with a 9.1M ohm or so in parallel (for just a bit of a controlled discharge), then add a 47 ohm in the source line (for the slow fade out), finally R2 could really be 1k or less depending on the LED efficiency. R1 in the gate line really isn't doing much at all. – Nedd Oct 31 '23 at 09:22
  • Nedd Thanks for your comments. Most Mosfets will work in the 2-3V range according that you use little current, in the mA (for currents of several amperes it's another story). I didn't test the circuit so I can't say if 1 or 10uF would be better (thought some formula exists). Mosfet gates use very little current. A resistor in parallel to the cap is perhaps needed to discharge, elde the Mosfet could very well stays on for ever or turns on unpredictably. Rather 1M than 9.1M. – Fredled Oct 31 '23 at 15:41
  • R1 is absolutely essential to protect the Mosfet gate. Without it the Mosfet can be damaged from the effect of the switch. I don't understand your suggestion about the 47 ohm "in the source line". Could you develop? – Fredled Oct 31 '23 at 15:43
1

Below are two circuits that build on some of the previously posted ideas but also satisfies most of the parameters of the original request.

The first circuit takes advantage of the slow charging of a cap and the high gain of a PNP Darlington pair. When S1 closes C1 slowly begins charging via R1, R6, and R3. The current coming from R3 initially turns on Q1 & Q2. An amplified current then flows through the collectors of Q1 & Q2 turning on Q3 which lights the LED, (D1 is a low current Red LED that operates at 2 ma). As the voltage on C1 continues to rise it reaches a point where the Q1 Q2 transistors begin to cut off. With the very low charging current the full cut off point for the transistors as well as the LED comes about very slowly.

The timing in this circuit will be susceptible to the uncontrolled gain and other parameter of the Darlington pair so there would likely be a good amount of variability from that, in addition C1 will likely have a 10% tolerance or greater, so a potentiometer (R6) can be used for adjustments to make up for these variations. R1 and R6 could actually be a single potentiometer. The 1 ohm resistor R7 is not required and is only included here to graph the LED current.

The third graph (in each group) shows the voltage across the 1 ohm resistor (and is a 1:1 equivalent for showing the LED current).

When S1 re-opens the circuit needs time to discharge, if S1 is closed again before C1 is fully discharge the timing will be reduced.

Other than the battery if all components were SMT types the whole circuit might only take up an area about the size of a 3 V lithium coin cell.

enter image description here

enter image description here

.
.

The second circuit operates a little differently than the first and makes use of a momentary NO push button (S1). Note that in some greeting cards a normally open switch will have a thin insulator between two finger contacts, opening the card slides the insulator out allowing the contact. To simulate a push button it may be possible to have a only a small hole in the thin insulator positioned exactly where the contact points pass. Then as the card is opened the switch briefly closes but is then re-opened as the card opens further.

When this circuit receives the brief switch closure C1 is quickly charged. When the switch re-opens C1 begins to slowly discharge via R1, R2, and R5. The small current through R2 & R5 turns on Q1 & Q2 (in a Darlington arrangement), Q2's emitter current then turns on Q3 that lights the (low current Red) LED. As in the first circuit the 1 ohm resistor is included only to graph the LED current and the pot allows for potential adjustment. Also as in the first circuit C1 will need to fully discharge before the same timing can be realized on the next card opening. (Actually with the switch described above the LED will flash once again as the card is closed).
.
In both circuits the use of very small currents allow the use of a smaller cap values and obviously a smaller body size.

enter image description here

enter image description here

.
.

Nedd
  • 10,391
  • 17
  • 24
  • 1
    Nice circuits. It has the advantage to be more predictable and allow the use of power LED of any size. I like the POT adjust. But it's not "extremely simple". I think a single Mosfet instead of 3 transistors and no POT would be already much more simple. And it would requires an even smaller capacitor. – Fredled Oct 30 '23 at 17:09
  • Wow, Nedd - I appreciate your work, but that's beyond my comprehension level... I also want to put a few LEDs in a greeting card, so one dim red LED isn't going to be enough. I know how to run several LEDs off a button battery or two, and I have a couple of switches in mind, including an inverse momentary switch. I was hoping to have an auto-off but it's apparently not do-able. So I'll adapt my design ;) Thank you for trying, though! – BirdsInMyBrain Nov 04 '23 at 21:22
  • @BirdInMyBrain - If you use the low current LED I linked to further up it wouldn't be dim at all. Also, if you really wanted a few LEDs going on a small card the best way would be to use a very small 6 or 8 pin SMT micro-controller. The circuit would be ultra simple, just the code would be a little complex, but even that step can be easily over come by searching popular forums that likely have code that can just be copied. You can even prototype a small design with a small USB board like this: https://electropeak.com/digispark-usb-board – Nedd Nov 05 '23 at 00:22
  • @Nedd - Apologies for the delayed reply - I've been trying to learn more. I had a bit of basic electronics when I took physics, but that was 50 years ago, so my learning curve is going a bit more slowly these days ;)
    • That small board looks interesting - and I realized that I can fit two batteries and go with 6 V, so this little board could work. I'll need to take some time to delve into that and see what it can do,

    Thanks for putting me on to that!

     – BirdsInMyBrain Nov 09 '23 at 05:36
  • @BirdInMyBrain - I've used a number of those small boards, the USB interface makes it very simple to get a new design started. There are also many look-a-like copies available too. If you get a design working with the proto-board it is easy to convert over to a single chip design. – Nedd Nov 10 '23 at 00:23
0

For what it's worth, I just had either a brainstorm, or a forehead-smacking "DOH!" moment, LOL!

Since my goal was to have a "twinkling stars" effect in holiday cards, and I was going to put die-cut vellum "stars" over the lights to soften them anyway, I just realized that I can use some 3mm candle-flicker LEDs plus an optical fiber setup to transmit the light...

It's difficult to describe the structure of the card itself because it's an early image in my mind, but I have used fiber to add "flames" to a fireplace that was far too tiny to even hold three or even two 3mm LEDs, so I know it works.

It barely qualifies as "electronics" at all, but I know it works and especially, I don't have to buy anything additional because I already have all of the components... It also might be something fun for younger children to try ;)

Thank you very much to everyone for all the good ideas you've shared; I'm not going to abandon the chasing and other LED effects, just because they're so interesting!, but given as close as it is to the holidays, plus all of the other gift projects I have in the works, this solves the immediate goal ;)

BirdsInMyBrain
  • 23
  • 5