I am in need of setting up a timing circuit to run for about 15 minutes. This circuit will be powered by USB. It will turn on a small motor that under load will draw less than 100mA and will run with as low as 3v. The main reason for posting in here, is the values of the resistor and capacitor. Reading at numerous site have basically scrambled my brain to thinking that maybe the circuit will not operate as wanted.

This timer does not have to be accurate to the second.

1. This circuit must have a switch. What type of switch should I use?

2. One site informed me that electrolytic capacitors can vary as much as +/-30%. Will I just have to plug in different resistor/capacitor combinations to get close to 15 minutes?

3. Are the values too high? How would I go about setting up this circuit to work properly then?

I knew basically nothing about electronics before diving into this. My project as it sits now will be just a push button on/off controlled by the user. With life being so hectic, knowing that it will shut itself down is one less thing to remember.

Find attached my first attempt treading into deep water.

 

The basic timer is basically sound, but please read the 555 Monostable article to set up the input switch, yours has some problems.

You also MUST have a diode across the motor, otherwise the 555 wil be blown.

This is not a timer, but it demonstrates the diode.

 

Thanks for the reply. I will have to start by reading that page again tomorrow. As I said, I am new to this and being so late, it read as blah blah blah to me. May as well been in a different language. Probably will make more sense with a clear head.

I was also thinking that it needed a diode, so have inserted one on my new attachment. Though I thought this, I have no idea as to what size/type of diode to use. Someone also suggested diodes to prevent back feed to the computer. Not even sure where to put these either.

Kenny

 

Use any of the 1N400x (where x = 1,2,3,4,5,6,7) diodes. The number represents the Max voltage across the diode before breakthrough. With 3V motor I doubt that it will ever create 100V PIV, but you never know. Rather be safe than sorry. I personally only have 1N4007 diodes in my spares box.

I would also use a transistor to drive the motor vs. directly wired to the 555 as you show. Choose a transistor that can handle the wattage consumed by the motor and greater, for example P=VI where V = 3V and I = 100mA. Thus P = 0.3Watt. The transistor should be a 0.5 watt device or bigger, and you might also consider that it might generate some heat, hence the need for a heat sink should also be factored in.

 

As shown you are turning the entire power supply on/off, you need to do this with pin as shown in my article.

 

The main reason for posting in here, is the values of the resistor and capacitor.

Too right - long RC time constants are problematic. To get a tau of 15 minutes with a 900 kΩ resistor (say) requires a 1000 μF cap, which usually have a tolerance of -20/+80 % and will be leaky besides. A 555 will tolerate timing resistors up to a few MΩ, at the expense of reduced noise immunity, but even so we're still in electrolytic country. Supercaps are low-leakage, but still have the same wide tolerance spread, and have a working voltage of 5.5 V or less.

Normally long time constants are done with a more accurate higher frequency clock and a counter. A few minutes is possible with RC techniques, but it's incredibly imprecise.

 

Don't forget the CMOS 555, it can go even higher resistances. 15 Minutes though, IMO, isn't too long for a 555, as long as you don't need tight accuracy specs. The base simplicity can be worth it.

I'll put up what I think the schematic should look like to save time...

Figure 1 gives the motor the full power supply voltage, while figure 2 gives the motor the power supply minus 1.8V. If the power supply is 5VDC then the motor sees 3.2VDC.

The duration will not be that accurate, but it will repeat with fair accuracy. In other words, in answer to your question #2, you may have to tweak the values of RC to get the duration you want, but once your there it will work within 20 seconds or so of the setting.

Betcha this is a USB powered fan.

 

Too right - long RC time constants are problematic. To get a tau of 15 minutes with a 900 kΩ resistor (say) requires a 1000 μF cap, which usually have a tolerance of -20/+80 % and will be leaky besides. A 555 will tolerate timing resistors up to a few MΩ, at the expense of reduced noise immunity, but even so we're still in electrolytic country. Supercaps are low-leakage, but still have the same wide tolerance spread, and have a working voltage of 5.5 V or less.

Normally long time constants are done with a more accurate higher frequency clock and a counter. A few minutes is possible with RC techniques, but it's incredibly imprecise.

Ditto to all of this. Years ago Maxim produced a specialized chip that Incorporated a 555 and digital divider. It made timing to hours possible using much smaller RC values. I can't remember the chip model though.

 

A long period timer circuit using a 4060: http://uk.geocities.com/ronj_1217/al1/24t.html

ken

 

Just keep in mind Gents, the OP wants simple.

 

Just keep in mind Gents, the OP wants simple.

Just curious. Is this a class project?

 

That 4060 circuit that KMoffett linked to (and CDRIVE mentioned in Maxim form) would be ideally suited, I'd have thought - the IC is nice and cheap, it can use caps with 10% or 5% tolerance, and it needn't have any high resistor values. I get nervous with resistor values in the megs as bias and leakage currents elsewhere in the circuit start to dominate, plus there are all sorts of leakages from flux residue, dirt and PCB/stripboard substrates. Cheap SRPB (paper) boards are very leaky indeed; FR4 (glass) is much better, but not perfect. Still, the leakiest thing would be the electrolytic, so it's nice not to have to rely on these things.

That said, I've slipped in a 1000 uF cap with a 10 uA current sink to generate a 20-minute ramp in this pigeon light, which I still owe joco some workable values for. Well, there's two at least... I've still about half of 'em yet to work out. I rather naughtily get away with it here as I've a dead band of 1-2V at the bottom of the discharge curve that I don't care about, and this is where the worst of the leakage current effects lie. Sadly I can't use a 4060 here as I need a timed linear ramp, not a timed pulse. Of course it will be a timed linear ramp with -20%/+80% timing tolerance...

There are some clever things it's possible to do with a CMOS 555 timer that aren't possible with the bipolar version, this I know, but beyond that I defer to Bill Marsden's greater expertise. They're a great IC - the most popular in the world, I hear, due to their price, reliability and flexibility, and they're the basis for many an ingenious circuit.

 

Thank you to all who have responded.

This is not a class project just a private one. One good thing I found out today, is the person who wrote up the timing specs dragged their pen on the paper. It only needs to be 5 mins not 15. Oh how life just got easier. The other advantage is accuracy is not required. Looking to hit this between 5-6 mins.

Will have to go with figure #1 though. Since USB specs out at 4.5v and my motor minimum is 3v.

Someone mentioned a transistor. While reading many set ups, I was informed that this circuit can handle up to 200mA with no problem. Since my motor even at start will stay under 100mA, is a transistor really required? Or is this again something I should play around with?

Bill - I will read your article again tonight. As I said, my brain was fried after all the articles I was inputting last night.

Thanks again,

Kenny

 

While the 555 can handle 200ma, it is always a good idea to allow extra specs. That and motors draw a surge current when they start up. The transistor can handle it, but the 555 is pretty close to it's limits. A 555 also has an odd characteristic in that the output (when it is high) is 1.2V less than the power supply, or 3.3V in this case. This is assuming direct drive from the 555 and 4.5V power supply (also near the 555 spec limit).

Not sure I'd recommend this design, but you you can try it out. I'd definately add a fuse (not shown), just in case. Something in the range of ½A.

 

The reason why I would use a transistor is for future applications. What will happen (for example) if someone should replace the motor originally designed for use, with a higher wattage motor? That will fry the 555, whereas if it fries the transistor it's cheaper to replace the transistor than the 555.

Just my thoughts. (That's what I would do )

 

Please reconsider and not use your USB port as a power supply. Unless the PC is an old junker that you don't care about, it's not worth worth the chance of blowing the port.

 

It seems to be the trend. Even devices that should be safe aren't always, such as the iPod that blew my neices port. I had to add a card to get the computer back up.

BTW, thinking about it, use a ¼A fuse for that circuit. It may save you some grief.