Yes, this is due to lack of available memory registers, right? Are you sure that PIC10F200 is able to deal with the task at hand to begin with? How can you tell, other the from experience? Is there some sort of metric you are going by? To how many separate LEDs can the PIC10F200 send instructions to?

10F200 has 3 outputs for LED control and 1 input. If you find the 10F202 is the same price, it is the same but with more program and data memory so may as well get that instead.
If you need more than 3 outputs the 12F508/509 has 5 outputs and 1 input and may not be much more expensive.

 

10F200 has 3 outputs for LED control and 1 input. If you find the 10F202 is the same price, it is the same but with more program and data memory so may as well get that instead.
If you need more than 3 outputs the 12F508/509 has 5 outputs and 1 input and may not be much more expensive.

Rock on Markd77, rock on!

Thank you!

 

I got this really bright idea....haha (thinking outside the box):

Glow sticks - probably for less than $0.30/each so in the worst case scenario...if I run out of time I think I can strap a pair of these to the sign and make it work for a day. I think I will not have any diffusion issues with these, so that is a plus. The negative of course comes from high price per hour of use ($0.60). If I can get that figure down to say half that much (larger order may help) I might get what I want.

I asked the seller of those sticks about how long they will stay at their brightest - I'll let everyone know what he says.

I think this may tie me over until I can create the circuit we have been brainstorming above and program it.

 

I thought I might mention that Texas Instruments has come out with a low-cost line of MSP430 microcontrollers. The first few models are out now. In low quantities they're around $0.75 and in 10,000s they're around 25 to 40 cents.

Also, TI just came out with the LaunchPad. It's a starter development kit for $4.30 including shipping. Only problem is they sold so fast, they've been rushing to fill the backorders.

Anyway, keep the MSP430 in mind as a low-cost option.

================

Don't forget bypass capacitor to the microcontroller. Probably a good idea even if it adds 3 cents to the cost. It's really important for pulsing on the LEDs, to provide the current.

================

For a power source, directly solder a coin cell to the PCB?

================

Using 3 multicolor LEDs with the anti-parallel arrangement for different colors, you may realize that you can theoretically get 7 different colors plus off using three microcontroller pins. I am sure you can get more but I haven't played around with it much.

================

There are some 11 cent (yes $0.11) dual color LEDs in qty of 3,000... check Newark at this link. (If link doesn't work, I basically searched for multicolor LEDs and sorted by price.) Most of these are 20 mA and around 150 mcd. I don't know if that's bright enough for you. Sure can see it at night.

=================


Hope these ideas help.

=================

It's funny how after all this, it comes down to glow sticks!! Anyway, I am having fun thinking about it.

 

I found some 5-6" glowsticks at home depot. rated for TWELVE (~$2USD) hours and pretty bright. I quickly made a rough draft of the sign and snapped few pics in various lighting conditions.

My camera on the iPhone 3G sucks big time - so pictured did not come out very well (and I will not post them because of it), but I was very happy with the results. I snapped some pictures from up close, 20 feet away and 10 feet away. I

I snapped pictured in the dark and dimply lit room (about as much light as I would expect in the places where I plan to place them).

Results are surprising:

Signs could be easily read with one glowstick, especially the green one at over 30+ feet away. Orange glowstick is a bit less bright so maybe at 25+ feet.



I was looking for more glowsticks and found that I could score them for less then $0.25 off eBay if I buy in bulk (300+). I will order a sample tomorrow and hope the light lasts out about as long as these SHOULD (I will report on the brightness level as the glowsticks burn. Under bright light the glowsticks show their color outline, but that is it.

=)

 

I thought I might mention that Texas Instruments has come out with a low-cost line of MSP430 microcontrollers. The first few models are out now. In low quantities they're around $0.75 and in 10,000s they're around 25 to 40 cents.

Also, TI just came out with the LaunchPad. It's a starter development kit for $4.30 including shipping. Only problem is they sold so fast, they've been rushing to fill the backorders.

Anyway, keep the MSP430 in mind as a low-cost option.

Nice! I just ordered EasyPic6, but I'll grab this as well. I ordered PIC first because I didn't want to rely on backorders coming through in time.
================

Don't forget bypass capacitor to the microcontroller. Probably a good idea even if it adds 3 cents to the cost. It's really important for pulsing on the LEDs, to provide the current.

HOT tip! Thanks!!!
================

For a power source, directly solder a coin cell to the PCB?

I am limited by the power capacity of coincells:http://en.wikipedia.org/wiki/List_of_battery_sizes#Button_.2F_coin_.2F_watch_.2F_micro_.2F_miniature

CR2477 are no less than a buck a piece.
CR2450 are pretty affordable (less than $0.20USD in bulk)

I do not know if this is the full list, or I have more efficient choices that I do not know about. I also do not know what my standard discharge current should be. I was going for AAA because I thought they have more juice to justify their price, but then again - I do not know that much ether.

I will have some issues with soldering to batteries. Unless I can spotweld or get batts with tabs I will have to get batt holders which ads $$$ overhead.



================

Using 3 multicolor LEDs with the anti-parallel arrangement for different colors, you may realize that you can theoretically get 7 different colors plus off using three microcontroller pins. I am sure you can get more but I haven't played around with it much.

- can you explain this again please. I
did not understand this well.
================

There are some 11 cent (yes $0.11) dual color LEDs in qty of 3,000... check Newark at this link. (If link doesn't work, I basically searched for multicolor LEDs and sorted by price.) Most of these are 20 mA and around 150 mcd. I don't know if that's bright enough for you. Sure can see it at night.

I should order few of these to test them out.
What do you think of these: http://cgi.ebay.com/100-Pcs-5mm-Red...itu=UCC&otn=5&ps=63&clkid=6694884020432656703
or these:
http://cgi.ebay.com/10-X-5mm-Red-Gr...tem&pt=LH_DefaultDomain_0&hash=item4cf017460b

=================

Hope these ideas help.

They helped a lot! I hope you chime in more often on my project's thread!
=================

It's funny how after all this, it comes down to glow sticks!! Anyway, I am having fun thinking about it.

K.I.S.S. hehe

 

What if I use UV LEDs and florescent paint or even glow in the dark paint?

 

Glow in the dark paint will also florescent usually. If you can get the LEDs cheap it will work, and if you turn the LEDs on/off it will look like parts of the picture is turning on/off.

 

I dont think the glowsticks held up for 12 hours...

 

Glow in the dark paint will also florescent usually. If you can get the LEDs cheap it will work, and if you turn the LEDs on/off it will look like parts of the picture is turning on/off.

I will probably have to front light them, right?

 

Do not know, if the glow paint is thin enough I suspect the back will shine through. It's a theory at least.

U/V LED are something I haven't had direct experience with, though I've played with florescent black lights now and again.

 

Yes. It will teach you much quicker. It has built in resistors and leds that will save you time developing and wiring. It will hold your hand through the design process.


There is also a very active forum for the EasyPICs that offer much in the way of help. The MikroC, BASIC, and PASCAL forums also offer tons of help in way of programming.

Its up to you.

I got my EasyPic6 yesterday and a couple of linear power supplies 0-18v


I'm putting an order for some resistors, batteries, and battery holders as soon as I figure out the power requirements.

Should I get the PICs now? like 10 of them or something to start with?

Can I use the same program on all PICs? or will I have to modify it for each PIC? I also have the 2 kilobyte hex limit on the compiler that came with EasyPic6 unless I upgrade, should I? Or will be OK with 2KB?


I'm waiting for some 5mm@5000mcd slow flashing RBG LEDs: http://www.youtube.com/watch?v=epoVPB6wOuo

 

I was thinking of implementing a system that turns on the sign and then turns it off. I need the sign to operate from 8pm to 3am each night. At first I was thinking of using a dark-detecting circuit, but I figured that I can save even more battery life if I can turn it off while it is still dark (3am). That is at least 3-4+ (think winter) hours saved each night if I can implement a timer circuit. But how? Any ideas?

 

Will I need a separate crystal for pic10F200? or pic12f508/509?

 

A 4 hour timer is not too hard, especially if split second precision is not required. There are chips like the 4060 that can do it. Microcontrollers can also do this well.

So you want a sign that lights up when it gets dark, and turns off after 4 hours?

 

Not precisely:

Situation 1: Turn on when it gets dark and work until 3am

Situation 2: Turn on when it is 8pm and work until 3am


The idea here is to only broadcast (light) the sign's content when the certain audience are around. I estimate that depending on the sign's placement location this could ether be when it starts to get dark or around 8pm. The question is how to track time. Or is there a more logical solution that I am missing?

 

A pic will do this nicely, otherwise you're going to need the equivalent of a digital alarm clock.

 

Thanks Bill,

Can you give me an idea how it will do this? I mean how will the pic know what time it is? I'm guessing that this will mean that it will always draw power from the battery no matter if the LEDs are on or off.

This brings me to my next question:will the power savings accomplished with a timer program (considering constant power draw of the PIC) will justify the [low level?] power draw by the PIC over few days time over a simple dark-detecting circuit that turns on the system when it gets dark and turns it off when it gets light out?


Design 1: PIC powered timer will draw power at all times but saves power(3am cutoff) during its most intensive use (lighting LEDs)
Design 2: Never draws power until its most intensive use, but draws it longer than in the Design1 above (no 3am cutoff)

 

How far away will this sign be from a mains outlet?

If it is only a short distance you could use a mains timer to control a relay to switch the sign on and off.

 

Hello,

I merged the threads on request of the OP:

Hello, I've made a mistake and made a separate post [ http://forum.allaboutcircuits.com/sh...ad.php?t=41665 ] instead of a reply to an existing post [ http://forum.allaboutcircuits.com/sh...ad.php?t=40926 ]. Could you please merge them for me?

Thanks!

Bertus