I woke up this morning with a cost reduction in mind: reducing the MOSFETs from two to one, and eliminating another resistor. I redrew the schematic and rebuilt the breadboard to be easier to understand. SW2 is now on the breadboard. See the attachments.

ETA: The LEDs have 1k current limiting resistors that are not visible in the photos, and I am still using 1N4148 diodes instead of the 1N4001 as shown in the schematic.

 

I woke up this morning with a cost reduction in mind: reducing the MOSFETs from two to one, and eliminating another resistor. I redrew the schematic and rebuilt the breadboard to be easier to understand. SW2 is now on the breadboard. See the attachments.

ETA: The LEDs have 1k current limiting resistors that are not visible in the photos, and I am still using 1N4148 diodes instead of the 1N4001 as shown in the schematic.

Seems like our disciplines are not too dissimilar, after all. A good 10% of my own, computer related, problem solution Eureka moments occur in the "downtime" period when I'm NOT in front of the offending equipment. The most memorable being when a clent (and myself) had been all but tearing our hair out over a particular, seemingly irreconcilable, problem following a power surge/outage at his premises. The potential equipment replacement costs wouldn't have left much change from $1m but that was the scenario he was seriously considering. Company technician after technician, brain after brain was charged with solving the problem. I was called in a as a freelance, last ditch attempt, to sort it and spent the bst part of the day, failing to do so.

Unable to erradicate the problem from my mind and equally unable to relax at home, I trotted down, late, to a local bar to shoot some pool and have a beer. I was closely beaten in my first game by a New Zealander who was due to go home, the next day. We got talking casually about time-zone differences...when it hit me. The date in the systems BIOS was American (as in January 2nd 2010 being 01.02.10) but the date in the Operating System was UK (as in January 2nd 2010 being 02.01.10). I could barely sleep, considering my theory/revelation, but arose early, attended at the client's office and had the entire system back up and running, sweet as a nut, by the time he arrived.

Enough waffling, thanks for the update and for being bothered enough to spend still more of your time considering, testing and posting it.

It's truly appreciated (it's still Cyrilic, at this stage, but I'm working on that ).

Thanks again. Will no doubt be bothering you again before long.

Jib T

 

I've created a drawing showing one way to wire your LEDs. I've also redrawn Tracecom's circuit to show how to connect the LEDs to said circuit. I added a 1uF capacitor in the event the power supply is questionable or if another circuit is added to this one.

The values of resistors R5-R69 will depend on the LEDs you're using. I can make a guess, but I'd much prefer to know the make and model of the LEDs you have. If that is not available, do you have the forward voltage and current rating for each LED (each color)?

Regarding your viewing requirements - I strongly suggest using diffused blue and white LEDs, not super/ultra/nuclear bright LEDs. The latter will blind viewers or at least make them see spots for an hour or so if viewing head-on as you've described. If for some reason you cannot replace the super bright ones you have, you can dim them by a) reducing the current to the LEDs, b) sanding the LED lenses, or c) putting a diffuser of some sort over the LEDs.

While reducing the current will help, they'll still be blindingly bright when viewed from head-on, so this won't help much. Sanding the LEDs will be tedious, especially since we're talking about 187 LEDs; it's also difficult to ensure each LED will be sanded exactly the same as the rest, so it's likely some LEDs would appear brighter than others. Using the frosted paint you showed earlier could work. I suggest using it on a piece of acrylic and putting that over the LEDs. I do not recommend spraying the LEDs themselves. You might also try a piece of wax paper over acrylic placed atop the LEDs.

If you're okay getting diffused LEDs, let me know and I'll point you toward some that should work well.

 

I've created a drawing showing one way to wire your LEDs. I've also redrawn Tracecom's circuit to show how to connect the LEDs to said circuit. I added a 1uF capacitor in the event the power supply is questionable or if another circuit is added to this one.

The values of resistors R5-R69 will depend on the LEDs you're using. I can make a guess, but I'd much prefer to know the make and model of the LEDs you have. If that is not available, do you have the forward voltage and current rating for each LED (each color)?

Regarding your viewing requirements - I strongly suggest using diffused blue and white LEDs, not super/ultra/nuclear bright LEDs. The latter will blind viewers or at least make them see spots for an hour or so if viewing head-on as you've described. If for some reason you cannot replace the super bright ones you have, you can dim them by a) reducing the current to the LEDs, b) sanding the LED lenses, or c) putting a diffuser of some sort over the LEDs.

While reducing the current will help, they'll still be blindingly bright when viewed from head-on, so this won't help much. Sanding the LEDs will be tedious, especially since we're talking about 187 LEDs; it's also difficult to ensure each LED will be sanded exactly the same as the rest, so it's likely some LEDs would appear brighter than others. Using the frosted paint you showed earlier could work. I suggest using it on a piece of acrylic and putting that over the LEDs. I do not recommend spraying the LEDs themselves. You might also try a piece of wax paper over acrylic placed atop the LEDs.

If you're okay getting diffused LEDs, let me know and I'll point you toward some that should work well.

Hi again (sorry for the late reply but my son arrived home unexpectedly from his 2 year sabbatical to Australia....he was supposed to have gone straight onto New Zealand....and I've been busy with domestic responsibilities, as a result).

Thanks so much for your continued help, though.

With regard to the LEDs, I bought them from a UK supplier (although I suspect it is simply a wholesaler of Chinese or other far Eastern packs of LEDs, which are forwarded onto the purchaser, such as me). They arrived in standard resealable zip-bags (I bought a mix of colors in packs of 100). No idea of brand etc., althought the supplier lists both the blue and white ones as having a forward value as 3.2 (typical) and 3.8 (maximum), so I selected those colors for ease of the project..

I've looked at the cost of diffused LEDs and it is horrendous, when you consider the quantity I need for this project - relative to non-diffused costs.

Thanks again.

 

I've created a drawing showing one way to wire your LEDs. I've also redrawn Tracecom's circuit to show how to connect the LEDs to said circuit. I added a 1uF capacitor in the event the power supply is questionable or if another circuit is added to this one.

Can I just add that your graphics are the first electrical diagrams I've seen that make some sort of sense to a layman such as myself and should become part of the schools' curriculum, for those studying electronics, in my humble opinion.

Not saying that it's all crystal-clear, quite yet to me, but it's as near as dammit.

Hats off to you. I'm extremely grateful for your help and that of tracecom, without which I am convinced I might have gone nuts.

Clouds are lifting and (although not quite Johnny Nash, yet) things are becoming clearer.

Thank (insert deity of choice) I found this forum!!

 

I've looked at the cost of diffused LEDs and it is horrendous, when you consider the quantity I need for this project - relative to non-diffused costs.

How much is too much? Check out these: blue and white. You require 40 white LEDs ($3.50 USD for 50) and 147 blue LEDs ($7.40 USD for 200).

Can I just add that your graphics are the first electrical diagrams I've seen that make some sort of sense to a layman such as myself and should become part of the schools' curriculum, for those studying electronics, in my humble opinion.

Thank you - I do take pride in my diagrams. Being a visual person myself, I figured it would be far easier to "see" a layout. Let us know as questions arise.

With regard to the LEDs, I bought them from a UK supplier (although I suspect it is simply a wholesaler of Chinese or other far Eastern packs of LEDs, which are forwarded onto the purchaser, such as me). They arrived in standard resealable zip-bags (I bought a mix of colors in packs of 100). No idea of brand etc., althought the supplier lists both the blue and white ones as having a forward value as 3.2 (typical) and 3.8 (maximum), so I selected those colors for ease of the project..

Do you have a mcd value and a viewing angle for your LEDs? Example: 25° and 7,000mcd? Since you're only concerned with head-on viewing, the angle won't matter much, but it helps to know both the angle and mcd (light output). An LED with a 10° viewing angle and 8,000 mcd is going to appear a lot brighter (and blinding) than an LED with a 30° and 9,000 mcd.

For now, use 1kΩ for resistors X19, Y26, Y27, and Z19, and 270Ω for the rest of the LED resistors. Assuming negligible voltage drop across the MOSFET, this will give about 9mA across each LED. If this appears to dim, you can use something like 560Ω and 150Ω, respectively, to get about 16mA to each LED.

I'd suggest first putting three LEDs in series for each color. Note the drawing shows each LED as a "D" - one side has a flat side if viewing the LED head-on or from the bottom. The lead closest to this flat edge is the cathode (-). The lead farthest is the anode (+).


To connect three LEDs, put three in a row on a breadboard.

1. The cathode from LED1 will be connected to the anode of LED2.
2. The cathode of LED2 will get connected to the anode of LED3.
3. Connect a 270Ω resistor to the cathode of LED3.
4. Connect +12VDC to the anode of LED1.
5. Connect the - or GND of your 12VDC supply to the other end of the resistor.
6. Set the board on its side so you can look directly at the LEDs.
7. Walk away from the board and look at the LEDs at the distances you expect the display to be viewed from.
8. Are the LEDs too bright (try a larger value resistor), too dim (try a smaller value resistor), or just right? Do for both colors.
9. As a bonus, try viewing both colors at the same time so you can make sure one color isn't much brighter than the other.

 

Thanks, guys, for all your help.

It's becoming far, far clearer to me now.

From a scared position of pure ignorance you've helped transform me into a (reasonably) confident project builder, although there will doubtless be some stumbles, hair-tearing and gnashing of teeth in frustration, ahead.

Thanks again, I'll let you know how it goes and rest assured shall yell "help!!" as and when I approach my nervous breakdown.

All the best, Jib T.

 

How much is too much? Check out these: blue and white. You require 40 white LEDs ($3.50 USD for 50) and 147 blue LEDs ($7.40 USD for 200).

Doh!! My mistake. I was confusing diffused LEDs (which I hadn't even costed) with FLASHING LED's (which I was going to use as a checkening-out alternative). I believe the flashing variety are considerably more expensive (at least that is what I found, when I searched ebay etc).

Thanks for correcting me and pointing out how inexpensive the diffused variety actually are.

I'm going to order some and compare the outcome of each.

Jib T

EDIT:

I don't know what I was on when I searched before as, following your lead, I've just discovered these which are also relatively inexpensive:

http://www.ebay.com/itm/100pcs-5mm-White-Flash-Flashing-Blink-Water-Clear-Bright-LED-Leds/221175558103?rt=nc&_trksid=p2047675.m1851&_trkparms=aid%3D222002%26algo%3DSIC.FIT%26ao%3D1%26asc%3D18240%26meid%3D2115400733410441867%26pid%3D100005%26prg%3D8344%26rk%3D3%26rkt%3D5%26sd%3D221128598277%26#ht_1295wt_1170

 

I was confusing diffused LEDs (which I hadn't even costed) with FLASHING LED's (which I was going to use as a checkening-out alternative).

Be forewarned that if you opt to use dedicated flashing LEDs, you'll probably need to wire them all in parallel which means you'll need a resistor for each and every LED and require more current. This will also prevent you from adjusting the flash rate and there is a good chance some LEDs may flash at different times/rates than others. Will be more difficult to find diffused flashing LEDs as well. Just some things to keep in mind.

 

Be forewarned that if you opt to use dedicated flashing LEDs, you'll probably need to wire them all in parallel which means you'll need a resistor for each and every LED and require more current. This will also prevent you from adjusting the flash rate and there is a good chance some LEDs may flash at different times/rates than others. Will be more difficult to find diffused flashing LEDs as well. Just some things to keep in mind.

Good enough reasons to abandon that notion then, I reckon.

Thanks for the advice.

 

Using a 100μF cap for the 555 timing produces a .48Hz flash rate, which means that the pattern alternates once per second. If you want it to alternate twice per second, change C1 from 100μF to 47μF.

 

Using a 100μF cap for the 555 timing produces a .48Hz flash rate, which means that the pattern alternates once per second. If you want it to alternate twice per second, change C1 from 100μF to 47μF.

Cheers for that.

This is all (sort of) starting to make some kind of sense to me (that said, I'm glad I didn't seek advice on building a rocket....a matrix of flashing LEDs is MORE than enough for my poor old brain at the moment!!!).

I'm constructing the matrix on a set of breadboards and plan to construct the controller on another breadboard, prior to linking them up.

A couple of questions arise (and I put my hands up to my likely apparent ignorance to you and many of the guys on the forum):

1) Do I have to link a separate connection from each and every one of the resisistors (I'm assuming should be on the controller-board) to each individual and relevant string of 3xLEDs? That's a lot of wires to get right (hence my breadboard dry-run decision).

2) Can I gradually add strings of 3xLEDs, testing as I go or should I construct the entire project, prior to administering the power-source?

Don't worry if you deem either question to be so dumb as to not warrant an answer, I'll simply trial-and-error it.

Really trying not to be a nuisance, here, in respect of all the help I've been afforded and to you chaps for caring enough to be bothered to - prior to which I was beginning to feel a little like a sheep that has gotten lost in the fog. Baaaaaa!

Many thanks.

Jib T.

 

All the resistors connect to 12VDC; only one wire required.

If I were doing it, I would construct the controller first and connect three LED/resistors temporarily to simulate groups x, y, and z. Each of these three can be one LED in series with one 1k resistor.

Once you have proven that works, you can add as many or as few LED strings at a time as you desire.

 

All the resistors connect to 12VDC; only one wire required.

If I were doing it, I would construct the controller first and connect three LED/resistors temporarily to simulate groups x, y, and z. Each of these three can be one LED in series with one 1k resistor.

Once you have proven that works, you can add as many or as few LED strings at a time as you desire.

So obvious (when it's been pointed out). I really do need to apply a little more common-sense. Am in danger of letting myself down and/or exposing myself as being a complete numpty.

Sometimes I scare even me

Thanks for the heads-up.

Jib T.

 

Do I have to link a separate connection from each and every one of the resisistors (I'm assuming should be on the controller-board) to each individual and relevant string of 3xLEDs?

All the resistors connect to 12VDC; only one wire required.

Let's make sure we're on the same page. Referring to the schematic on post #23, +12VDC gets connected to one side of each string of 3 LEDs. Note there are a few LEDs that are individual (not connected in a string). The other side of each set goes to a resistor. The other side of the resistor is connected to the other resistors in the same fashion which all connect to one side of switch SW2.

So yes, each string/set of 3 LEDs gets its own resistor. That is 65 wires (plus one big +12VDC wire) between the display and the control board if you opt to put the resistors on the control board. If you plan to put the resistors on the display board, you only need 4 wires - one wire for each of the three sections going to SW2, D2/D3 and a +12VDC wire.

I agree with Tracecom that doing things in piecemeal is far simplier and less stressful. Much easier to troubleshoot too.

Let us know the specs for the LEDs you decide on and we can recommend the appropriate values for R5-R69. Specifically, forward voltage and current.

 

Let's make sure we're on the same page. Referring to the schematic on post #23, ABSOLUTELY.............

....................If you plan to put the resistors on the display board, you only need 4 wires - one wire for each of the three sections going to SW2, D2/D3 and a +12VDC wire. That's the way to go, then. Less is more as far as I'm concerned (and you, too, I'd wager).

I agree with Tracecom that doing things in piecemeal is far simplier and less stressful. Much easier to troubleshoot too. Agreed and thankfully it's been confirmed I can do that.

Let us know the specs for the LEDs you decide on and we can recommend the appropriate values for R5-R69. Specifically, forward voltage and current. Will do, as soon as I get that far. This may be easy for some here, but not so for yours truly. But if/when it all goes to plan and my grey matter is breached by this new knowledge I'm appropriating, I hope and expect my understanding will lead to bigger and better things.[/QUOTE]

All I can continue to say, to the point of repetitive-exhaustion, thanks again guys.

If it weren't that I'm such a chatterbox, I'd be speechless regarding the help you've been extending.

Faith in human nature suitably and fully restored

 

Hi everyone,
First time poster (please be gentle).
I have a small(?) LED project and not the slightest idea where to start except in that I have researched what I believe I need and am armed with Soldering Iron (which I am used to) along with stuff I’m not used to as follows: Breadboard and cables, battery holders, AC/DC power adaptor, a shed-load of 5mm Ultra-Bright LEDs, Resistors, Capacitors, Diodes, Ne555 and Ne556 chips, Crocodile clips etc., etc.
I currently have no idea how to read an electronic schematic diagram (but am reasonably confident that I can learn).
I want to produce an up/down, flashing, LED sign and have attached my own (non-schematic) diagrams of what I hope to achieve.
I have no idea whether this is a simple or a difficult project, I am so “green-behind-the-gills” and do not even understand the advantages or otherwise of running the proposed LED project in either Series or Parallel mode.
I would wish to power the project from either AC/DC adaptor or a 12 volt car battery and only for very short (on-demand) periods of time, similar to how often the button is pressed at a Pedestrian Crossing, say.
Is anyone able to offer any help and advice, please?