TESTING THE CIRCUIT
To assist in testing the circuit, you may need a Logic probe. The RGB LED Dice circuit is very simple and the following Logic Probe can be put together in less than an hour. It will indicate HIGHs and LOWs as well as pulses.
The author did not need the Logic Probe to test the RGB LED Dice project as the circuit worked the first time it was switched on.
However the circuit was tested by getting a 6v supply with 470R series resistor and probing the pins of the 8-pin IC socket without the chip being installed.
As each set of pins was tested, the corresponding colour illuminated to prove the wiring was correct.
The author makes a point of only using a simply multimeter and a Logic Probe to test any of the projects in this series because these are the only pieces of test gear that will be available to the average hobbyist.

So, we have two areas of interest. Construction and Programming and it's up to you to take it on.
The project is designed for all sorts of uses, including models such as train layouts, alarms and similar effects.
But the real thing we want to get across, is programming.
This is another example of using a simple 8 pin chip to provide a number of features that would take many logic chips (such as counters and gates) and lots of components to duplicate.
It also highlights our method of hand-coding as an effective way to produce a program.
This project uses about 400 instructions to produce the effects and it uses the EEPROM to store the sequence produced by the user (sequence 1) - and show it at turn-on.
In this respect, some of the sub-routines in the program are quite complex and suitable for the advanced programmer. However, if you are a beginner, you can read through the program and most of the sub-routines will be easy to follow as each line of code is explained. You have to start somewhere and this project offers a challenge.
Most projects with a program of this complexity are only available as a pre-programmed chip or only the hex code is available. There is usually no attempt at educating the reader in programming.
That's the difference between our projects and all others.
We offer a learning curve.
For every hour of effort you put into reading, building and using one of our microcontroller projects, you get the experience of 100 hours of effort that has been put into the design to make it appear simple.
All you have to do is start . . .
 

PROGRAMMING THE CHIP
The kit comes with a pre-programmed PIC chip but if you want to program your own chip or modify the program, the .hex file is available as well as the assembly file, so you can see how the program has been written and view the comments for each line of code.
The PIC12F629 is one of the smallest micros in the range but you will be surprised how much can be achieved with such a tiny micro.
The program contains sub-routines to produce delays, sequences on the display and both read and write EEPROM jobs that require accurate code - including a special sequence - called a handshaking sequence that prevents the EEPROM being written due to glitches. 
Even a program as simple as this is not easy to put together and to assist in this area, we have provided a whole raft of support material.
Not only do we provide a number of programs with full documentation but our approach to programming is simple.
It involves a method of "copy and paste" whereby sub-routines are taken from previously written code and copied into your program. Any modifications are made in very small steps so that each can be tested before adding more code.
This is exactly how we produce a complex project. Each step is written and tested before adding the next step.
This saves a lot of frustration as it is very easy to add a line of code that is incorrect and get an unsuspected result. 
If you follow our suggestions you will buy a programmer ("burner") called a PICkit-2 if you are using a laptop. It is the cheapest and best on the market and comes with a USB cable and 2 CD's containing the programs needed to "burn" the chip. If you are using a desk-top and/or tower with a serial port, you can use a cheaper programmer called MultiChip Programmer from Talking Electronics. You will also need NotePad2 to write your .asm program. This can be downloaded from Talking Electronics website. You will use RGB LED Dice.asm  or RGB LED_Dice-asm.txt as a template for your program, plus a 6 pin to 5 pin connector that fits between the burner and the project. This is also available on Talking Electronics website.
As we said before, this project is for medium-to-advanced programmers as it is very compact and does not have in-circuit programming pins.
To be able to modify the chip you will need a programming socket and this can be obtained from one of our other projects that contains the 5 pins for in-circuit programming.
You can then put the chip into the other project to be programmed and modified and re-fit it into this project for execution.

PROGRAMMING LANGUAGE
There are a number of kits, programs and courses on the market that claim and suggest they teach PIC Programming.
Most of these modules and courses use a PIC microcontroller as the chip carrying out the processes, but the actual programming is done by a proprietary language invented by the designer of the course.
Although these courses are wonderful to get you into "Programming Microcontrollers" they do not use any of the terms or codes that apply to the PIC microcontroller family.
All our projects use the 33 instructions that come with the PIC Microcontroller and these are very easy to learn.
We use the full capability of the micro and our pre-programmed chip is less than the cost of doing it any other way.
In addition, anything designed via our method can be instantly transferred to a PIC die and mass produced. And we use all the input pins and all the memory of the chip. The other approaches use less than 25% of the capability of the memory and one of the pins is not available.
In fact it would be difficult to reproduce this project via any of the opposition methods. It would require a larger chip and more expense. 
You can use our method or the opposition. Just be aware that the two are not interchangeable.
Ours is classified as the lowest "form" (level) of programming - commonly called machine code - invented in the early days of microprocessors - and now called mnemonic programming as each line of code is made up of letters of a set of words. The opposition uses a higher level language where one instruction can carry out an operation similar to a sub-routine.
But you have to learn the "higher level language" in order to create a program. And this requires a fair amount of skill and capability.  
It sounds great and it is a good idea. But if you want to learn PIC programming, it does not assist you. It is "a step removed" from learning PIC language. The other disadvantage of the opposition is the "overhead." The 1,000 spaces allocated for your program is filled with pre-written sub-routines. You may require only 10 of these sub-routines but ALL of them are loaded in the memory space. And they take up all the memory.
You have no room for your own program.
To get around this the opposition uses the 128 bytes in EEPROM to deliver instructions on how to apply the sub-routines. This provides about 30 powerful instructions using their language called BASIC (or a similar language). 
It's a bit like selling a diary filled with all the paragraphs you need to express yourself, and leaving a few blank pages at the back for you to write single lines such as: see page 24, paragraph 7, see page 63 paragraph 4, to create your diary entries.
It depends on how much you want to be in charge of writing a program. Using our method is like writing your own auto-biography. Using the opposition is like getting a "ghost writer."
When using a higher level language to create a program, you have absolutely no idea how the code is generated for the micro.
In some of the developmental kits, the code is "locked away" and you are NEVER able to access it.
Everything runs smoothly until a fault appears. With our method you can see the code. With the other methods, you cannot see the code - it's like doing key-hole surgery without the advantage of an illuminated endoscope to see what you are doing.
Everything has its place and our method of hand-assembly is only suitable for very small micros and you will eventually need to "learn a high level language."   The PIC12F629 has over 1,000 locations for code and this equates to more than 20 pages when printed, so this is about the limit to doing things by hand.
But our drive is to show how much can be done with the simplest devices on the market, at the lowest cost.
Anyone can show you high-technology at a high price but this is not where you start and this is not where you get enthusiasm.
We provide the things to get you started. That's the difference.

CHARLIE-PLEXING
When ever you invent a product or idea, GIVE IT A NAME.
That's the first thing I do when I produce a new project.
It gives you a "point of reference."
And that's what has been done by the inventor of the circuit we have used in this project.
It's called CHARLIE-PLEXING and is basically the concept of connecting as many LEDs as possible to the outputs of a micro.
Each output of a micro can deliver current (25mA) when it it HIGH and sink 25mA when it is LOW.
This means it will illuminate a LED when two outputs are in particular directions and another LED when the outputs are reversed.
With 3 outputs, 6 LEDs can be illuminated, but not all at the same time.
With 4 outputs, 12 LEDs can be connected and sometimes more than one LED can be illuminated at the same time.
With our RGB LED Dice project, two LEDs are connected in parallel (called Display LED "A" and "A+"), two more LEDs are connected in parallel (called Display LED "B" and "B+") and two more LEDs are connected in parallel (called Display LED "C" and "C+"). 9 leads are produced from these 6 LEDs.
The final LED (the centre LED of the display ("D") has 3 more leads. Thus we have a total of 12 leads and by charlieplexing we can illuminate any of the colours from any of the LEDs.  We have already mentioned that all the LEDs cannot be turned on at the same time, however sometimes more than one colour can be turned on at the same time and this will be discussed later.
We now come to another interesting feature.
The maximum current from each output of the micro is 25mA.
If you drive any of the colours of the RGB LEDs with 25mA, the light output will be blinding. The LEDs will be far too bright to view and the colours will be blurred.
We have a choice. We can drive them with 12mA or put 25mA through them for 50% of the time. You would think this comes to the same brightness. But no so.
Our eyes are non-linear and by delivering 25mA for 50% of the time, produces a brightness almost equal to the full 25mA - so it is a much more effective way of illuminating the display.
In fact we can reduce the time to about 10% and still produce a viewable brightness. The reason is the LED comes on at full brightness @ 10% of the time and our eyes see this peak and retain the brightness until the next peak arrives. If we pulse the LED very quickly, the LED will appear to be bright ALL THE TIME.
We need to employ this trick because we cannot turn on all the LEDs at the same time. Depending on the LEDs we want to illuminate or the effects we want to show, we may have up to 8 different colours being illuminated, one after another.
All the complexities of how the colours are turned on has been worked out and each number, from 1 - 6 has been placed in a separate sub-routine. All you have to do is call the sub-routine and the appropriate colour and particular set of "pips" on the display will be illuminated..
And the same applies to the effects. These are worked out separately and placed in sub-routines.
There is only one simple rule to remember.
To turn on a colour in an RGB LED, one output has to be HIGH and the corresponding output has to be LOW.
No other LEDs on these two outputs can be illuminated. However LEDs on the other two outputs can be activated and if this can be done at the same time, the display will be slightly brighter if it reduces the size of the "run." This is the number of individual "illuminations" required to produce an effect. Very good brightness can be achieved with a "run-of'-four" as the LEDs are high-bright. 
You will notice that some of the unused colours inside the LEDs are connected to the outputs we are driving and to prevent these LEDs illuminating, the unused outputs are turned into INPUTS. This means the line is neither high or low and is equivalent to the LED being disconnected from that particular line.
The outputs can be changed to inputs at any time during the running of the program and this gives us great flexibility with driving  the LEDs. 
It is quite easy to work out which lines have to be made HIGH or LOW or turned into an INPUT. Simply look at the circuit.

MODIFYING THE PROGRAM
To modify the program you will need a PICkit-2 programmer and this comes with 2 CD's containing all the software needed for In-Circuit Programming.
You will also need a lead (comes with PICkit-2) to connect the programmer to your lap top via the USB port and an adapter we call 6pin to 5 pin Adapter to connect the PICkit-2 to your project.

6pin to 5pin Adapter


Adapter connected for In-Circuit Programming
(the chip is placed in another project for in-circuit programming
or any PC board with 5 In-circuit Programming pins)

The PROGRAM
The program looks complex because the LEDs are being accessed individually and each illumination requires a separate sub-routine. Two LEDs marked "A" are in parallel, two LEDs marked "B" are in parallel and two LEDs marked "C" are in parallel. LED "D" is a single LED. This means the display is limited to driving and illuminating these pairs of LEDs and you cannot produce random effects on the display without taking these limitations into account.

The program



does a bit of detecting when turned on. It detects to see if a bit has been set in EEPROM to tell the micro to go to a required sequence or start with sequence 1.
It also detects if switch A or C has been pressed at the instant the project is turned on so that the micro is directed to the sub-routine where the user-sequence can be entered or if the EEPROM bit is to be cancelled.
All this gets done in the SetUp routine and then the micro goes to Main.
 
In Main, the program increments a "jump" file and calls a table where it finds a directive to go to a particular sub-routine.
The sub-routine is executed and the micro goes back to Main where it looks for a release of SwA. This forms part of a key debounce as the key must be fully debounced as it is advancing the micro through the sequences.
To provide a totally reliable debounce, the key is detected as not being pushed for the duration of a whole cycle of a sequence and a separate loop is then executed where the key can be detected as being pushed, to advance the program to the next sequence.
To create your own sequence as sequence1, the project is turned off and SwA pressed while turning the project ON.
This sends the micro to a sub-routine called Attract.
As soon as SwA is released, the program starts to time the duration when a switch is not pressed and it "times-out" after 2.5 seconds.
The program also times the duration when a LED is illuminated. It also accepts 2 or 3 LEDs illuminated at the same time. These are all clever instructions that need to be looked at to see how they operate.
Up to 15 steps can be entered and each step occupies three bytes. The first value identifies the illuminated LEDs, the second byte identifies the ON duration (in increments of 5mS) and the third byte identifies the OFF time.
These 45 bytes are contained in files 30h to 5Fh.
When a switch is not pressed for 2.5 seconds, the program "times out" and sends the values to the EEPROM. It then shows the sequence on the LEDs.
If the project is turned off and on again, this sequence will be displayed as sequence1.
To replace the sequence with something else, simply repeat the steps above.
If you want one of the pre-programmed sequences to appear each time the project is turned on, simply advance through the sequences by pressing SwA and when the desired sequence is playing, push SwB.
This will record your choice. Turn the project OFF then ON again and the chosen sequence will be displayed.
To remove this feature, press SwC when the project is off and at the same time, turn the project ON.
All these feature have been added to the program, one at a time, and it is important to add them in the correct order. For instance, you can only add a removal feature after the initial feature has been produced. Reading and writing to the EEPROM is a most complex operation and the instructions must be laid out as shown in the program, as they include a hand-shaking sequence. When you need this code it is copied and pasted in its entirety, to prevent a mistake.
Nearly every instruction has a comment to explain not only what it does, but why it was chosen.

NEW:

To illuminate each set of LEDs, the in-out resister (TRIS) must be loaded and the out lines must be taken HIGH or LOW. Any line configured as an input is effectively removed from the circuit and does not have any effect.
The first thing to do is create a sub-routine to drive each of the LEDs: "A," "B," "C," and "D."
The produce a sub-routine that uses 1, 2 or 3 of the previous sub-routines.
We have already mentioned the fact that the LEDs are too bright when driven at 25mA and to reduce the brightness, a short delay much be included.
As you increase the number of sub-routines, you will notice that previous routines can be used and this saves a lot of space.
Programming is like owning a LEGO factory. As you introduce new products, you can use many of the previously developed Lego blocks and bricks.


 

Here are the files you will need:
RGB LED Dice.asm
RGB LED_Dice-asm.txt
RGB LED Dice.hex
 

;RGB LED DICE.asm ;**************************************************** ;RGB LED FX.asm * ;25 sequences to demonstrate the possibilities for * ;an RGB LED * ;22-5-2011 * ;**************************************************** ; 2100h END

HISTORY
LED Dice projects have been around for a long while.
The simplest circuit uses 6 LEDs and they get scanned by a CD 4017 chip. This is not very impressive.

 The next circuit comes from a kit by Talking Electronics. It shows the LEDs in the formation of the spots on a dice and the circuit slows down to give the impression of the rolling of a dice:

The next circuit uses a very interesting feature of the 4017 chip. The "carry out" (pin 12) is HIGH for the first 5 clock pulses.
The first output is pin 3, so that just pin 12 is HIGH and thus LEDs "B" illuminate to produce "2" on the display.
The next output to go HIGH is pin 2 and this makes LEDs B and A illuminate to produce "3" on the display.  Using pin12 reduces the complexity of the project while producing the "pips" on a dice.
This circuit shows the importance of reading a datasheet thoroughly and understanding what is going on with the outputs of a chip. The circuit does not have the "slow down" feature.

The following circuit uses a PIC12F629 to drive a 7-segment display. It has the "slow down" feature.

The next project uses a PICAXE-08 chip that is really a PIC12F629 but you do not use any of the PIC instructions to create a program. You use instructions created by the supplier of the chip and you are effectively learning nothing about programming a PIC chip. The chip could be an Atmel or NEC or Philips microcontroller.

Here are 4 more LED Dice projects:

		The first photo shows a project using an oscillator chip and 4017. The second photo uses a 555 a 4071 OR-gate and a 4017 counter. The circuit is below. A few diodes can take the place of the 4071 chip and the project is over-designed. The third photo shows a microcontroller project using an Atmel micro. The fourth photo shows a multivibrator made from 2 transistors and a 4017 chip.