"In Circuit"
This project comes in two versions:
as an "add-on" for the 5x7 project
(pages 1&2) and as a "stand-alone" project (pages3&4)
Read both sections before deciding on which version you will build.

Use this Site Map for guidance

This project adds three features to our Microcontroller Course. It shows:
The versatility, cheapness and adaptability of a microprocessor project.

Firstly, we will describe the "add-on" version for the 5x7 Display as this is the cheapest version and, quite frankly, it only deserves a few dollars as a piece of test equipment. 
It's "all the rage" to have an Electrolytic Tester for servicing equipment and yet I have serviced over 36,000 appliances without one. Possibly it took me a long time to realise electrolytics have the capability of drying out because old-style electros were inherently very reliable and, of-course, they were not old when the equipment was being repaired. 
Now . . .  it's not impossible to live without one. All you have to do is replace all the electros in a faulty piece of equipment, then turn it on and see the results. 
Charles had a stereo amplifier in for repair recently and he started at one end of the board and replaced every electro. He tried the amplifier at regular intervals and as the electros were replaced, the output improved. After replacing the last few, the output was fantastic. 
Russell had the same experience. He was servicing a fax machine. After about 50 electros, the screen came on with the correct start-up instructions, the beep tone was correct, the machine automatically answered the line and the fax was readable. All these faults gradually developed over the years. And they were all due to electro's drying out!
Faulty electros create enormously unusual faults. All due to drying out.
Only last week Charles had a 4 watt amplifier (operating from a plug-pack) and the background hum was excessive. A new 1,000u electro in the power supply reduced the hum to near-zero. 
In this case ripple voltage was getting into the amplifier. Exactly the same thing can occur with a more-complex piece of equipment. The motor and thermal heaters in a fax machine produce glitches and "noise" on the power rails and this gets into other sections of the circuit. Electrolytics are designed to absorb these ripples by acting very similar to miniature rechargeable batteries. They absorb the spike when it is higher than normal, and deliver energy to the power rail when the rail voltage is lower than normal. 
Sometimes (very rarely) an electro is designed to pass a signal from one stage to another. If the electro is dry, the amplitude of the signal delivered by it will be lower than expected and faulty processing may occur. There are a few other areas where electros are used (such as integrating and separating signals) and it is not connect to either power rail, but most of the time they are connected across the power rails. 
In most cases the circuit surrounding an electro can be classified as low-impedance. This means it is difficult to test electros while they are "in-circuit" (because components such as chips are closely connected to them - and they have a low resistance). But if the voltage on the electro is below rail voltage, the effect of the surrounding components is minimised.
Basically electros cannot be tested with a multimeter unless you provide a voltage to charge them and then measure how long it takes them to discharge through a known resistance. That's why you need an Electrolytic Tester. 

The PC board for the Electrolytic Tester is shown below. It is connected to the 5x7 Display project with 4 leads. The termination on the 5x7 Project is a 5-pin plug (one dummy pin for alignment)). Two "test leads" are taken from the Electrolytic Tester PC board to the electrolytic-under-test. One has an earth clip (crocodile or alligator clip) and the other has an E-Z clip (a long clip with a very small hook on the end that clamps around a lead). 

The figure below shows the Electrolytic Tester PC board connected to the 5x7 Display. The leads are identified as positive, RA1, RA0, earth.

The ELECTROLYTIC TESTER connected to the 5x7 Display

There are two ways to test an electro "in-circuit." 
The first method is to determine the time needed for it to charge to a known voltage. 
The other is to charge the electro then discharge it through a known resistance. 
Both methods have an inherent inaccuracy due to the surrounding circuit components being connected to the electro. When the electro is out of circuit, the current from the tester will charge the electro to a certain value in a known time. From this we determine its value. When the electro is "in-circuit", some of the current will flow to the surrounding components and the electro will take longer to charge. The test will think a larger electro is being tested and will produce an over-value. 
To avoid this we have opted for the discharge method. 
If the surrounding components have a very low impedance, the reading from the tester will be lower than expected. In this case, the answer is to remove the electro from the circuit and re-test it. 

This project is an add-on for the 5x7 Display. It has been designed to show how easily you can add to an existing project. 
The Electrolytic Tester section consists of 7 components on a small PC board. This board connects to 4-pins on the 5x7 Display with 4 lines. 
The operation of the project is done by the program in the microcontroller. The components on the Electrolytic Tester PC board are merely interfacing components. They connect the lines of the micro to the electrolytic-under-test. 
The 220R limits the charge/discharge current to a known value and the transistor detects when the electro has discharged. 

The circuit looks very simple but the two components in the "font end" provide a very accurate discharge-time.
These two components are the 220R and the voltage-reference (made up of the two green LEDs). A green LED produces a characteristic voltage across it of 1.9v. Two LEDs have a voltage of 3.8v across them and this voltage remains very constant. This voltage behaves exactly the same as if a 3v8 zener diode was in circuit.   The advantage of the LED(s) is visual. You can see when the electro is charging.
The electrolytic-under-test is charged to 3.8v.  This gives it a known amount of energy and if it is 1u, it will have "one-unit" of energy. If it is a 100u, it will have 100 units of energy. If we place a resistor across this electrolytic, it will take one unit of time to discharge the 1u and 100 units of time to discharge the 100u. The resistor we are talking about is the 220R when the micro line is taken to 0v. Our second accurate voltage-point is the lower voltage-level. This is detected by the base of a transistor. The electrolytic is discharged between 3.8v and 0.7v. Between these points the discharge-rate is fairly linear. 

This project shows how add-ons can be connected to a project. You have already seen how the 5x7 matrix can be used to display figures and letters as well as animation. Using this knowledge gives you an enormous scope for creating a really impressive project by merely adding a front-end.  The 4 lines to the 5x7 use two port lines that can be configured as input or outputs and although the lines are already connected to another device, with clever circuitry you can add other things to these lines. If you need any more lines, they can be taken from under the 5x7 board via a separate plug-and-socket. 
Examples of projects that can be connected to the 5x7 are: 
A Heart-Rate monitor, an Ultrasonic Tape Measure and a Blood Alcohol Detector. There are lots of other ideas and anything that needs a readout can be connected. 

No matter what you design, Talking Electronics has made it easy to get you started.  We have a small proto board about 4cm x 3cm containing long pads and a positive and negative rail as shown below. It has no holes and has been designed for prototyping. It is called Experiment Board MkIII. 
Our method of prototyping may be new to many readers, but has been described many times in our 20 years of publishing. 

Experimenter Board MkIII $3.30 plus post. 

The only real way to prototype a project is on a board shown above. It is very difficult to build a project on a board with holes and circular lands because you have to turn it over to join up the components and this takes a lot of complex thinking. You may have seen a lot of "scratch-pad" areas on development systems consisting of holes and circular lands. But have you tried to use them!  They are not really suited for development work. They are for the next stage in a design. Once you get a circuit working you can transfer it to an area where it can be added to a micro. But where do you do the actual designing? This is the purpose of the Experimenter Board. It allows great flexibility and ease in placing components without much effort. This is important when you are designing something. You don't want to be distracted by the placement of the components. 
With our design, the parts are soldered (with long leads) to the lands and almost no jumper wires are needed. All the parts sit in a "birds-nest" or "rats-nest" arrangement and you can see exactly how everything is connected without looking under the board. No only is this quicker, but fewer mistakes are made. At the end of the development work, the parts can be taken off the board and re-used or fitted onto a "scratch-pad" as explained above. This is what I used in developing the Electrolytic Tester and it makes prototyping so much easier. 
The next stage in the development process is making the PC board. But firstly a circuit diagram is drawn  and the board is created on one of the PC board drawing packages. An overlay or top-layer is always included on all our boards and this is one of the things that stand us out from most other magazines and books. Our overlays include the value of every component and the board has the name of the project. Many magazines have a code-number on the board and  this might be ok for the month of issue, but who can remember a code number in 2 year's time!  That's why so many magazines have fallen over and the same with kit suppliers, and even manufacturers! It's simple things like this that make your product accepted or rejected. 
You can get PC boards manufactured for as little as $100, on a panel 25cm x 15cm, so it's not impossible to produce a project yourself and get it fully developed without any outside help. 
Obviously our 5x7 concept is very "experimentally orientated" but it's the only way to get an idea up-and-running. Once you get the concept working, the final design can take a completely different form. But without prototype-ability, it is very difficult to start. 
From start-to-finish, the Electrolytic Tester took 2 full days of programming. Most of the program was taken from the 5-Digit counter and one of the software items you need to develop a project like this is MPASM. This program takes your ASCII file (.asm file) and produces a .hex file for loading into PIP-02. To download MPASM, click HERE.

 Go to: Construction. 
Or:   Burning the program into the chip and using the Tester.