You are here

A Prototype LC-Meter based on an Atmel Mega163 Microcontroller - english summary

Error message

Deprecated function: Array and string offset access syntax with curly braces is deprecated in include_once() (line 20 of /mnt/web216/e3/48/5667948/htdocs/includes/file.phar.inc).

On the page i have briefly described my LC-meter prototype, some ideas and up to a certain degree some theory. In the end, the basis of all is Thomsons formula: w = 2pf = (L C)-1/2, where f is the frequency (or w is the angular frequency), L is the Inductance and C is the Capacitance of the circuit. On this page, i want to focus on some details of the implementation.

I have developed the prototype board with eagle 4.14. The schematic is spread over 4 sheets. Sheet 1 deals with the microcontroller, the LC-display for output and some keys for input. Sheet 2 shows the LC-oscillator with its controls. The third sheet shows a monoflop based on the famous 555 device. Measuring capacitance is traced back to measuring the pulselength. The last sheet deals with power supply and voltage regulator for 5 V (7805).

Since this is a prototype, the wiring between the blocks mentioned above is not yet implemented. You will not find them in the schematics. All signals are available on pin headers (or ... to be more precisely ... on ... what's the opposite of pin headers? ...). You have to connect the right port of the microcontroller with the right signals of the building blocks on your own. For details see this picture or that picture.

The board is shown here. I have tried to design it as a single layer board. Don't worry about the tracks on components side. As you can see on the photos, the tracks are replaced by short wires. You do not need a double layer board.

The unknown inductance is connected to JP7 or X3/X4. Since it is in series to L3, Lx can be very small. The oscillator will work in any case, because L never becomes smaller than L3. On the other hand, the oscillator might not work, if Lx and/or the losses therein become too high. In addition, we assume, that the device under test consists of the unknown Lx and some parasitic (or even true) parallel capacitance. This capacitance has to be small (what small ever means - a few pF? - i have to admit that i do not know). The reason is, that we do not have a simple LC-circuit, but a circuit consisiting of a L3 in series to a parallel Lx||Cx-circuit. I did not yet investigate this circuit. If L3 is small compared to Lx, probably it can be neglected. But L3 is 47 uH. Therefore below 500 uH, Lx should have only small parallel capacitances. Above 0.5 mH, there is a chance to neglect L3 and to see the parallel capacitance to Lx, if there is any.

If you connect more complec circuits - maybe networks or IF-filters - to this LC-meter, results will become meaningless. The basic assumption for this LC-meter is, that we have an ideal L and and a small ideal C in parallel to L. Other configurations are not considered or may lead to completely inaccurate results. In particular, losses in L are not considered!

As you can see in the schematics, there are some SIL (single in line)-relays. However, currently software supports only a few of them: K2/K7 and K1/K8.

The main purpose of K2 is to add an additinal, well known capacity to the circuit consisting of (L3+Lx)||(Cy+C13). The parasitic capacitance of Lx is assumed to be small. Cy is something like the parasitic capacitance of Lx+L3. This capacitance is not known. It is completely absorbed in C13. In the end, we can not distinguish between C13 and Cy. Therefore, K2 adds an additional capacitance, C15, which will lower frequency of the whole circuit. Then C13+Cy can be determined. If we know C13+Cy, we can determine from the frequency L3+Lx. Now, K7 comes into play. Since the accuracy of inductances, purchased from electronics dealers usually are of low accuracy (10%, 20% or even worse ), we have to determine L3 just like an unknown inductance. It seems to be easier, to get capacitances with low tolerances instead of getting such inductances. K7 will short cut Lx (and Cx, if there is any) and L3 can be determined from C15 (K2). Then we will find out the value of Lx. But keep in mind: Cx has to be small!

For NE555, the relays serve as range selectors. One out of three should be active at the same time. But i did not yet implement it in the firmware.

Here the link to the Eagle files (schematic and board) is here.

Undefined