I already did a capacitance meter for arduino and test it on some capacitors and it outputs almost the correct value, however when I try to connect the arduino based capacitance meter in the parallel plates capacitor it outputs a very small value compared to the original capacitance meter.
/*
MeasureCapacitor
Measures the capacitance between D7 and A2.
Prints the result to the Serial Monitor.
This example code is in the public domain.
*/
#include <Capacitor.h>
// Capacitor under test.
// Note that for electrolytics the first pin (in this case D7)
// should be positive, the second (in this case A2) negative.
Capacitor cap1(7,A2);
void setup() {
Serial.begin(9600);
}
void loop() {
Serial.println(cap1.Measure()); // Measure the capacitance (in pF), print to Serial Monitor
delay(1000); // Wait for 1 second, then repeat
}
What are the input impedances of the meter and Arduino based meter?
Sorry, but I don't have idea what impedance is
but I took some photo maybe it could help.
The material used for parallel plate capacitor is two copper
Low capacitance values are difficult to measure. I wouldn't really trust either measurement. 
If you have a known-value capacitor of approximately the same value as your homemade one, you can measure it with both setups to see what you get.
But if you have a "good" multimeter the specs should give you the range and accuracy. Note that specified accuracy is often "full scale" so if there is a 1000pF range and it's rated for 1% accuracy, the readings could be off by 10pF on any reading with the meter set to that range.
I would trust a laboratory grade, calibrated, capacitor meter or LCR meter to be within it's published specs,
What frequency did each use, that will have a big effect on reading. Cabling is also important.
Depending on the size of the plates and the separation, the capacitance WILL be a very small value. Google will even point you to a formula to actually determine the capacitance of your capacitor much more accurate than your meter.
