 | Enter Your Electronics & Design Project for a chance to win a Grand Prize for Originality, a Tool Set, and a $100 Shopping Cart! | Project14 Home |
| Monthly Themes | ||
| Monthly Theme Poll |
Just for fun I thought I would consider various colours to spray the 3D printed case for the TinyDVM. The 3D printed colour of white is OK but I thought it might look better with some alternative colour. So I tried yellow and black sprayed onto the first version of the case. I decided the black would be better. (My apologises for the video which is a bit blurred towards the end. This is something to do with the camera, I dropped it once and now sometimes it gets blurry. Maybe a new camera is needed.)
I could not get the 3D printed case completely smooth and the flat black paint didn't look as good as the trial, probably because there was no yellow undercoat. Still it looks OK. I left the bezel white for a better contrast with the display and then fitted everything inside, drilled a small hole in the end for the connecting wires, stuffed it all into the case and added the bottom. I also programmed in a simple program that just measures the voltage and displays it on the OLED display.
I'm still using the example OLED clock programme provided in the documentation for the OLED and just inserted a while loop near the beginning before it gets into any of the clock display statements. I might even leave the clock display part in and make it a combined DVM and clock (maybe not!). The while loop is listed below:
while (1)
{
oled.setCursor(0, 0); // points cursor to x=0 y=0
oled.print("TinyDVM"); // Print a const string
value = analogRead(0);
oled.print(" "); // Print a const string
TinyV = (3.30 * value)/1023;
oled.print(TinyV,3); // Print a float
oled.print("V"); // Print a const string
oled.display(); // Draw to the screen
delay(300);
} /* while */
It is all working nicely now and displaying voltages. I am currently using 3 decimal places for the display, just to see how well it works. With a 10 bit ADC this provides a resolution of (1/1023) * 3.30V = 0.00322V, so by displaying 3 decimal places the display changes in steps of approximately 0.003V. It might be better to just use two decimal places. The voltmeter used for comparison (DT-830B Digital Multimeter) is a cheap one and only uses two decimal places on the 0-20V range needed for measuring voltages in the range 0.00-3.30V. Also, the TinyDVM display fluctuates fairly rapidly on the third decimal place and hence sometimes on the second decimal place as well. This is most likely due to some sort of noise being collected by the wiring inside the case or the leads used for making connections. The multimeter probably uses a double integration filtering method running at 50 Hz to reduce this type of noise, which is why digital voltages change at such a relatively slow rate. I do not think I will get into that much filtering but I might try averaging over two or three readings which should help considerably. This variation also affected the refresh rate. When running at maximum speed the final digit or two even are just a blur, so I inserted a delay of 300 ms just to slow everything down. Once I start filtering I might reduce this delay.
The next step was to calibrate the TinyDVM which I did with the DT-830D multimeter. This multimeter is good enough for everything that I get involved with so is suitably accurate to calibrate the TinyDVM. Below is an Excel graph of DT-830B voltages measured against TinyDVM voltages. The controlled variable, which is the voltage measured by the DT-830B goes along the X-axis and the measured variable along the Y-axis.
As this shows it is completely linear apart from at the maximum voltage. This is because the largest value the Arduino ADC can provide is 3.30V no matter what the input voltage. Also, although it cannot be seen on the graph there is a small positive offset of 0.02V. I'm not entirely sure what this is, possibly some small voltage offset in the wiring somewhere, or maybe just something to do with the Arduino ADC. However, it appears to be consistent and I could easily subtract this from the readings before displaying them.
The next step will be to improve the display. It is a graphical display so it might be fun to try and make use of the capability. Also, although I hadn't intended to, I could possibly make use of the WiFi capability of the Arduino MKR 1010 WiFi. I have not tried this out yet so I'll have to see how difficult this will be in the time remaining.
Top Comments