Earth resistance meter [150084 ]

With this project, I like to build an Earth resistance meter (Megger).
As this is not possible to do with a regular multimeter, and commercial devices are very expensive, I believe DIY is a good alternative. And … it's a lot of fun to do !
I did some research on the internet and found a way to measure earth resistance with a regular 24V transormer. This way, you can obtain the earth resistance value "Rx" by applying an AC current "I" between the measurement object "Earth electrode under test" and "Current injection electrode" and finding out the potential difference "V" between "Earth electrode under test" and “Voltage drop sense electrode”
This is fully explained on: http://wiki.diyfaq.org.uk/index.php?title=TT_Earthing#Using_temporary_current_injection_and_voltage_sense_electrodes
Instead of using a standard volt- and ampere meter and Ohm’s law, I like to implement everything in an Arduino Uno project to make a ergonomic meter.
I use two standard metal threaded bars of 40cm for the electrodes (see image)
I like to use a 16x2 LCD and one “START” button to trigger the measuring sequence. Also, the meter must be software and hardware protected for over current, open circuits, etc….
Because we measure currents and voltages in an “earth” load, we need proper optical isolation between Arduino and the electrodes.
See schematic “Megger1.pdf”
I use 2 Isolated linear sensing IC by AVAGO: HCPL-7520 ( http://www.avagotech.com/docs/AV02-0956EN ) to isolate both voltage and current values. This IC needs 2 separate 5V sources: primairy side = Vaa and secondary side = Vbb = 5V Arduino. I use 2 small AC adapters by Velleman and a 7805 regulator for Vaa.
I use a shunt resistor of 1.5 Ohm to measure current (IC1) and a high impedance voltage divider (R3 R4) to measure voltage (IC2) On both inputs, I use 2 antiparalleled diodes to protect the inputs for over voltages. I also use a PTC resistor to protect the transformer for short circuits.
Both AC voltages go thru a LPF of 1.5 kHz to minimize HF noise. (R6, R7, C5, C6)
I use 2 ADC’s in the Arduino to measure both current and voltage values. Because these signals are 50Hz AC, I need to measure for a few seconds and pick out the max values.
Both AVAGO IC’s have a DC voltage of 2.5V on the output (1/2 VCC)
Because I like to maximize the resolution of the ADC’s in Arduino, I use a reference voltage of 2.5V. This way I only measure the negative half of the 50Hz signals, so each step of the 10bit ADC = 2.5V / 1023 = 2.44 mV
As we measure only the negative half of the AC signals and the “0” level is at 2.5V, I need to subtract these measured values from 1023 (max value of the ADC) to obtain the true value.
In order to measure correctly I have add some Ohm law formula's that are implemented in the software:
Voltage:
Uin = (Urx * R3) / (R3 + R4) where Uin = voltage on input IC 1
Current
Iin = Uin / Rshunt where Uin = voltage on input IC 2 and Rshunt = 1.5ohm
(1,46 Ohm more specific)
Software:
See file attached !
Note: this is my first Arduino project, I believe I have an “analog” thinking brain.
At school I got traumatized by having to program a 8751 processor with UV light window and “machine code” and... never got something working. I must say, Arduino open's a new world to me :-)
I have a feeling that this is a good working project, I measured my own ground connection at home and is seems be be correct (compared to the value I got from Vincotte)
Please feel free to comment or send your remarks !
Discussie (8 opmerking(en))
MAK007 5 jaar geleden
redas 5 jaar geleden
frederik 9 jaar geleden
I have add a new schematic.
I will generate a square wave of 220Hz out of port D0 of the arduino.
(220Hz is right between 3rd harmonic of 50 Hz and 60Hz)
I send this to an optocoupler, so all logic is isolated.
The square wave goes to a simple lowpass filter of 300Hz (R15 / C19) so we get more or less a sinewave.
Now we amplify the signal with a simple audio amp (NE380, 2.5W)
I use 18V (2x 9V battery) as power for the amp and the optocouplers (via 7805 for the Avago's)
I use isolated DCDC convertor to generate isolated power for the arduino.
I will build this in the coming week, more news soon.
I found a tool to calculate a digital filter in Atduino, might be helpfull: http://www.schwietering.com/jayduino/filtuino/
grtz,
Frederik
sbsb 5 jaar geleden
ClemensValens 9 jaar geleden
frederik 9 jaar geleden
A few idea's for improvement:
- I like to use arduino to generate a 'non interfering' with 50Hz or 60Hz frequency to avoid interference with existing earth leaks. Anyone a suggestion about what's the best frequency? I found online that 137Hz might be good.... not clear to me why.
- I like to use an optocoupler to isolate arduino with the outside world (TL113)
- I like to use LM386 audio amp to drive the signal into the ground.
- I don't need a constant current (as stated previously) , because im using the 'fall of potential' method!
- I like to use 2x 9V battery in series to power the lm386, and use mini isolated DC DC convertor (5V) to power the arduino. This way all measuring remains isolated.
- I like to implement digital filter in the arduino software, so i only measure the 137Hz component. I might need some help in doing this :-)
Any suggestions or tips are very welcome. I keep posting new findings :-)
ClemensValens 9 jaar geleden
frederik 9 jaar geleden
ClemensValens 9 jaar geleden
Redfloyd 10 jaar geleden
four-point-method.jpg (24kb)
fall-of-potential-method.png (38kb)
Chris.KB3CS 6 jaar geleden
an RF techician would find a complex impedance measurement at various frequencies from 137 kHz to 25 MHz very useful.
iirc, if you detect not only amplitude from your measurement bridge (yielding the magnitude of the impedance) but also phase from your bridge, then you can compute Z = R +/- jX
KEN MACDONALD 6 jaar geleden
PAVAN KUMAR KUMAR BHUKYA 6 jaar geleden
frederik 10 jaar geleden
frederik 10 jaar geleden
I did some tests regarding accuracy of the measurements.
I found out that there's an measurement error at low resistances.
I attached an xls file with a line graphic of what I measured, also I figured out how to correct this with a simple math formula.
Soon I will implement this in the software!
UPDATE 01/08: new xls file with more accurate calculations and new sofware with correction implemented.
Frederik
megger1_9.zip (2kb)
frederik 10 jaar geleden
New version of the schematic, with ATmega328 processor and power supplies in it.
Also add transistor to control backlight, still need to program this!
Im planning to draw a pcb so I can use the meter without having to give up my arduino board.
When i boot the arduino, the relais output (D0) goes up and down a few times, is there a way to avoid this?
Cheers,
Frederik
frederik 10 jaar geleden
Hey Clemens, all,
I have one little problem with my design where I might find the answer here:
When I push the start button, sometimes i get random caracters on the LCD. After resetting the arduino, all is working fine again. This happens at random, not clear what causes this.
I implemented a small delay in the software after the button is pushed, to avoid bouncing.
All suggestions are very welcome!
I also like to make a PCB so i can finalise the project.
Im planning to build a standalone pcb, without Arduino attached, good idea? Or is a shield a better idea?
grtz,
Frederik
ClemensValens 10 jaar geleden
Nice! In your case, what were the distances between the three rods? 10m like they mention in the diy faq?
Regards
Clemens
frederik 10 jaar geleden