Circuit to convert 230VAC to 0-1.225V microprocessor input?

Thread Starter

jmckenzie

Joined Feb 4, 2016
6
Hello,

I'm new to the forum. I'm currently studying for an MSc in renewable energy. I have a little experience with basic electronics and arduino.

Ive just started (well i should have!) a year long experiment which involves a 16 channel data logger. Unfortunately I've underestimated the complexity of data logging!

The inputs to the logger need to be 0 - 1.225V. I need to measure the power consumption of 2 AC circuits, I have 2 CT Clamps for measuring current and need to measure voltage as well.

I'm considering adapting the first circuit on this page: http://microcontrollerslab.com/ac-voltage-measurement-using-microcontroller/

I'm not sure what to use as the ground connected to the 2200uf caps? Do the caps and the 2m2 resistors allow for a potential divider between AC and DC negative? Or is it a coil tap ground or something similar? If someone could clarify how this will work I'd really appreciate it! before I plug something in just to go bang!

Thank you,

Jim
 

Thread Starter

jmckenzie

Joined Feb 4, 2016
6
okay now Ive read it again I realise the caps are for filtering harmonics. But could someone explain to me how the voltage reduction is achieved please? It seems like the voltage would just be too high for the op amp?
 

Dodgydave

Joined Jun 22, 2012
11,277
The two 1M2 resistors and the 2K2 resistor form a voltage divider for the op amp input, and the two resistors on its output half it.

Personally i would use a step down transformer to measure the voltage, and use a precision rectifier.
 

Thread Starter

jmckenzie

Joined Feb 4, 2016
6
The two 1M2 resistors and the 2K2 resistor form a voltage divider for the op amp input, and the two resistors on its output half it.
Okay, thanks.

Personally i would use a step down transformer to measure the voltage, and use a precision rectifier.
I'm thinking this could cause issues with accuracy? the transformer characteristics will change due to frequency variation and I'm not sure if the variation in voltage would be truly linear? I haven't looked into it so I suppose it might be negligible, however I've come across others saying that they gave up on that option due to errors.

That is a strange number. What is the part number for the A/D converter, multiplexer, o microcontroller with this analog input?
Its an ADS1258, but the voltage ref has been set at 1.225 with an LM4041-AIM3-1.2. I think the multiplexer is integrated. Its an atmel micro cant read the model atm because the screen is in front of it. The data logger is made by Controlord, its the Gigalog S. Heres the manual: http://www.controlord.fr/en/home.htm (in download section).

I thought it was a bit odd as well. Seems like it would of been a lot easier to use if the ADC was set at -2.5V to +2.5V or 0V to 5V - as it suggests in the ADS1258 data sheet. Could it be for greater accuracy?


I have this switching power supply which will be powering the data logger and the opamp circuit:
http://www.mouser.co.uk/new/meanwell/meanwell-lrs-family-powersupply/

Its the LRS-75-12. I'm not familiar with this type of supply, it seems to be combination of transformer and switch mode?

When I tried to measure the AC voltage between the negative DC and the AC live - there were sparks! Is there a reason why I shouldn't be doing this? Maybe because of the switch mode part? If I cant measure the voltage across it then surely I cant put a load of resistors across it either? Whats the impedance of a multi-meter? >1Mohm?

Thanks for replying guys.
 

Thread Starter

jmckenzie

Joined Feb 4, 2016
6
The two 1M2 resistors and the 2K2 resistor form a voltage divider for the op amp input, and the two resistors on its output half it.
Okay, thanks.

Personally i would use a step down transformer to measure the voltage, and use a precision rectifier.
Ive read others saying that they had issues with accuracy with this method. I guess due to the reactive impedance of the transformer it becomes inaccurate due to frequency change and maybe there would be a non-linear voltage relationship? Also the losses of the transformer would need to be taken into account when calibrating which could also be non-linear.

That is a strange number. What is the part number for the A/D converter, multiplexer, o microcontroller with this analog input?
Its an ADS1258 with integrated multiplexer feeding an atmel micro (cant see the model number atm because of lcd screen).
The data logger is a Gigalog S made by Controlord, the manual is available in the download section of this site:
http://www.controlord.fr/en/home.htm

Its says in the manual the AREF has been set to 1.225V with an LM4041-AIM3-1.2 so the analogue input is -100 to +1225mV. I'm not sure why, would of been much easier if it was 0-5V or -2.5V to +2.5V as the ADS1258 can be used. Maybe to do with resolution?

The power supply I'm planning to power the circuit and logger with is a 12V switch mode power supply. Is this going to cause issues?
These are the voltage readings across the AC & DC side:

(L) to (V+ or V-) = 205 VAC
(N) to (V+ or V-) = 43 VAC

AC Voltage was 242 VAC
DC Voltage was 0V across all combinations except of course V+ and V-.

I cant really make sense of that? Could anyone shed light on this?


Thank you for you replies guys :)
 

Thread Starter

jmckenzie

Joined Feb 4, 2016
6
Sorry i thought I'd lost my original message,

so it turns out that when I originally tested the voltages I must have shorted on the chassis or something and looking at the datasheet of the power supply it is switch mode.
 
Top