The zener diode can not be there. The whole point is that we measure the voltage drop accross the resistor R2. Otherwise the voltage applied to the ADC pin will be the same as the zener and it will never change. Yes this is my idea too, but with MOS transistors instead. In particular SI9926CDY.

 

The zener diode can not be there. The whole point is that we measure the voltage drop accross the resistor R2. Otherwise the voltage applied to the ADC pin will be the same as the zener and it will never change. Yes this is my idea too, but with MOS transistors instead. In particular SI9926CDY.

I agree with dana that most ADC's nowadays work using differential signals, and make the use of a separate ground moot. But if you must have a separate ground, I suggest you take a look at dc-dc converters, such as this one. But the output of this kind of converter would have to be filtered somehow, because it's like a switched mode power supply, which are extremely noisy for their use in ADC circuits.

 

I am already using the LM2576 which should be a DC to DC step down converter?

 

The zener diode can not be there. The whole point is that we measure the voltage drop accross the resistor R2. Otherwise the voltage applied to the ADC pin will be the same as the zener and it will never change. Yes this is my idea too, but with MOS transistors instead. In particular SI9926CDY.

Yes, I just want to protects the ADC and almost forgot the zener diode will clamp the voltage which because R1 is there.

If you want to protect the ADC then you can use the circuit below, otherwise you just use your voltage divider.

 

So the circuit with the 2 grounds will not work?

If you can keep the battery ground completely separate from the micro ground, then your circuit in post #10 will work.
In effect you are just shifting the micro ground with the relay.

 

How can that be done? How do they do it on the arduino forum?

EDIT:
This is the circuit, I accept my chargers are with an isolated power supply (charger is ABC-1220D FST).

https://www.dropbox.com/s/zt3bkw6guij0etv/Battery.monitor.02.04.pdf?dl=0

I am trying to use a finder finder 32.21 x000 relay. But its rated for AC current and I can not find DC specification, can it be used, how much will the voltage drop be? Since I am measuring voltage it is important.

 

Why is LM117 (317) not good for this circuit?

 

Why is LM117 (317) not good for this circuit?

Good for what circuit, where?

 

Here is the circuit. It will be turned on for about 30mS every 10minutes. After that it will enter sleep mode. I was thinking of a capacitor paralel to "R7" in pico farads in order to filter any noise from the switching or anything else. Also an opto transformer to separate the batteries from the MCU, so it can have a different ground, despite being powered from the batteries indirectly.

https://www.dropbox.com/s/j3z0hlacjeblf5a/Circuit.pdf?dl=0

 

Please upload the circuits to our forum, otherwise when the site lose the connection then this part of thread will be become useless.

 

Sorry for that, I can not upload PDF and the picture quality is bad.

This is the circuit with 2 relays and an isolated DC to DC converter, while using the symbold analog ground (GNDA) as the second ground on the schematic:
https://www.dropbox.com/s/i0b0xe25w3ojd9b/Battery.monitor.2.relays.2.grounds.pdf?dl=0

 

Sorry for that, I can not upload PDF

Why?

 

.Creating a Virtual Power Supply Ground

Wendy's Index

 

I dont think the "pdf" extension is allowed?

 

I dont think the "pdf" extension is allowed?

Click on the "Upload a File" button at the bottom of the post and you will see all the files that can be uploaded.
.pdf is one.

 

Thanks for the explanations, they are very useful. But the problem is already solved I will use the circuit below, what bothers me is will the relay transfer the voltage and current, epecially since this model is for AC I think?

I also have given 2 options for a power supply. With LM117 and LM2576, which do you think is good enough? And the "22pF" capacitor on the reset button bothers me.

Tomorrow I will soder it and test it.

 

Its working. I used a different transistor and switched it from "3.3V", because it was all I had. I used the LM 317 power supply, but I did not upload the whole code to the chip, because its long. So I only used the ADC code to test, without the wifi part. It was draining about 30mA. With the wifi code it might be more.

 

Should I use Lm117 for the power supply or LM2576? I am worried that LM117 might overheat if the whole circuit drains about 100mA.

 

The power consumption of LM2576 is better than LM117.

 

Should I use Lm117 for the power supply or LM2576? I am worried that LM117 might overheat if the whole circuit drains about 100mA.

The dissipation in the LM117 will be (Vin-Vout) * 0.1A.
The LM117 is good to about 1W without a heatsink.