It did spin off from the other thread you mentioned, but I thought this warranted its own thread since my Arduino ADC inaccuracies are there regardless of what's feeding it. I think I've got a decent handle on the current-to-voltage conversion aspect of the project for now (many thanks to everyone in the other thread!)

I absolutely agree.

I figure it this way. In its basic state the Arduino is a 10 bit A/D resulting in 1024 quantization levels between 0 and Vref so using 5 Volts as Vref the best resolution we will see is just about 4.88 mV. With that in mind if I convert 0 to 50 Amps RMS to 0 to 5 VDC and do it linear I can scale things for 100mV/Amp. Even if my 10 bit resolution was 5 mV I could still resolve about 1/20 Amp. That being good enough for measuring household current.

Ron

 

I had to alter my reference in my sketch as it was giving a value of 1023 at just 4.7V. I think it's probably because it's a rip-off clone-duino nano.

If you require a more stable reference voltage, you can purchase the AD584 on a breakout board, and it gives 2.5V, 5V, 7.5V, and 10V at very stable reference levels. Mine cost £2 from eBay.

 

The reference comes from the ATMega328 so I doubt it's the clone.Especially if you are running off USB. I'd check your supply voltage - that's more likely the source of the problem. Arduinos are basically just a glorified breakout board so the clones will have few differences from the genuine ones (usually a cheaper USB serial bridge, VReg, and connectors).

And, yes, if you care about the ADC reference, get an external one like you pointed out - genuine Arduino or not.

 

By the way, the AVR8s have vrefs built in. The arduino library surfaces them in the analogReference call. Worth looking at so you don't have to use Vcc as your Vref. It looks like the Vref on the 328 is only guaranteed ~10% accurate (1.1V +/- 0.1) though likely to be better than that. You'd have to divide the input down to 1.1V - I'd use 1% resistors.