Right, I agree - that’s another thing that will have a little revision in the next version. Some people had difficultly calibrating to 240V if using a 9vac transformer (which is a number 0-65,535). To get to 120v it’s around 30,000.
Apart from that, I calculated, and compensated for, a 2.5 degree error with the 9vac transformer that comes with the kit for 120V people.
Yeah, the ATM90E32 is 6000:1 for active energy, according to the datasheet. But yeah, still no reason to not let in some more mV. Even if it were to get a bit warmer, there is temp compensation too.
The statutory voltage limit in the UK used to be 240 V + 6% = 254.4 V. It’s very likely that this voltage will still be reached occasionally, for some consumers. It’s now 230 V + 10%, but I don’t think there’s been a rush to replace all the transformers.
I followed one of your links above re 240V and came across…
////// VOLTAGE
if (LineFreq == 389) {
//if 50hz then voltage is 240v and needs to be scaled
voltageA = eic.GetLineVoltageA()*2;
voltageC = eic.GetLineVoltageC()*2;
I don’t think that’s a safe assumption. There are a handful of countries (Jamaica maybe?) that use 50Hz 120V.
Although we (those of us in the US) tend to think of our system as 120 VAC, it’s actually
a 240 Volt system as well. (a single phase split into two 120 Volt “legs” WRT residential service)
I’ve seen my voltage (I’m in rural Oklahoma, so the power system here tends to be flaky)
go as high as 256 Volts, which is in line with what Robert described.
When I receive my board, should I attempt to replace the voltage input resistor in the divider to attain higher ADC voltages for the power chip? If so, what value? I think this question is equally addressed to all experts in this thread, not just @jdeglavina. Obviously, startup parameters would also need to be changed to accommodate the change.
The point about attenuating the voltage and then amplifying it has more to do with noise and interference than anything else.
If you (say) attenuate the voltage by a factor of 100 and then amplify it by a factor of 100, what it means is any noise or rubbish that gets onto the signal after the attenuator can be 100 times smaller to have the same effect as it getting onto the signal before the attenuator - or more to the point, getting onto the signal in the absence of both attenuator and amplifier.
If you need to up the voltage calibration by more than 65k, and you don’t want any of the output values scaled in software, then you can if you want to. With a 12VAC output AC Transformer you shouldn’t have to scale anything in software. You could change R14 & R18 to 51k 1/4W 1% 0604 resistors - like ROHM ESR03EZPF5102
The resistors can easily be swapped with a hot air soldering gun. For the software, you would only have to calibrate the voltage again, and phase angle offset, if you want (this would be a change in the ATM90E32 library).
Sorry, I didn’t see that you posted this previously. Yes, swapping the resistors would allow more voltage in. I meant setting a phase angle offset in case your AC transformer introduces some error. You probably won’t need to do that though.
FYI, for the next revision, I’m going to use 20k 1/4W 1% resistors in place of the 100k. This gives an input voltage of about 520mV with a 12VAC transformer.
No se si me recuerdas, Te escribi hace un par de dias con una screenshot similar al tuyo en esta publicación con mis datos totalmente diferentes, podrias ayudarme con eso?