Current Monitoring

General project help for Adafruit customers

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
mauifan
 
Posts: 174
Joined: Sun Aug 12, 2012 11:13 pm

Current Monitoring

Post by mauifan »

Here is a link to a commercial product that I am interested in:

www.emporiaenergy.com/how-the-vue-energy-monitor-works

The "problem" with this product (and others like it) is that it stores data in "the cloud." (There is also a terms of service issue, but I will ignore that for now.) Do you carry the supplies I would need to build one of these on my own? You have a number of microcontrollers and current sensors, but I don't see "CT Clamps" listed on your website.

User avatar
adafruit_support_mike
 
Posts: 67454
Joined: Thu Feb 11, 2010 2:51 pm

Re: Current Monitoring

Post by adafruit_support_mike »

We don't carry clamp-style current sensors, but you can probably find something at Digikey or Mouser.

Given the number of inputs shown in those product photos, I'd guess they're using Hall Effect sensors. Those are much less expensive than magnetic clamp-on current sensors:

https://www.digikey.com/en/products/fil ... 4spi4aHZAA

https://www.mouser.com/c/sensors/curren ... rs%7C~Type

I'd further guess that most of the board inside the enclosure is devoted to reading all those sensors. It may have a multiplexer, or an Analog Front End (AFE) chip that does most of the work and reports to a microcontroller or small computer.

That multiple input handling will be the hardest part to match. It's possible, but is a system-design problem. We can't say, "here, use this N-many channel input device".

User avatar
mauifan
 
Posts: 174
Joined: Sun Aug 12, 2012 11:13 pm

Re: Current Monitoring

Post by mauifan »

Could I ask you to elaborate a bit? I am not sure I follow your logic of using Hall Effect sensors over rectification of an AC signal and measuring the corresponding "DC" voltage with the ADC converter built into many microcontrollers.

It seems to me that the clamps would be needed no matter what. Each clamp wraps around a "hot" wire to be measured. An AC current flowing through the wire will induce an AC voltage proportionate to the current flow, which the board somehow measures.

User avatar
adafruit_support_mike
 
Posts: 67454
Joined: Thu Feb 11, 2010 2:51 pm

Re: Current Monitoring

Post by adafruit_support_mike »

mauifan wrote: Wed May 10, 2023 10:16 am I am not sure I follow your logic of using Hall Effect sensors over rectification of an AC signal and measuring the corresponding "DC" voltage
Hall effect sensors cost about $1.50. Current sense transformers cost about $10. If you want to monitor several channels, you're looking at a large difference in cost.

Rectifying and filtering to DC in hardware will also give you additional cost. It's certainly possible, but you also have the option to measure the AC signal with the ADC and do all the filtering in software. The software option works well in this case because the signal of interest is so slow: a microcontroller's ADC can capture so many samples of a 60Hz AC waveform that the error is negligible.

That ability to trade hardware for code is one of the major design issues of our era. Code has many advantages -- it occupies no physical space, the cost of redesiging and replacing code is very low, and the cost of reproducing it is basically zero -- to the point that people used to think 'analog' and 'hardware' would eventually disappear. The past 30 years have shown that 'making information about the world digital' is an inescapably analog hardware problem though, and that the digital approach gets harder as the signal of interest gets faster. These days, experienced designers know how to estimate the relative difficulty of both options and will choose the one that's likely to be easiest.

Hall effect sensors provide an ADC-ready signal at a lower cost and with many fewer parts than transformer-based sensors. The multichannel nature of the project multiplies that advantage.

60Hz wall power is easy to digitize and to process digitally, suggesting the whole project will probably benefit from moving as much of the work to software as possible. And the project will get the best possible advantage from software's near-zero reproduction cost.

With those advantages for a design with as much code and as little external hardware as possible, it's worth asking whether a transformer-based design has advantages of its own.

Frankly, I don't see any in this application. The transformers won't give you better isolation or produce a signal that's easier to digitize. Rectifying will introduce its own error terms (the exact loss to diode voltage, for instance), and simple filter circuits still leave a fair amount of ripple. You'll end up doing much of the conversion to DC in code anyway.

Overall, it looks to me like the project will benefit from the mostly-digital approach.

Locked
Please be positive and constructive with your questions and comments.

Return to “General Project help”