Using BB to measure power

[jred...@gmail.com]

I'd like to use my BeagleBone Black to build a small power efficiency station for my school. For that I plan to measure power consumption using a current transformer and a voltage transformer, feed the data into a small & fast database and show it through a web interface to the school staff.

I see that 50 Hz are not a problem for BB ADCs, but I'm not sure at all using them is a good idea.

Intially I planned to use the same voltage I must measure to power the BB. I'm starting from a circuit like the attached one (replacing the UA78M33 by a UA7805CKCT which provides up to 1.5 A output)

Changing the R1/R2 divisor I can make the 220V signal lower , but I'd always get a signal with a negative side (-0.9V - +0.9V) in the best case. Same for the current transformer, after applying its output to a 100 ohms resistor. So, first, is it an awful idea to use BB for this purpose? If not, how can I avoid the negative part of the signal before feeding the BB ADCs?

Thanks in advance

[Przemek Klosowski]

Ouch, and another ouch since you seem to live in a 220VAC country. You can't just connect 220V to a voltage regulator---it has maximum allowed input voltage around 35V---you'd exceed that by a factor of almost 10.

You probably should either do some reading about line voltage electronics and 220V power supplies (hint---what you propose could work if you used a transformer to get 220V down to 12V or so).

My suggestion to you would be to consider a low-cost commercial power meter like Kill-A-Watt ($20 or so) then point a BBB with a webcam at its display, and do
a little image processing to read out the power. People also cracked them open and interfaced directly to their internal circuitry.

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to the Google Groups "BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Dieter Wirz]

Really ouch. I you do it like this and connect your BB(B) to the lan,
then all computers in your school are under tension (Brrrzzzz)!
Therefore: NEVER USE A VOLTAGE DIVIDER WITH NET CURRENT, unless you
really now what you do!
You need to have a galvanic separation like a transformer or a diode.

Have a look at this:
http://www.instructables.com/id/Real-time-Web-Based-Household-Power-Usage-Monitor/?ALLSTEPS

On Wed, Oct 30, 2013 at 10:31 PM, Przemek Klosowski
<przemek....@gmail.com> wrote:
> Ouch, and another ouch since you seem to live in a 220VAC country. You can't
> just connect 220V to a voltage regulator---it has maximum allowed input
> voltage around 35V---you'd exceed that by a factor of almost 10.
>
> You probably should either do some reading about line voltage electronics
> and 220V power supplies (hint---what you propose could work if you used a
> transformer to get 220V down to 12V or so).
>
> My suggestion to you would be to consider a low-cost commercial power meter
> like Kill-A-Watt ($20 or so) then point a BBB with a webcam at its display,
> and do
> a little image processing to read out the power. People also cracked them
> open and interfaced directly to their internal circuitry.
>
>
> On Wed, Oct 30, 2013 at 2:57 PM, <jred...@gmail.com> wrote:
>>
>> I'd like to use my BeagleBone Black to build a small power efficiency
>> station for my school. For that I plan to measure power consumption using a
>> current transformer and a voltage transformer, feed the data into a small &
>> fast database and show it through a web interface to the school staff.
>> I see that 50 Hz are not a problem for BB ADCs, but I'm not sure at all
>> using them is a good idea.
>> Intially I planned to use the same voltage I must measure to power the BB.
>> I'm starting from a circuit like the attached one (replacing the UA78M33 by
>> a UA7805CKCT which provides up to 1.5 A output)
>>

[José Luis Redrejo]

El miércoles, 30 de octubre de 2013 22:31:43 UTC+1, Przemek Klosowski escribió:

Ouch, and another ouch since you seem to live in a 220VAC country. You can't just connect 220V to a voltage regulator---it has maximum allowed input voltage around 35V---you'd exceed that by a factor of almost 10.

You probably should either do some reading about line voltage electronics and 220V power supplies (hint---what you propose could work if you used a transformer to get 220V down to 12V or so).

Maybe I didn't explain it correctly: I'm connecting a current transformer and a current voltage transformer (220/9) to the circuit in the scheme. So the voltage regulator is receiving 9v (about  12.7 V , but with about 0.6v less because of the diode rectifier)

 

My suggestion to you would be to consider a low-cost commercial power meter like Kill-A-Watt ($20 or so) then point a BBB with a webcam at its display, and do
a little image processing to read out the power.

I  don't like this solution, too many things to add...

 

People also cracked them open and interfaced directly to their internal circuitry.

mmm, do you know if there are any publications of the results of these cracks? They might provide me some ideas.

[José Luis Redrejo]

El jueves, 31 de octubre de 2013 13:28:12 UTC+1, Dieter Wirz escribió:

Really ouch. I you do it like this and connect your BB(B) to the lan,
then all computers in your school are under tension (Brrrzzzz)!
Therefore: NEVER USE A VOLTAGE DIVIDER WITH NET CURRENT, unless you
really now what you do!
You need to have a galvanic separation like a transformer or a diode.

I do, I do, maybe I didn't explain it . In my first message, the CONN_2 connector is connected to a 220/9 AC Transformer, so, I'm using 9 V AC to feed that circuit.

 

Have a look at this:
http://www.instructables.com/id/Real-time-Web-Based-Household-Power-Usage-Monitor/?ALLSTEPS

Great, it's very similar to what I want to do. But, re-reading the article I see he's using a power factor of 0.75 and a voltage value of 110 V, to calculate Active Power. I want to measure these parameters to get the real active /reactive power values, that's why I have the ADC_VAC line in my circuit. That's exactly the line I don't know how to pass from negative/positive values to only positive values lower than 1.8 v to feed the BB ADC.

[AndrewTaneGlen]

Hey Jose,

You could implement a very basic rectifier http://en.wikipedia.org/wiki/Rectifier#Single-phase_rectifiers to get an always-positive voltage

Alternatively, you could use something like the ADE7763 (http://www.analog.com/en/analog-to-digital-converters/energy-measurement/ade7763/products/product.html), in combination with your step-down transformers to get all of the volts/amps/power/phase infor you're after. I would recommend this option.

Regards,

Andrew.

[José Luis Redrejo]

Thanks for your answer Andrew, I'm going to checkt the ADE7763. In fact I was studying now the MCP3909 or MCP3901. These products look similar to the ADE7763. Any advice to choose the best one to interface with BB is very welcome.

Regards.

José L.

[Przemek Klosowski]

> mmm, do you know if there are any publications of the results of these
> cracks? They might provide me some ideas.

For instance,
http://lowpowerlab.com/blog/2012/12/28/wattmote-moteino-based-wireless-killawatt/killawatt-moteino-internal-connections/

and there are others, just google internal connections to kill-a-watt

[jerr...@utp.edu.co]

Lo solucionaste?, si es asi, como lo hiciste, gracias

El contenido de este mensaje y sus anexos son únicamente para el uso del destinatario y pueden contener información  clasificada o reservada. Si usted no es el destinatario intencional, absténgase de cualquier uso, difusión, distribución o copia de esta comunicación.

[Alan Corey]

First of all you really are planning to use transformers for isolation, right?  Current transformer in this sense: https://en.wikipedia.org/wiki/Current_transformer

There are also clamp-on ones, but you don't need to make any physical connection to the 220 volts AC you're trying to measure.  It needs to be just one wire though, not a pair, because the second one has current flowing the opposite direction which will cancel the first one.  So you'd probably need an electrician to get into the electrical panel and hook it up for you.

I think it's possible to flip the reference voltage around so negative would be the default.  If you look at the pinouts there are 2 pins for voltage reference, just ground the positive one instead of the negative.  I've mostly only used dual slope integrating A/Ds and those would read negative just fine, but this is a successive approximation one.

A wacky idea, if you've got a sound input, would be to feed the AC out of the current transformer into a sound input and monitor the amplitude of the waveform.  You'd probably want to do a bunch of averaging to smooth it out.  These A/Ds are much faster than you need so you'll probably want to average out some noise anyway.

I dunno, I just got my Beagle in the mail yesterday, but I bought it because of the A/D inputs which I want to use to monitor solar panel outputs.  I think the reason they're there is for touchpad X,Y inputs but there are 6 at least, you could borrow one for something else.