Power monitoring in the workshop?

[Jasper Wallace]

Hi,

The Not just arduino group is talking about doign some group projects and
one idea is power moniting in the workshop.

Has anyone else been looking into setting it up? presumably we'd need 3 x
current clamps and something to do the A/D conversion. Has anyone started
gathering bits?

I've started a wiki page:

http://wiki.london.hackspace.org.uk/view/Unit_23_Power_Meter

Related:

This is the wiki page for the power monitoring in unit 24:

http://wiki.london.hackspace.org.uk/view/Electricity_meter

--
[http://pointless.net/] [0x2ECA0975]

[Nigel Worsley]

> The Not just arduino group is talking about doign some group projects and
> one idea is power moniting in the workshop.

The easiest way is probably to hook up to the meter's data port.

> Has anyone else been looking into setting it up? presumably we'd need 3 x
> current clamps and something to do the A/D conversion.

You need to measure the voltage as well, and then multiply it by the current at a fairly
fast rate, 100 samples per cycle (5kHz) is pobably enough, but faster will give a more
accurate answer.. A lot of the workshop equipment is going to present a significantly
inductive load and multiplying average current by average volts will give significant errors.

Another thing to note is that those current clamps will be current trasnsformers (inless
you pay big money for hall effect baed ones, which are usually only used for DC).
These have a current output rather than voltage, and neeed a burden resistor across
the output to work correctly. Some of them include this internally bbut many do not.

Nigle

[Dean Forbes]

I was looking at this and went from a simple solution to being confused .... not difficult

I geuss I am left with two question 

What have we used for the existing setup "Question" anyone know ??

And what level of acuracy do we need as opposed to want and will we go with want or need

[Russ Garrett]

On 7 November 2012 15:10, Dean Forbes <m...@deanforbes.com> wrote:
> What have we used for the existing setup "Question" anyone know ??

Our electricity meter in Unit 24 has a serial port. We're using that:

http://wiki.london.hackspace.org.uk/view/Electricity_meter

--
Russ Garrett
ru...@garrett.co.uk

[Ken Boak]

Openenergymonitor.org have just released an Arduino shield based wireless energy monitor

http://openenergymonitor.org/emon/Modules

Ken

[Nigel Worsley]

Ken Boak wrote:
> Openenergymonitor.org have just released an Arduino shield based wireless energy monitor

That would be a good start, but it only supports single phase. Probably wouldn't
be too difficult to add in 3 phase though.

Nigle

[Ken Boak]

Nigel,

It supports up to 4 current transformers. - so 3 phase not a problem.

[Nigel Worsley]

Ken Boak wrote:
>It supports up to 4 current transformers. - so 3 phase not a problem.

Yes it is, it only has a voltage input for one phase. This means that the
software needs to phase shift the signal by 120/240 degrees to
synthesise the other phases before doing the RMS calculation. It will
also fail to take into account the difference in voltage between the phases,
which can be quite significant. According to the cacti graphs the figures
for unit 24 at present are:

L1 = 232.82V
L2 = 240.47V
L3 = 243.75V

An energy monitor that was only looking at L1 would be out by about 4.5%
for the power on L3

Nigle

[David Sullivan]

On Wednesday, 7 November 2012 16:17:26 UTC, Nigel Worsley wrote: According to the cacti graphs the figures 

for unit 24 at present are:

L1 = 232.82V
L2 = 240.47V
L3 = 243.75V

An energy monitor that was only looking at L1 would be out by about 4.5%
for the power on L3

 

Would incoming voltage vary much from the readings from unit 24? Cheat and assume L1-3 are the same for both?

Sully.

[Dean Forbes]

In a previous life I was taught to look at the problem/requirement before providing a soloution 

The second part of my comment was 

And what level of acuracy do we need as opposed to want and will we go with want or need

What are the views on this ??? 

[cepm...@yahoo.co.uk]

I have made a three phase current transformer, This will make the
measurements more consistent than using clamp probes and will fit into the
dis. board.

There is an old penrecorder type power meter with clamp probes on the
shelf near the bike mountain. It contains some amplifiers that are matched
to the probes. Might be an interesting thing to resurrect.

Phil

--
" et cognoscetis veritatem et veritas liberabit vos. "

[Adrian Godwin]

Rather than continuously calculating instantaneous power, how about measuring the phase angle (by timing zero-crossings) and then measure voltage and current separately on 3 phases using a slow ADC ?

This would be much easier to calculate but wouldn't be accurate if we have a lot of non-sinusoidal loads. However, in the workshop it might be good enough - both because we probably only want a rough figure, and because motors do generally present sinusoidal loads, albeit with a phase shift. A big pile of computers would be a different matter. The welder could be an exception, but it's a transformer rather than an inverter type so may not be too bad.

I haven't tried this nor ever built a power meter. It's a suggestion to invite discussion and educate me.

[Nick Johnson]

Did I miss something? Didn't Russ say the power meter has a serial port, rendering all of this planning completely unnecessary?

-Nick

On Thu, Nov 8, 2012 at 9:41 AM, <cepm...@yahoo.co.uk> wrote:

[Kimball Johnson]

For Unit 24, not unit 23.

[Russ Garrett]

On 8 November 2012 14:15, Nick Johnson <arac...@notdot.net> wrote:

> Did I miss something? Didn't Russ say the power meter has a serial port,
> rendering all of this planning completely unnecessary?

The power meter in Unit 24 has a serial port. We're talking about Unit 23.

--
Russ Garrett
ru...@garrett.co.uk