"Smart" consumer units?

[listerofsmeg]

Hi all,

We are getting a full rewire on our 1960's bungalow, including a new consumer unit.

I was wondering whether you can buy "smart" consumer units that allow you to monitor power usage on each circuit, rather than just on the entire supply?

Either they don't exist, or my google powers have failed me, so thought you guys might know of any?

Cheers

 

[ban-all-sheds]

Consumer units do not allow you to monitor the power usage on the entire supply.

What you want to achieve could be done, but why?

What do you intend to do with the data?

How much are you prepared to pay to get it? Do you have an ROI in mind?

Are you aware that most domestic electricians won't have a clue how to do it?

What skills do you have which are relevant?

 

[OwainDIYer]

Yes, sold for commercial units which have to monitor power loads over a certain level.

eg

http://www.smartpowershop.co.uk/electricity-meters/din-rail-mount.html

Discussed at
http://www.automatedhome.co.uk/vbulletin/showthread.php?1699-Sub-Meter-Electricity-Monitoring
although that thread is now a few years old

 

[stillp]

You could use something like the Acti9 system http://download.schneider-electric....e_Id=107688093&p_File_Name=Acti9-Panorama.pdf

 

[SimonH2]

One of our customers at work makes such equipment - for commercial environments. IIRC I had a discussion with one of their guys a few years ago to find out what they were using - with a view to putting it in our server room. The consideration for it was quite short once I found out the cost of the equipment - which would have been 4 figures even for our small DB

However, the OP may wish to consider something like this.
http://www.delorie.com/electronics/powermeter/
Unlike most "add on" energy monitors, this does calculate true power - there's space on the board for 16 power measurement chips which do all the measurements (voltage, current phase angle/power factor, and totalise it all up) together with a microprocessor that interfaces between these and whatever you want to use to interrogate it.

While I was looking into this some time ago, I came across a few options.

http://openenergymonitor.org/emon/
Don't know if most of the hardware can do true power measurement, or if it's just limited to measuring current and guessing about voltage/power factor. But I did spot mention of counting pulses from your lecky meter.

http://www.billporter.info/2010/12/19/not-so-tiny-power-meter/
Does true power measurement - in software. Only one channel.

http://smartenergygroups.com/samotage/posts/72-The-SEGmeter-specification
Another DIY system - dunno if it does real energy monitoring.

http://www.brultech.com/products/ECM1240/default.htm
Something commercial.


A common things whatever you choose is that to get true power measurement (rather than a guess) requires the system to measure all three of voltage, current, and power factor. The chip used in the first project at delorie.com does (from memory) many measurements per cycle and does instantaneous power calculations - which it then integrates to get a value useful to us. Some simply look for something like zero crossings to work out phase relationship - which is better than nothing but isn't necessarily accurate with non-linear loads. The cheaper options don't bother - so all of those "energy meters" that just clip round one of your meter tails are absolutely NOT energy monitors as the only thing they can measure is current - no voltage, no power factor, and so only a guess as the actual power.

Also common to all projects that measure each circuit in your CU - you'll need to be going into the CU to attach the CTs (current transformers). It shouldn't need saying that inside a CU is a hazardous place and you shouldn't go in there if you don't know how to work safely. Unless you use "clip on" CTs, you'll also need to be disconnecting cables - and it's easy (more easy on some boards than on others) to miss the clamp when putting them back.

 

[ban-all-sheds]

http://www.delorie.com/electronics/powermeter/

The NEC has the same sort of requirements for cables being insulated for the highest voltage present as BS 7671.

Anyone know what the individual cores of Cat5 are rated at?

 

[listerofsmeg]

Many thanks for all the replies. Just the kind of stuff I was looking for. Lots to read through now!

Thanks again.

 

[SimonH2]

Anyone know what the individual cores of Cat5 are rated at?

Not 240V IIRC But then they only have 110V there.
Not the neatest board installation either - plenty of bare copper showing.

 

[bernardgreen]

from http://www.drakauk.com/products/special_cables/special8.php

DC LOOP RESISTANCE: ≤ 190 Ω/km

RESISTANCE UNBALANCE: ≤ 2%

INSULATION RESISTANCE (500V): ≥ 2000 MΩ/km

CAPACITANCE AT 800Hz: nom. 48nF/km

CAPACITANCE UNBALANCE (PAIR TO GROUND): ≤ 1500 pF/km

CHARACTERISTIC IMPEDANCE (1-100 MHz): (100 ± 15)ΩΩ

NOMINAL VELOCITY OF PROPAGATION: Approx 67%

PROPAGATION DELAY: Nominal ≤ 535 ns/100m

DELAY SKEW: Nominal ≤ 20 ns/100m

TEST VOLTAGE (DC, 1 min) CORE/CORE: 1000V

 

[ban-all-sheds]

"Special cables" - so is that performance guaranteed to apply to any Cat5 cable?

Insulation resistance 2kΩ/mm - what does that tell us?

"TEST VOLTAGE (DC, 1 min) CORE/CORE: 1000V" - does that mean that it can't withstand that for more than 1 minute?

What really matters is does the maker certify that with unsheathed cores the cable meets the regulatory requirements for being in containments with 230VAC present?

 

[ban-all-sheds]

Not 240V IIRC But then they only have 110V there.

Not so.

http://en.wikipedia.org/wiki/Split_phase#North_America

 

[JohnW2]

Insulation resistance 2kΩ/mm - what does that tell us?

I presume they have chosen the wrong way to think about and express it, and hence also the wrong units - if it were purely a matter of 'resistance' (without considerations of insulation breakdown), resistance would be inversely, not directly, proportional to length - i.e. ≥2000 MΩ for 1km of cable would correspond to ≥2,000,000,000 MΩ for 1 mm of cable! The reality is, of course, that this concept of IR/length is really not valid/relevant.

"TEST VOLTAGE (DC, 1 min) CORE/CORE: 1000V" - does that mean that it can't withstand that for more than 1 minute?

I presume that is a reference to the duration of the test. However, I think the nature of the materials is such that if it can stand 1000V for 1 minute, it probably can withstand that voltage (pd) indefinitely - I'm not aware of any effects of long-term existance of a (sub-breakdown) potential across insulation, are you?

What really matters is does the maker certify that with unsheathed cores the cable meets the regulatory requirements for being in containments with 230VAC present?

Indeed.

Kind Regards, John

 

[SimonH2]

Not so.

http://en.wikipedia.org/wiki/Split_phase#North_America[/QUOTE]
I know they have split phase, but the highest voltage the insulation on the sensor cable would need to withstand would be 110V. Think about it ...

EDIT: OK, it's 120V not 110V, that's a minor detail.

Insulation resistance 2kΩ/mm - what does that tell us?

I presume they have chosen the wrong way to think about and express it, and hence also the wrong units - if it were purely a matter of 'resistance' (without considerations of insulation breakdown), resistance would be inversely, not directly, proportional to length - i.e. ≥2000 MΩ for 1km of cable would correspond to ≥2,000,000,000 MΩ for 1 mm of cable! The reality is, of course, that this concept of IR/length is really not valid/relevant.

My assumption was that they mean a 1km length of cable will have > 2GΩ IR. The implication being that 2km would have an IR >1GΩ or 1/2 km would have an IR >4GΩ.

However, I think the only way to express that as something you can multiply by length to get a value would be to specify the conductance - ie insulation conductance < 1/2000 µ&#937;/km. Something tells me that fewer would be able to work out WTF that means !

 

[JohnW2]

My assumption was that they mean a 1km length of cable will have > 2G&#937; IR. The implication being that 2km would have an IR >1G&#937; or 1/2 km would have an IR >4G&#937;.

Indeed, that's what I said, However, it does lead to some conclusions which I find rather hard to believe. Can you really believe that, say, the IR of 1m length of this cable really is >2T&#937; (i.e. >2,000,000 M&#937?

However, I think the only way to express that as something you can multiply by length to get a value would be to specify the conductance - ie insulation conductance < 1/2000 µ&#937;/km. Something tells me that fewer would be able to work out WTF that means !

As a point of detail, the unit of conductance is not &#937; (or &#956;&#937 - the unit it the Seimen (S), which is &#937;^-1. As you say, is the proper was of expressing it in a ‘per km’ manner. However, even when one does the calculations ‘the right way up’, I’m still suspicious of what the figures we’re looking at imply in terms of the IR of short lengths of the cable. What do you think?

Kind Regards, John

 

[SimonH2]

However, it does lead to some conclusions which I find rather hard to believe. Can you really believe that, say, the IR of 1m length of this cable really is >2T&#937; (i.e. >2,000,000 M&#937?

Why not ? If the cable is indeed uniform, then that is exactly what it means - regardless of how difficult it might be to measure. I think the only problem is a mental one - in that (in this context) there is a general attitude of "> xG == unmeasurable &#8776; infinite" where x is some fairly low number.

As a point of detail, the unit of conductance is not &#937; (or &#956;&#937 - the unit it the Seimen (S), which is &#937;^-1.

Ah yes, it's a long time since I use those.

I’m still suspicious of what the figures we’re looking at imply in terms of the IR of short lengths of the cable. What do you think?

I think it's just a mental hangup from dealing with "numbers we can't get our head round". Don't take this as a personal comparison, but remember Blackadder trying to teach Baldrick to add.
2 beans + 2 beans = "some beans", or a very small casserole
We can handle "2 beans" fine - most people have a meter that will measure up to at least several hundred meg, or a gig or two. Unless you've had reason to work with more sensitive gear (I haven't BTW), then it's hard to mentally picture what higher numbers mean.

Think what's it's like talking voltages to "the man in the street". We've got 'low voltage', like you get from a battery that won't kill you. We've got 'mains voltage' that will give you a belt and possibly kill you. And then there's 'everything more than mains voltage' and once you get there then the average man in the street would probably struggle to picture how (say) 132kV relates to 415V - both are "something more than my mains, and it'll kill".