How to read analog power meter?

Post by **H4X0R5** » Sun Aug 20, 2006 8:08 pm

Hi Guys, im am wanting to replicate the power monitoring but i dont have a digital power meter mine does not have a blinking led or anything like that, it only has a moving circular disk with a black line and every time the black line passes the window it is a cirtain value, there is no way of replacing it with a digital one, does anybody have any means of measuring this? Thanks

Post by **b_weijenberg** » Mon Aug 21, 2006 7:53 am

It is planned by RFXCOM to have a remote meter reading available in about 5 to 6 Months.

Post by **Richard** » Mon Aug 21, 2006 10:03 pm

b weijenberg,

Where did you find this info? Is there more info availble?

Richard

Post by **b_weijenberg** » Mon Aug 21, 2006 11:49 pm

This is inside information from a very trustworthy source.

Bert

Post by **Richard** » Tue Aug 22, 2006 7:26 am

i trust the (re)source

so i will wait full of curiousity.

Richard

Post by **ame** » Wed Aug 30, 2006 1:24 pm

Hi,

I have the same problem. Here in New Zealand I have two meters. One is for the general supply to the house: lights, sockets, oven, etc. The other is for hot water and an electric storage heater, and is switched on at night by ripple control. I leave the hot water turned on all year, and the storage heater on in winter only.

Both meters have a spinning disk with a black mark on them. You can see the edge of the disk through the glass on the front of the meter. The meters are different models from different manufacturers (night: Sangamo-Weston, general: English Electric).

Here is my solution. I started working on it yesterday, and it's quite easy.

My assumption was that it should be possible to detect the difference between the silver disk and the black mark on the edge of the disk. I made a simple circuit with a high brightness LED and a phototransistor on a small piece of stripboard, then I attached the board to the front of the meter with Blu-Tak and bent the LED and photodiode until the light from the LED was reflected from the edge of the disk back to the photodiode. I put a piece of black heatshrink cable around the phototransistor to reduce interference from stray light. When the black mark is absent, the phototransistor conducts due to the reflected light. When the black mark is present, the phototransistor does not conduct, as no light is reflected.

Here is the basic circuit:


+5V --+---------+
..... | ....... |
..... \ ....... \
.150R / ... 10k /
..... \ ....... \
..... | ....... +----- Out
..... | ....... |
..LED V -> -> |/ Phototrans
..... - -> -> |\
..... | ....... |
0V --+---------+-----


(ignore the dots- the forum code strips out multiple spaces, so I have to space things out with dots).

The output is low when the black mark is absent, and high when it appears.

There are a couple of problems, but I think I can work round them:

1) The edge of the English Electric disk is not smooth. It is rough, like a hacksaw blade. This means the light can flicker a bit. I have put a low-pass filter on the output, but I think it is too slow- now I miss the black mark if it moves too fast. Next, I will try a schmitt-trigger.
2) The edge of the Sangamo-Weston disk is very thin (but smooth). It is not possible to get the LED and phototransistor lined up accurately. My solution here is to use a cheap laser pointer- the reflected light is very bright even from such a thin edge.

I will probably build two identical circuits with laser pointers, then hook them up to a DS2423 one-wire counter. That way I can read both meters from one device.

Anyway, I hope this helps someone. It really works, and if you have a thick disk you can do it with an LED for a light source. It is cheap to build the circuit, and you can test it with a multimeter. If it works, add some filtering and hook it up to a counter or what have you.

Andrew

Post by **Bwired** » Wed Aug 30, 2006 3:01 pm

Hi Ame
Thanks for your solution!
Check this topic, same subject. http://www.domoticaforum.eu/topic.asp?TOPIC_ID=101

Post by **Digit** » Wed Aug 30, 2006 11:03 pm

You could also consider a DIN rail mountable KWh-meter with S0-interface like this one:

http://www.eltako.com/dwl/zaehler_e_typ ... er%20S0%22

The S0 can then be used as input for a counter.

Regards,
Digit.

Post by **ame** » Wed Aug 30, 2006 11:25 pm

Yes, but I already have a meter, provided by the electric company. I don't want to pay them to replace it, and they won't do it for free. At the moment my detector has cost approximately 2 Euro. I have used these components to verify that the idea works, so my next step is to build two 'final' designs based on what I have learned. I expect the total cost to be much less than 10 Euro. I shall be using a DS2423 to count the pulses, and I will connect this to the one-wire bus that I already have (currently with two temperature sensors connected). The software I am using, Digitemp, is capable of reading the DS2423, so there is very little extra work involved.

Andrew

Post by **Bwired** » Wed Aug 30, 2006 11:34 pm

Hi Ame
How does your photo transistor react when you open the door of the fuse box and some bright light comes in?

Post by **Digit** » Wed Aug 30, 2006 11:36 pm

Well,

My KWh-meter with S0 (still on the shelf somewhere, waiting to be installed...) was 115 Euro, and if you already have a working concept, the choice is not very hard.

Just FYI, you can put this KWh-meter in series with the existing one. Just cut the electricity cables somewhere after the existing meter and put the second one there. Of course, i don't know how your electric company feels about you cutting the cables...

Regards,
Digit.

Post by **ame** » Wed Aug 30, 2006 11:57 pm

<blockquote id="quote">quote:<hr height="1" noshade id="quote">Originally posted by Bwired
 Hi Ame
How does your photo transistor react when you open the door of the fuse box and some bright light comes in?
<hr height="1" noshade id="quote"></blockquote id="quote">
That is a good question. I have found that the heatshrink tubing around the phototransistor is very good at rejecting external light, and the phototranstor itself seems to be very focused (perhaps this is due to the shape of the clear plastic housing). Also, the meter cupboard is in a dim corridor in the house, and the light nearby has a lampshade that casts a shadow over the meter.

The worst case would be that the bright light would prevent the phototransistor from sensing the black bar, so when the cupboard was open, pulses would be missed. However, with low load the pulses are about 1 minute apart, and with high load about 20s apart, so if you are quick (don't leave the cupboard open) you won't miss many pulses. If I use cheap laser pointers then they are so bright that I can turn the gain of the circuit way down so that external light is unlikely to affect it.

I have attached a picture of the prototype sensor. The legs of the components have been bent to make the light reflect properly, and the board is stuck to the meter with blu-tak. I will make a bracket for the sensor later when I have finalised the design.
Andrew

Post by **ame** » Thu Aug 31, 2006 12:02 am

<blockquote id="quote">quote:<hr height="1" noshade id="quote">Originally posted by Digit
Just FYI, you can put this KWh-meter in series with the existing one. Just cut the electricity cables somewhere after the existing meter and put the second one there. Of course, i don't know how your electric company feels about you cutting the cables...
<hr height="1" noshade id="quote"></blockquote id="quote">Indeed...
Generally, cutting the cables *after* the meter is acceptable, but here in New Zealand it is illegal for a homeowner to do their own electrical work. There are some jobs a homeowner can do, but everything must be inspected by an electrician before it is made live.

Andrew

Post by **Bwired** » Thu Aug 31, 2006 12:37 am

<blockquote id="quote">quote:<hr height="1" noshade id="quote">That is a good question. I have found that the heat shrink tubing around the photo transistor is very good at rejecting external light, and the phototypesetter itself seems to be very focused (perhaps this is due to the shape of the clear plastic housing). Also, the meter cupboard is in a dim corridor in the house, and the light nearby has a lampshade that casts a shadow over the meter.
<hr height="1" noshade id="quote"></blockquote id="quote">I had really problems with the lights when I opened the door. So no I have now a photo diode which only reacts on the IR light of the LED. I can open the door now and all goes well. The fuse box closet is important form me, I have all kinds of devices in it like LAN patch panel, router, 1-wire patch panel, Camera patch panel, alarm etc.
Regards Pieter

Post by **MindBender** » Thu Aug 31, 2006 11:29 am

I've been working on this problem too and I also had the problem of extra light from an opening door. This could be solved using a second - balancing - phototransistor that's pointed in the same direction as the first one, but not pointed to the wheel. This second phototransistor could adjust the DC offset of the op-amp.

I have also tried to use the (Agilent) sensor of an optical mouse to detect the movement of the wheel. But this sensor will need another lense than used in a mouse, because the standard mouse lense is focussed too close and doens't have enough focus depth.

And for those who want a ready-to-use solution; Buy a (extra) meter with a W/h pulse output:
http://www2.produktinfo.conrad.com/date ... eeicht.pdf
Out mutuals friend Klaus Conrad (http://www.conrad.nl) sells them for 269 Euro and 299 Euro, resp. non calibrated and calibrated. I think the calibrated meter may even be used to replace your regular standard issue meter, but ask your local electricity suppliers first. You're most definitely not allowed to switch meters yourself (but that wouldn't stop me).