Gas usage monitoring

[cliffwright]

Hi Folks,

I'm very new to Idratek and haven't even fully installed my initial setup yet . .

But I'm wondering . . . by using the pulse counting capabilities of modules such as the DRB002, how (and is it even possible) could I go about using a photo-reflector such as this;

http://www.maplin.co.uk/photo-reflector-sy-cr102-12709

to count the number of revolutions of the final dial on my gas meter that has a reflective silvered section on the thousands of a cubic meter number wheel . . .

Any ideas?

I see this chap;

http://www.hekkers.net/domotica/HowT...rGasUsage.aspx

has done something similar but I don't want to write my own application to handle things . . . I want something failry plug and play that can be fed into the pulse counting capability of an Idratek module . . .

Anyone any thoughts or experience of this?

Cheers

Cliff

[soitjes]

Don't know where you live, but in Holland they sell this device :

http://www.enymate.nl/english

It can measure gas, water and electricity. It comes with software but also with an API so you can access it yourself. Currently it's using Bluetooth for communication protocol. They told me an IP version would be available soon, once it's available I will buy it to integrate it in my application.

You can contact them to see if you meter is compatible with their product.

Note that I'm not affiliated at all with the company, and I have not used the product myself.

Soitjes.

[Karam]

Its really just a matter of whether the sensor is able to reliably distinguish the change in reflectivity, i.e to give you a reasonably shaped pulse when the silvered section passes by. Once you get a pulse the rest is handled by the pulse counting feature of digital inputs on DRB (or most other modules) and the metering side is already catered for in Cortex so there should be nothing else to do other than get the sensor to work reliably.

The SY-CR102 looks to be an active photo reflective detector i.e you supply some current to the LED side to deliver some light (not sure whether its IR or not) and then the receiver is a phototransistor which you can connect directly to a digital input (pullup resistor is already provided in the DRB). Remember to get orientation of connection correct - emitter goes to 0V and collector to digital input. The LED part can be powered from the IDRANet 12-15V line perhaps most conveniently derived from the module connector. You will need a current limiting resistor between this and the diode. The limited specs suggest a value of 20mA. I expect that this is a working value rather than max. nonetheless would suggest something between 680-1000ohms. Again beware of orientation

Since its a very basic signal you will need to also take measures to prevent incident lighting from causing false outputs. So taping the sensor to the gas meter and covering the area with black tape may help. Another possibility is that the reflectivity of the wheel in the non silvered areas is high enough to cause the phototransistor to turn on even when the silvered section is not in view. In this case you may have to reduce the current to the LED and see if you get a point where the difference between silvered and non silvered is well defined. The converse may also be true - so I guess its a case of experimentation. If you have a multimeter you can typically test the transistor's output before connecting to the DRB, i.e to see clearly that it is going 'on/off' as the silvered section passes by.

[ludditeal]

on UKHA_D recently there was a discussion on CurrentCost and I am sure when I looked at their web-page at the time there was wireless Gas meter reading device in the Envir range but now it has gone back to promising it. The current cost can be connected to a PC and whilst I haven't looked into it I believe there are a number of open source programs to process the data and produce reports/graphs.

Allan

[cliffwright]

Its really just a matter of whether the sensor is able to reliably distinguish the change in reflectivity, i.e to give you a reasonably shaped pulse when the silvered section passes by. Once you get a pulse the rest is handled by the pulse counting feature of digital inputs on DRB (or most other modules) and the metering side is already catered for in Cortex so there should be nothing else to do other than get the sensor to work reliably.

The SY-CR102 looks to be an active photo reflective detector i.e you supply some current to the LED side to deliver some light (not sure whether its IR or not) and then the receiver is a phototransistor which you can connect directly to a digital input (pullup resistor is already provided in the DRB). Remember to get orientation of connection correct - emitter goes to 0V and collector to digital input. The LED part can be powered from the IDRANet 12-15V line perhaps most conveniently derived from the module connector. You will need a current limiting resistor between this and the diode. The limited specs suggest a value of 20mA. I expect that this is a working value rather than max. nonetheless would suggest something between 680-1000ohms. Again beware of orientation

Since its a very basic signal you will need to also take measures to prevent incident lighting from causing false outputs. So taping the sensor to the gas meter and covering the area with black tape may help. Another possibility is that the reflectivity of the wheel in the non silvered areas is high enough to cause the phototransistor to turn on even when the silvered section is not in view. In this case you may have to reduce the current to the LED and see if you get a point where the difference between silvered and non silvered is well defined. The converse may also be true - so I guess its a case of experimentation. If you have a multimeter you can typically test the transistor's output before connecting to the DRB, i.e to see clearly that it is going 'on/off' as the silvered section passes by.

Thanks Karam -

Through a bit of browsing, I've spotted this chap; http://phk.freebsd.dk/Gasdims/n . . . Who uses an OPTEK OPB704 (which annoyingly you can buy for a couple of quid from CPC but have to swallow £5.95 postage for it to be 'handled'!) . . .

This looks a nice neat unit . . . as he's proven it's applicablity to this practice, I might order one and give it a go . . .

My Gas meter is boxed in within Node 0 so it's nice and Dark (Currently have a Webcam facing it to take snapshots every 1/4 day) so this shouldn't be an issue.

on UKHA_D recently there was a discussion on CurrentCost and I am sure when I looked at their web-page at the time there was wireless Gas meter reading device in the Envir range but now it has gone back to promising it. The current cost can be connected to a PC and whilst I haven't looked into it I believe there are a number of open source programs to process the data and produce reports/graphs.

Allan

This might work but I'm keen to keep as much as possible within Cortex / Idratek . . . I figure a Photo-resitor type approach should be fairly cheap in that the pulse counting can be handled (and therefore the cost included) within an existing Idratek module.

[Geps]

Try RS online for the OPTEK OPB704

Higher unit costs but free postage.