Radioshack IR detector LED vs Sunlight vs faint IR pulse

eagletalontim

I am trying to redo one project I did a little over a year ago and it consisted of picking up an IR pulse on the top of the electric company Smart Meter that replaced the old Analog meter. The current circuit I have right now works pretty dang good and I can calculate my bill to within $1. The problem I am having is I lose some of the meter's IR pulses due to sunlight. Nothing has been said yet, but I have electrical tape stuck all over the meter to block as much sunlight as possible. It looks a little tacky and I want to redo this pickup to look a little more professional.

Is there a SIMPLE way to make the RadioShack IR detector ignore sunlight and only detect a ~100ms faint IR pulse? I want to avoid buying any parts for this revamp and use what I already have on hand. Here are a few components that I am thinking I could use and already have.

LM1815 - Adaptive Variable Reluctance Sensor Amplifier
5v ADC
A bunch of various radioshack transistors.
Thousands of resistors all various values.
Ceramic Capacitors (.001uf to .1uf)
1 RadioShack IR detector with Chrome panel mount holder (This allows me to mount it in an enclosure and helps pick up the faint IR pulse.)
A random pack of Photo Resistors.

Ok Macgyvers! That's the list! Any help is appreciated!

Hal Albach

Here is how one guy did it.

http://hackaday.com/2011/05/13/monitor-your-homes-power-usage-on-the-cheap/

eagletalontim

I have read that before which inspired me to build my own, but I can't find schematics for it.....

kwinn

eagletalontim wrote: »

I have read that before which inspired me to build my own, but I can't find schematics for it.....

Download the data sheet (http://www.ti.com/lit/ds/soes008a/soes008a.pdf) and take a look at the schematics on it. The sensing portion of Figure 9 on page 6 (5-6) should work for your application.

The sensor is only sensitive to IR so that should reduce the effect of sunlight.

eagletalontim

I think I am going to need a better IR sensor. I have one that I hacked out of an old DVD player and it picks up my test IR pulses from a remote perfectly and ignores sunlight but it will not pick up the pulses from the meter. The RadioShack one picks up just about any light and is very finicky. Why does the radioshack IR detector pick up sunlight, but not the DVD sensor? Is there a way to tell what IR receiver I will need to get? I do have a DigiKey parts order I need to get eventually unless RadioShack has a better sensor available that I missed on my last trip.

The radioshack one has 2 leads and the DVD one has 3 leads. I have been using a piezoelectric speaker or an LED to get some feedback if the sensor picks up a signal just for testing purposes.

PJAllen

eagletalontim wrote: »

The RadioShack one picks up just about any light and is very finicky. Why does the radioshack IR detector pick up sunlight, but not the DVD sensor? Is there a way to tell what IR receiver I will need to get?

The "RadioShack one" One is a receiver, it's looking for the 38kHz carrier that remotes use.
The other is a phototransistor.
Different devices.
The meter probably isn't using the 38kHz carrier.

eagletalontim

Would there be a way to figure what kHz receiver I need or do I have to buy a bunch and hope for the best?

PJAllen

Here's a crazy idea: Take what you've got out to the meter and see what, if anything, goes

eagletalontim

I already have a circuit working right now and it has been working for over a year. Problem is, it looks hideous and electrical tape is all over the place I was hoping to make a single box that can mount above the unit and pick up the IR pulse without using electrical tape, sponge (to block sunlight) and whatever else I used to block as much sunlight as possible.

To take this project even farther, I was looking into making the pickup run from a solar charged battery and broadcast the pulse via a wireless setup to the main "brain" in my house instead of a 50ft cable running under my house up to the meter.


EDIT : I can visibly see the IR pulse on my cellphone camera and cannot see it with the naked eye. I have noticed that some IR lights emit different colors on the same camera. Some are blue, some are red, some are close to orange. Can I use this info to determine what receiver I need?

eagletalontim

Here is a screen shot of my existing software I took yesterday. It runs off a PROP connected to my computer via the serial port. The VB program I wrote sends signals to the PROP and "asks" for the data back, If the data is successfully pulled, it displays the information on the program window. Every hour, the VB program sends the collection of power usage to my server so I can view it anywhere, even on my phone.

You can also see a few really low drops in the log where the IR detector did not work for a short period of time. Usually because of sunlight at just the right angle. This would mean more sponge or electrical tape or a better IR receiver / redesign....

PJAllen

You'll have to go out there with an o'scope to see if the meter's emitter is modulated or scour the web for related info.
If it is modulated: maybe you can locate an IR receiver for that carrier freq or design a circuit to take advantage of that.
If it's not modulated: "This will mean more sponge and/or electrical tape" or other shielding.

eagletalontim

I cannot take the cover off of the meter as that is a federal offence. How would I hook up the O-scope to one of the receivers to pick up the signal and see if it is modulated?

PJAllen

Use a simple IR photo-transistor to "see" what the raw output of it is.
[Collector pulled up to +V, keep it low current, and Emitter to Gnd; scope looking at the C-E junction.]
I don't know how often those pulses or pulse-trains come along, it'll take some doing.
Should be fairly aggravating.

Take advantage of darkness by night, rig up a shade, or whatever you're up for.

Beau Schwabe

I think a derivative of the IR Heart-Beat detector circuit could be used here with some success as it takes care of auto biasing the IR. It uses the IR, in reverse-bias mode as a capacitor, the same way that a standard LED has been used as a sensor here on the forum.


http://forums.parallax.com/showthread.php/149216-DEMO-IR-Heartbeat-detector

brucee

Just so happens, I am working on something similar today. A 3 lead IR detector has the amp/filter/demod circuit built in. A 2 lead device is probably just a photo diode or photo transistor.

Both of those might be washed out in direct sun, but assuming you can shield it from direct sun, the way to ignore the background is to filter the output. IR controls send out a signal usually between 30 and 60 KHz, with 36 being the most common. So you want to take the output of the detector and run that through a bandpass filter of 30-200 Khz and amplify it to something you can read. You'll have to build a filter out of your on hand components. I found this website recently that give you great cookbook designs for filters.

http://www.changpuak.ch/electronics/calc_09.php

eagletalontim

Beau Schwabe (Parallax) wrote: »

I think a derivative of the IR Heart-Beat detector circuit could be used here with some success as it takes care of auto biasing the IR. It uses the IR, in reverse-bias mode as a capacitor, the same way that a standard LED has been used as a sensor here on the forum.


http://forums.parallax.com/showthread.php/149216-DEMO-IR-Heartbeat-detector

I think this has gotten me really close to what I need!!!!! My neighbor has been letting me use their meter as a test as mine still has electrical tape all over it. After building the circuit in the listed thread, I was finally able to pick up pulses without electrical tape and blocking as much sunlight as possible from the RadioShack IR receiver!

Beau Schwabe

Excellent !!