Detecting Air Conditioner Power Status
william chan
Posts: 1,326
Hi,
Anybody knows what would be the easiest and cheapest way to detect whether an air-conditioner is on or off?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
Anybody knows what would be the easiest and cheapest way to detect whether an air-conditioner is on or off?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
Comments
Yes, I am thinking of using SX or Propeller for this smart home project.
But then the air-con power led would be blocked from being seen by the user then... Hmmm....
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
OTHER USES:
This general-purpose LED circuit can operate at relatively high speeds making it quite suitable even for serial communications. In fact the same LED can be used for bi-directional communications. You can choose to use an INFRARED LED or any LED as long as it's the same "color" at both ends. This means that that little power indicator on the front panel of your project could be a secret transmitter/receiver. These circuits work as I have used them on vending machines to accept commands from IR remotes and transmit reports etc via IR to a handheld printer as well as standard bidirectional serial communications.
*Peter*
I do it two ways. An optocoupler connected to the compressor supply (240v), and a DS18B20 mounted in the discharge vent.
The optocoupler is a dead giveaway and pretty basic, but the DS18B20 gives me about ~25 Deg C when the compressor is not running and about 6 Deg C when it is.
That way I have insurance that the compressor actually has power (ie my solid state relay is working) and it is actually running and has not iced up the evaporator (When the evap ices I get a discharge temp of about 18 Deg C and I know it's time to service it).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
You·need a resistor on the transistor's base.· Otherwise the transistor will turn·on and the LED won't since the forward bias of the transistor is typically .6V and the forward bias of·an LED is typically 1.7V (<-depending on color and make).· Something like a 100 to 330 Ohm resistor should be fine, delivering about 10mA to 3.3mA to the transistor.
Ther's always an Opto-isolator... Aside from a relay, that would be your safest bet to avoid any ground loop problems.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
you might want to take a look at http://forums.parallax.com/showthread.php?p=762477
Mark
This is very easy to do just hook up a Optic Sensor··like 4N25·to·the ·contactor coil which is 24 volt· on your
air-conditioner·outside unit·or you can do it at air handler
At air handler
One side of the transformer which is·not·the·red·wire side·hook to the transformer
Because the red· which the one that goes to the thermostat·all of these colors· green·and white·and orange·and yellow wires·are·on the same side of the transformer··
The· yellow wire which is supposed to be the wire to the compressor contactor coil and the other wire which can black or blue· or any other color on the other side of the transformer are the two wire that you will need
This is how you can tell if your unit is a heat pump· if the outside unt come ON for both heat and cool·then it is a heat pump· unit
If this is a Heat pump
Make sure that in·HEAT·or·COOL·MODE·· that the same two wire work for both setting
·
If this not a· HEAT PUMP unit then this will only be true in the COOL MODE
There·is a very easy way to check that you have the right two wires is to use a volt meter on this if you get 24 volt when the out side unit is running and you do not have 24 volt when it is not running you have the right two wires
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
··Thanks for any·
·
·
·
·
Sam
Post Edited (sam_sam_sam) : 12/10/2008 3:15:09 AM GMT
A base resistor is usually used to limit the current which will be clamped by any transistor to around 0.6V. When a current source is self limiting there is absolutely no need for a current limit resistor. Sure, you can add one but you are never ever ever going to get very much current at all from operating an LED in photo-voltaic mode as a current source. The LED forward voltage is immaterial in this configuration. All devices exhibit some amount of ESR (equivalent series resistance).
Trying to operate the LED in purely voltage mode requires a much higher input resistance than is achievable by a simple resistor+transistor and typically an opamp is used, but the response is slow.
Remember we are using the LED as a passive sensor that generates a voltage/current in response to light not as a "current to light converter".
Have a look at how photodetectors are coupled directly to opamp inputs which is effectively a nulling input at 0V, no series resistors, no voltage.
*Peter*
·
I misunderstood... I thought you were piggybacking off of an existing LED that illuminated.· Even still, it takes a fairly bright light to generate voltage from an LED, you might need to bias the transistor some for proper detection.
·
william chan,
·
Back to the question..."Anybody knows what would be the easiest and cheapest way to detect whether an air-conditioner is on or off?" .... My Red-Neck approach would be to place something in front of the vent that activated a switch.· Properly positioned, something like this would work...
·
http://www.parallax.com/Store/Sensors/PressureFlexRPM/tabid/177/CategoryID/52/List/0/Level/a/ProductID/432/Default.aspx?SortField=ProductName%2cProductName
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
I just mentioned the LED as a sensor thing because photo resistors aren't just normally lying around and then I got carried away with OT "other uses".
*Peter*
You idea to place something in front of the vent that activated a switch isn't that redneck. When I installed my whole house humidifier, one of the elements of the control circuit was a sensor to detect airflow in the return vent to the furnace. It was a flat piece of sheet metal spot welded to a microswitch mounted on a plate that fastened it to the vent wall. You just had to line up the arrow on the plate to make sure that the air flow closed the switch. If I can find the instruction sheets and part numbers, you might be able to just by the switch. Installation is simple - cut hole in vent, mount unit in hole with arrow in direction of air flow, connect wires.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
The device you're speaking about is called a "sail switch", for rather obvious reasons. I only know because I have a
friend in the HVAC business, and I asked him what they were called :-)
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When all else fails, try inserting a new battery.
If I use a white LED, then it would be able to detect all LED colors right?
In your opinion, what is the maximum distance between the 2 LEDs for reliable detection?
Should I use a high HFE transistor?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
I know my circuit works for standard and IR leds but I don't think that a white LED would work in the same way. A green LED is good for detecting general lighting so I would go for that if that's all I had, although I have plenty of little 3mm daylight phototransistors that look like leds that I would use first I guess.
Some of the other solutions involve wiring into the unit so I would avoid that as much as possible. You don't need anything special for the transistor as any general-purpose type (2N2222 etc) will work fine. Since you don't have any control over the intensity of the A/C LED you might find that sensor will have to be placed almost on top of the LED or within 1" or so.
*Peter*
How come nobody tried detecting that?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
http://forums.parallax.com/forums/default.aspx?f=25&m=305434&g=305466#m305466
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
Yes, I think a Hall Effect sensor would be a good idea. (In fact I even referenced your project higher in the thread). But I do have a question about the particular sensor that you chose to use. Are there Hall Effect sensors that can just nuzzle up against the power cable without having to coil up the wires? For example, I've got a pocket doo-hickey that can sense AC current running through a wire that tells me it's on or not. I always presumed it was a Hall Effect sensor on the tip of it. It didn't cost very much - maybe $15, I think. Could just a plain ole Hall Effect switch of some kind do the trick?
[noparse][[/noparse]Edited: I should add that instead of placing a Hall sensor on the power cable, perhaps one could be positioned on the motor??? Maybe that would provide a stronger magnetic field but somehow you would have to find a sweet spot where the motor's own magnetic field wouldn't keep the Hall sensor tripped all the time. Is that possible???]
Mine is not a rhetorical question. I've got a freezer that kicks on in my lab and I think its magnetic field is messing with my photomultiplier tubes, so I'd like to disregard any data that I take during those times the freezer kicks on or off.
thanks for your help,
Mark
Post Edited (ElectricAye) : 12/17/2008 3:06:39 PM GMT
A simple reliable way to detect any alternating current load is on would be a current transformer. Take a small ferrite core, put the incoming line or neutral (not ground) through the core (loop it around once if you have the space) and then wind several turns of thin wire on it. How many turns you need will depend on the ferrite core, the current draw, and output voltage desired. One side of the coil goes to ground and the other through a resistor ( 1K-10K ) to an input pin. When the program detects pulses on the pin the load is turned on.
Can it work by just coiling the small wire around the current carrying wire without any ferrite core?
Not all white LEDs are phosphor based. Only those that are more than 0.1 watt are phosphor based.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my