Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[shorty73]

The PA011 directions say it can sense when a load is applied, then automatically turns on.  So, once the photosensor restores power, shouldn't the PA011 sense the transformer load and turn on?

We're going to have quite a few light on the circuit, so we'll definately be over 40watts.  This is the sensor I ordered:

http://www.amazon.com/Westek-SW103CTC-4-Outdoor-Swivel-Control/dp/B000Z9DCF6

[Brian H]

Thanks for the link to the photo sensor. I found the users manual on the www.AmerTac.com site. It does appear to be able to do the load you want to control.

Local sensing requires the power to be applied to the PAO11 and then the switch on the load toggled from On to Off and back On. I am not sure if it will trigger On when power is applied with the LV transformer connected. Actual test would show if it would or would not go On.

[Backward Engineering]

I believe the PA011 is the same as the SR227 which uses a stepping relay to control the output.  I have posted a schematic of the SR227 if your interested.  I use the SR227 in my system and I can confirm that it will stay in the state it was in when the power is removed and re-applied.  To design the device to switch from off to on automatic when power is restored would be very undesirable.

http://forums.x10.com/index.php?action=dlattach;topic=9437.0;attach=2078

The stepping relay will remain in the same position if power is removed and reconnected.

Regarding the directions that state that the PA011 can sense the load, I believe that assumes the input voltage is present when the load is switched.  In your configuration, the load would be present when the input to the PA011 is re-applied

I believe you will need to connect the 115 vac light sensor to the output socket on the PA011.  The light sensor will not be able to switch on until power is re-applied to its input.  If you connect the light sensor ahead of the PA011, you will not be able to re-close the relay in the PA011 until after dusk.

Low voltage light sensors are more difficult to produce because of the higher currents they would need to switch.  I don't know if they are available or not.

Don

[Brian H]

The users manual for the SW103C,SW103CT indicates when first installed {power applied ?}. It takes several minutes for it to go Off. My guess is it is adjusting to the light levels.

That may effect where you have the Light Sensor connected in respect to the PAO11.

[Backward Engineering]

Most of the light sensors use cadmium photo resistive elements as the sensor.  The resistance is highest at low light.  They also have a long delay, using a resistor and capacitor to integrate any sudden changes in the light level.  The light sensor is probably charging the capacitor when initially powered. (adjusting to the light)

I will look at the manual on the light sensor, but for the moment I have no new ideas.

Don

[shorty73]

Tested the PAO11 in the garage; when power is restored to it after being disconnected, it remembers the previous state.  So, it appears I can either cut the transformer plug wire and put the photosensor there, or put it before the outlet and set the mini timer to send on signals every 10 or 15 minutes from about 5pm to 9pm to kick on the lights after the photo sensor restores power.

I also tried to get the controller to sense an applied load (shop drop light), but it didn't work. 

[Backward Engineering]

I tested my SR227 to see if it would turn on when the load is switched.  i noticed that it would not switch on the first time the load was connected, however, dis-connecting and re-connecting the load caused it to turn on.  I repeated the test two times with the same results.

Looking at the schematic of the SR227 that I reverse engineered, it does have load sensing.

I think I have a simple mod to an SR227 that would cause it to switch from off to on if power to the SR227 is switched on.  If the SR227 is already on, the circuit does nothing.  Let me know if you want the details.  I have not tested this mod and I don't have a spare lying around that I could test it on.  If anyone wants to modify an SR227 and test the circuit I will provide the details

Don

 

[IPS]

This is how I turn my Xmas lights on. Am sure a lv transformer and a light sensor can be incorporated in the circuit without using PA01.
I installed two LM 11 modules in a outdoor flood light fixture and instead of bulbs ,installed screw in outlets for the the lights.
Ahp macros and cm 15a control all the lights flawlessly. It also works with BVC. For the x10 modules to work properly, I mixed the light strings with at least one of the old incandescent type. It's been working for over a year.

Sometimes simple things work too.

[Backward Engineering]

I can't find any data on an LM11 module.  It sounds like it's a screw in lamp dimmer.

Putting a transformer load on a lamp dimmer module is a good way to blow out the triac in the dimmer module.  Transformers should only be connected to appliance modules, especially high current transformers.  The transformer can not handle the slightest dc component and will saturate causing high current spikes which damage the triac.  Low power transformers can be damaged by high current and overheating if the transformer should saturate due to an unbalanced waveform on the input to the transformer.  The last thing we want is an electrical fire.

[IPS]

Thank for poining it out.
Sorry for quoting the wrong number. It should read  SocketRocket Screw-In Lamp Module (LM15A). I don't know if it is dim able. It would if interesting to know if it can be used with LV transformers.

[Brian H]

LM15A uses a triac to turn the AC On and Off.
Though it does not dim. Just full On and Off.
X10 says Incadescent Loads Only, but some have had good results with some brand CFLs.
Driving a transformer maybe a problem.
They have a tendency to have the triac short from bulb burning out surges.

[Backward Engineering]

ALL ABOUT TRIACS

Most of the triacs used in lamp modules are not rated for inductive loads where the current is lagging the voltage.  Some newer triacs are better suited and specified for inductive loads, however the current still needs to reach the holding current for the device.

The waveform that supplies a 60Hz transformer needs to be free of any dc component.  The waveform with a triac switch will have a short period with zero output right after the zero crossing due to the turn on and holding requirements of the triac.  This dead time, after zero crossing must, be the same for both haves of the cycle or a dc component will be present.

Transformers normally exhibit some inductance.  The problem with inductive loads is the lagging current prevents the triac's anode current from rising to the minimum holding current during the very short time the triac's gate pulse is present.  This is also why the triac modules have a minimum lamp wattage rating.  This failure to reach the triac holding current is one of the reasons low power lamps tend to flicker.

CFL and LED lamps don't care much about the small dc components, however the current draw by these lamps is less than drawn by the incandescent lamp of equivalent  output.  This can result in problems with the triac reaching the minimum holding current.

With the standard 2 wire X10 wall switch module, the current that the module passes to power the circuit in the module, causes CFLs to flicker because the low current charges the lamp's internal inverter filter capacitor until the voltage reaches a value that allows the lamp to turn on.

The result is the CFL will not turn off completely.

[dave w]

FWIW

Silly me,

I used Socket Rockets to control inductive wall warts which fed some "flameless candles" in the bedroom used for accent lighting. It took a year but both transformers finally blew their safety link in their primaries. The thing to note is the wall wart plastic cases had melted and deformed around the plug. This happened slowly over the course of the year they successfully powered the candles.

A lesson was learned. Even when using a triac driven only in "full on" state (no dimming possible) the turn on distortion created by the triac and the inductive load, ain't good for the transformer. My switching wall warts work fine.

[Backward Engineering]

NEW SOLAR POWERED LIGHT SWITCH DESIGN

How I addressed my solar light switch issue.

I have a ceiling lamp above a stair way landing that also has a sky light.  The stairway has no windows, so at night it would be very dark.  We like to keep a low power light on at night.  The circuit has a wall switch at the bottom of the stairs and again at the top.  I had installed an automatic light sensor in the skylight area to turn the light off during daylight hours.  I recently installed an X10 WS4777 to replace both light switches.  I modified the WS4777 adding a white wire connected to neutral.

The light sensor had been installed between the wall switch and the light fixture.  I purchased a 9 watt dimmable LED bulb at Home Depot for just under $10 and installed it in the fixture.

The light sensor did not like working with the X10 wall switch.  During the day, the light sensor would turn the light off which would cause the X10 module to stop conducting (the triac would not stay on).  This would cause the light sensor to turn on again.  The result was, the light would cycle on and off every few seconds.

My solution:

I have designed a solar powered light switch to replace the existing solar sensor.  My design uses a 6 volt photovoltaic solar panel which is only 2.2" x 2.2" and will attach to a single gang switch cover.  Inside the single gang box is the new circuit.  I used a 1.5 farad super cap that is charged during the daylight by the solar panel  CMOS logic drives a MOS FET connected across a bridge rectifier to switch the light on.  The capacitor discharge current should be less than 1ua.  At that rate, the super cap should be able to keep the FET on for a week.

I have ordered the solar panel and super cap.  I will report on the project again in about a week.

[IPS]

Thanks BE for a nice write up.