Christmas candle lights that come on at night

Discussion in 'The Projects Forum' started by Dpetervan, Dec 17, 2007.

  1. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    I have several, but a couple stay on all day; i.e. do not go off during the day like the rest of them. As a fun electronics project, I am trying to "fix" them. I took them apart and am trying to figure out how they work. THey connect to 120 VAC, contain what looks like a photoreseistor (CdS), what looks like a transistor, one resistor, and one other small, black cylinder shaped electronic component, about twice the size of the resistor. THat is all! The bulbs are 7 watts. I would appreciate someone of experience to have a quick look and tell me how you think it works, and what the small black component is! I made an attempt at trying to draw out the schematic, and have taken a few photographs.

    For some reason I find this extremely interesting; my wife thinks I am a nerd! :rolleyes: Thank you very much!
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    in the second photograph, the device in the upper right corner is a diode, looks to be cathode down. The device in the lower right corner is a resistor. I can't quite make out the bands, but it appears to be in the megaohm range. The central device with the wavy line under the window is a CdS cell. It's a light dependent resistor that becomes more resistive when light diminishes. The device at the left side is a semiconductor in a TO-92 package. At a guess, it's a SCR, silicon controlled rectifier.

    The CdS cell lets the rectified AC voltage rise on the gate of the SCR until it fires, allowing half the AC current cycle to pass through the lamps. Unless it's a TRIAC, in which case all the AC current goes through the lamps.
     
  3. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Wow; thank you for the prompt response! I think it is incredible how clever these little circuits can be. Do you think my attempt at the schematic is accurate and makes sense?
    I would like to build this on a breadboard for trouble shooting. (I am a little leery of working with line voltage).

    Thank you again!
     
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    No, the controlled device can't be a transistor. It should either be a TRIAC or an SCR. See the reference here - http://www.allaboutcircuits.com/vol_3/chpt_2/11.html - for how these devices work. You may be able to make more sense of the circuit when you read the material.

    I can't be more specific, as the detail in the photos isn't enough to make out part number, etc. Your resistor stripes make it 5.1 Mohm, but that's all I can say for sure.
     
  5. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Thank you very much.....I will fool around with it some more. Your insight into that being an SCR makes perfect sense. I will try to build it and see what happens!:)
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Try to read the markings on the TO-92 three-legged critter, and then see if you can find a datasheet for it.

    Trying to breadboard something that runs on line voltage can be quite dangerous.

    However, since the circuit is so simple, it shouldn't be too difficult to determine which part is faulty.

    After making sure it's unplugged and removing the light bulb

    1) Set your meter on the lowest OHMS setting, likely 200 Ohms.
    2) The diode should conduct electricity one way, and not the other. If it's open both ways or shorted both ways, it must be replaced.
    Suggested replacement diodes: 1N4003 through 1N4007

    3) The resistor is of too large of a value to be read by most inexpensive multimeters. It looks like the color bands are green-red-green, which would be 5.2 million Ohms, or 5.2M.

    4) The CdS cell will vary in resistance with light. Try clipping your meter leads to it's leads, set your meter to read somewhere between 2,000 and 20,000 Ohms, and change the light level on it. If it changes readings significantly, it's likely OK.

    That leaves the TO92 device and the resistor. Resistors are cheap and available everywhere, but they usually are OK if they're not burned or cracked.
     
  7. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Great advice!....I will try and test each component separately, as you suggest. I am also reading up on "Thyristors" and "SCR's" in the textbook, to try and understand and properly draw out the schematic.
     
  8. gerty

    AAC Fanatic!

    Aug 30, 2007
    1,153
    304
    If you're leery of line voltage, and it's good to have a healthy respect, you can re-create it with low (12-24) voltage ac. You'll have to use the appropriate voltage lamps. Some of my students have no fear of line voltage(not good) and some start off with low voltage untill they are comfortable.You'll have to change resistor values as well.
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Great idea, gerty. :)

    HVAC (furnace control) transformers are available everywhere; they put out 24vac. There's always Radio Shack, where price is exchanged for convenience.

    Come to think of it, some super-bright LED's wired up with a PWM circuit to make them "flicker" could be interesting ;)
     
  10. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Thank you again for the input.....I am still working on this. I believe I do have a step-down AC transformer I could wire up into the breadboard.

    I hope to re-draw the schematic now that we think it is an SCR or TRIAC, and will post it for criticism. I am wondering if the other diode is a DIAC?
     
  11. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Does it have any numbers on it? Part numbers are usually informative.
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    That's why I like to scratch them off; it makes it more of a challenge for the next go-round. ;)
     
  13. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Here is what I have found out: The TO-92 packaged device has the markings" PCR 406 NJ6K" on it. I tried searching by part number at a number of internet electronics parts distributors and could not find a match, so I am not sure if it is a SCR or TRIAC, or something else.
    The other black component is definitely a DIODE with the markings 1N4007 DC.
    The other resistor is definitely 5.2 megohms, and the light sensitive component is a CdS photoresistor.

    Given that, I drew out the electrical connections in my updated schematic attempt (attached).

    The CdS resistor, the 5.2M resistor, and the diode all seem to test fine with my tester.

    Any thoughts on how this circuit works, and the proper way to draw the schematic?

    Thank you very much!

    Dpetervan
     
  14. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    I am not sure if my circuit drawing attachment uploaded, so here it is again.

    As mentioned, the TO-92 packaged device has "PCR 406 NJ6K" on it; the diode is "1N4007 DC", the resistor is 5.2 M, and the other a CdS photoresistor. All parts seemed to test OK, except I was not sure how to test the TO-92 device.

    Any thoughts on how this circuit works, or how the schematic should be drawn?

    Thank you very much!
     
  15. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    OK; After spending a few more hours on the internet and reading the Thyristor chapter, I have come up with what I believe to be the real schematic for the circuit (attached). I found the data sheet for the "PCR 406" TO-92 packaged component, and indeed it is an SCR.

    Am I correct in my understanding of this circuit that during the day, the photoresistor conducts and therefore short circuits the cathode and gate, cutting off the SCR. Therefore, all the current now goes through the photoresistor, into the 5.2 megohm resistor, then the diode, to the ground. The megohm resistor at 120 volts would only allow about 23 microamps current (120/5.2x10exp6), therefore probably not enough to allow the filament of the lamp to light (is this correct?).
    In the dark, the photoresistor reistance increases, allowing a voltage difference to build between the gate and the cathode, ultimately triggering the SCR and allowing it to conduct through the diode to ground, relatively unimpeded, therefore alowing higher amperage through the lamp filament, and turning it on.
    Am I correct in this?

    The other question I have is what is the purpose of the 1N4007 diode? Is it just to halve the power for the overall circuit?

    Thank you for your help. If this is correct, then I think I am all set going ahead with this project!:D
     
  16. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    The diode is there to starve the SCR of current during every other half-cycle of the AC power voltage. Once triggered, the SCR would stay on constantly if it were not for the diode.

    hgmjr
     
  17. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    RIght! Got it! Do you think my explanation and schematic is otherwise correct?
     
  18. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Your explanation of the operation of the circuit is correct.

    hgmjr
     
  19. Dpetervan

    Thread Starter Member

    Feb 26, 2007
    18
    0
    Beautiful! Thank you very much for the feedback....great to know I finally "got" something.

    You guys should be commended for an incredibly useful web project, "All About Circuits".

    Great effort!
     
  20. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    SCRs can only conduct current in one direction (unless they break down), so the diode is not there for the purpose of interrupting the current. I think the diode is there to prevent reverse breakdown of the SCR.
    The datasheet does not specify reverse breakdown voltage, although it does spec reverse gate voltage at 1V max. This spec would be exceeded if the diode were not present (shorted), but the 5Meg resistor would limit the current, so I doubt that would be a problem. I think it's there to prevent reverse cathode-to-anode breakdown, although I checked a similar NTE part, and it has symmetrical forward and reverse breakdown.
    I suspect the circuit would work just fine without the diode.
    EDIT: Teccor parts also have symmetrical breakdown characteristics.
     
Loading...