Page 2 of 2 FirstFirst 12
Results 11 to 19 of 19

Thread: OT - Need sparky for tecnical question - florescent lamp ballast

  1. #11
    Join Date
    May 2002
    Location
    SE Texas
    Posts
    11,984

    Default

    The squirrelly behavior and additional blown fuses scare me a bit. You are definitely playing with some unknowns here. I would suggest that you either stop using that lamp or if you do continue to use it, keep it at least a foot (30cm) from anything combustible. That includes sitting it on a wood shelf, as in the photo. And do not leave it on when you are not present.

    Lamp = $25

    House = $xxxxxxxxx

    Your and your family's lives = Priceless
    Last edited by Paul Alciatore; 02-08-2019 at 12:33 AM.
    Paul A.

    Make it fit.
    You can't win and there is a penalty for trying!

  2. #12
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,413

    Default

    it appears that the capacitor in question is connected across the input line wires, after the fuse. It would then be, just as you suggest, a filter on the power line, working with the common mode inductor The capacitor would short the two lines together at high frequencies. Yours seems to have shorted them at any frequency, which is why it blew off the board.

    Might have happened as a direct result of the "chattering" switch. that could have created some high voltage spikes in the common mode inductor that is in series with the input power. and that might have broken down the insulation in the capacitor.

    It seems to be an SMT substitute for C12 which was a thru-hole part on the topside, and is not present. I do not know of any SMT capacitors that are rated "X" for across-the-line use, but there could be some good for 120VAC. I assume the light carries a UL recognition..... but maybe not with that capacitor in it.

    The lack of the capacitor would not be a fire hazard.

    The fuse is a questionable value... a 25W lamp should be OK with a fuse in the range of 0.5 to 0.75 A, and was no doubt tested at (chinese) UL with that value. The fuse you have in it may be OK, or may not. All we know is that it is not the one UL used.

    Your 0.5A fuse probably blew due to the switch problem, a fast blow fuse would not like that variation, which would lead to a lot of surges and the spikes of voltage as above that may have killed the capacitor.. I'd suggest returning to a 0.5A or 0.75A fuse and using it.

    Use the smallest fuse that does not blow in use.
    Last edited by J Tiers; 02-08-2019 at 02:19 AM. Reason: fixed extra sentence that was supposed to be gone
    1601

    Keep eye on ball.
    Hashim Khan

  3. #13
    Join Date
    Oct 2013
    Location
    USA MD 21030
    Posts
    4,961

    Default

    I see a place for a C12 on the top of the PCB, while the blown cap underneath is C12A. C12 looks like it might be for an X2 rated across the line capacitor, while the small SMT capacitor is probably not rated for such service. Turning the cord switch in the wrong direction should not have been possible, unless it is damaged or defective. Yes, be careful with this. A 1.5 amp time delay fuse can hold as much as 6 amps for 10 seconds or so, and that is 720 watts at 120 VAC. More than enough for fire or explosion.

    You might want to read the voltage on the lamp itself, and then maybe get a new current regulated switching supply that can provide that voltage and current. LED lamps typically have strings of 10-20 white LEDs in series, which would be 30-60 VDC or so. For a 38 watt lamp there could be 10 x 3.8 watt LEDs, or two strings of 10 each, 2 watt LEDs.

  4. #14

    Default

    Electronic ballasts for flourescent lamps (the lamp under discussion is of the flourescent breed?) put out a high frequency AC. It will ignite the gas discharge in the lamp without having to apply high voltage spikes across it like when using the olden type magnetic ballast that operated at line frequency. Measuring the voltage level of that high frequency AC signal with a simple multimeter may not provide a reasonable result. A scope is needed....
    The ballast essentially works as a current limiting device for the gas discharge through the lamp tube. The plasma in the lamp tube offers very little electrical resistance, so a ballast is needed to operate the lamps at their rated power. One can over or underdrive a lamp tube just by using a ballast that is rated for more or less power....it is the ballast that determines the output power of the system and not really the lamp tube.

    The simple electronic ballasts also rely on detecting the filaments in the lamp tube as an indication that a lamp is actually connected to the ballast circuit. If these filaments are broken, the ballast shall not turn on. If it did turn on without a lamp connected to it, then a huge bang and transistors blown to bits is the result. The voltages on switch transistors would spike sky high due to inductive action of the "transformer" windings. Don't ask how I know
    So it would be beneficial to check wether both filaments on the tube are in tact and have the same resistance.

    I messed around quite a bit with this type of simple electronic ballasts....used to overdrive them to light planted tanks by increasing the air gap in the transformer cores and replacing switching transistors with beefier ones. It was a cheap option to get the light tubes going as "true electronic ballasts" were and still are quite expensive.

    The conclusion that I came to about fixing these type of circuits when they started acting strange and blowing fuses: cut it out, replace with a new working ballast circuit and forget about it. Don't spend a minute on trying to fix it....99% of the time it will keep acting up and burning components which cost more than a complete replacement unit from ebay in the end result.

  5. #15
    Join Date
    Oct 2013
    Location
    USA MD 21030
    Posts
    4,961

    Default

    Oops - I was thinking about LED lamps, not CFLs or fluorescent tubes. You are correct. I fixed a magnifying desk lamp with circline fluorescent bulb, using the ballast from an old burned out 9W CFL. I made a video:


  6. #16
    Join Date
    Jan 2003
    Location
    Chilliwack, B.C.
    Posts
    11,525

    Default

    The problem with the cord switch caused multiple ons and offs to the ballast in a short period of time- thus there would have been many surges happening in a short period of time. Fuses (and rectifiers) don't like repeated surges, and out they go. Capacitors have a characteristic called ripple current. If a constant charge and discharge is applied that's beyond the rating of the cap, it will heat up and blow. Maybe it takes out a fuse when it blows, maybe not, but in any properly designed circuit the parts are rated to handle only what they need to and not the unusual- which in this case was probably a large number of supply interruptions and restorations. The input cap was draining quickly during times when the voltage was interrupted, and charged quickly when voltage was restored- causing a large number of repeated surges through the fuse. Other caps in the circuit could be vulnerable to this as well, especially smaller ones. The large input cap is pretty robust, but it certainly has limits as well.

    This is little different than the advice that used to be given concerning your home stereo system. It was advised that you don't turn it on and off repeatedly, thus avoiding a problem like this blown fuse and blown cap.
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  7. #17
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,413

    Default

    The topside C12 if it was a type "X", would have been good for spikes up to maybe 1200V, and would not break down and fail from typical voltage spikes. Very questionable whether the surface mount part was of that type.

    The failure and blown fuse are explainable, and are unlikely to lead to fire an so forth once fixed. If the unit has worked for some hours, it is not likely to have a hidden problem that will cause a dramatic failure

    The wrong size fuse is a different matter, and btw, that fuseholder is not usually rated for mains voltage by UL, although it might "work".

    The fuse is there to prevent large currents in case of a failure from any cause, and you ought if possible to use the same type originally supplied, certainly not a higher rated type.
    1601

    Keep eye on ball.
    Hashim Khan

  8. #18
    Join Date
    Mar 2010
    Location
    Metro Detroit
    Posts
    257

    Default

    Thank you all for the replies.

    I have no intention of leaving the light unattended in the future. I typically use this light for ambient background lighting when I need to work in somewhat confined spaces such as inside a cabinet under a sink or in the front footwell of a car if I need to access something under the dash. The reason I shut the light off after 2.5 hrs was that it was dinner time (was not leaving the light unattended in the basement). The 1.25A slow blow was installed to see if I could duplicate the condition that caused the 0.5A to fail. I would not have suspected the switch until I installed the 1.25A and saw the lamp flicker/dim/make noise when I misused the (now former) cord switch.

  9. #19
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    30,413

    Default

    Turning wrong direction is an "expected condition of use" for those switches, and they should not do the intermittent contact deal. But, many of them will. I have seen that also.

    I would not call it "misuse", unless the switch was used in a situation exceeding its limits.
    1601

    Keep eye on ball.
    Hashim Khan

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •