Here's one for the experts, hour metering an AC bulb

Discussion in 'General Electronics Chat' started by trader007, Jun 28, 2015.

  1. trader007

    Thread Starter Active Member

    Feb 27, 2010
    222
    19
    So... yeah... the first one we tried worked great.

    What we have is a UV bulb running off a 12vdc power inverter... from what my meter says, it oscillates at 26khz on the output to the bulb. I don't know the voltage, but that seems irrelevant.

    The hour meters we are using are ones we won't be finding a datasheet for- http://www.ebay.com/itm/Digital-LCD...799?pt=LH_DefaultDomain_0&hash=item4195014997

    Like I said, the first one we tried did work fine. The bulb's wire, either one, did not even have to be in the hour meter slot for the induction to be detected and the hour meter counted great.

    But everyone we try since then doesn't work. The meter sees the signal, the little counter goes HIGH/LOW when I physically move it close and away from the bulb's power wire, but if I hold it still it stops counting.

    My best guess is the hour meter is made to see up to a 15,000rpm signal, because its made for gas engines... and since the bulb is getting a 26khz power feed the meter simply sees it as an always HIGH signal.... But that doesn't explain why the first hour meter we tried DID see the oscillation and counted fine. I fear that even though these hour meters all look the same, they aren't... and since they are encased in epoxy there is no way to tell what the guts are made of or what the specs are.

    Anyone have any idea what I'm talking about, and maybe a possible dirty fix to solve this issue?
     
  2. #12

    Expert

    Nov 30, 2010
    16,325
    6,818
    Here's one of the specs: Designed to detect pulses in a spark plug wire. That's about 40,000 volts of change with a very small duty cycle. Why does that not resemble a 12 volt DC power line with, at most, a few tenths of a volt of ripple?

    No schematic, no photos, no written description of the circuitry or any principle of operation.
    I'm stumped.
     
  3. #12

    Expert

    Nov 30, 2010
    16,325
    6,818
    Good evening HP. :) I think you got your AC and DC mixed up. He's trying to get a DC feed from a 26KHz DC converter to trigger a sensor designed for 40,000V pulsed.
     
  4. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    Yet he calls the 'sensor' an hour meter???:confused::)
     
    #12 likes this.
  5. #12

    Expert

    Nov 30, 2010
    16,325
    6,818
    You can only understand the situation if you go to the eBay page and read up on the device. It's an hour meter for internal combustion engines. It is supposed to sense the spark plug wire.
     
  6. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    Oooops! -- I missed the 'connect to spark plug wire' 'feature':oops:

    So it would seem the problem will be insufficient EMF and, perhaps, D/C related...

    Best regards
    HP
     
    #12 likes this.
  7. #12

    Expert

    Nov 30, 2010
    16,325
    6,818
    The problem is also question related. The original post is so difficult to simplify that nobody would touch this one for three (3) hours.

    Better questions get better answers, quicker. :)
     
    Hypatia's Protege likes this.
  8. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    +1000:D
     
    wmodavis and #12 like this.
  9. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    @trader007

    Assuming I now understand your application: The desired result may be achieved via connection of a 12V DC hour meter to the input of the inverter such that the meter is energized only when the inverter/lamp is operating...:D

    Best regards
    HP
     
  10. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    You need to learn how to become a mind reader, a lots of newbies need you to be ... :D
     
    wmodavis, Hypatia's Protege and #12 like this.
  11. #12

    Expert

    Nov 30, 2010
    16,325
    6,818
    I'm still confused! It seems the 12Vdc inverter is changing 12V DC to 26KHz at an unknown voltage to run a UV bulb.
    A poorly written question can take hours to decipher! :confused:
     
  12. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    FWIW I take the OP to mean 'inverter' in the generic sense of the term (i.e. power Oscillator, etc..) as opposed to an off-the-shelf "DC to Mains" converter... But then my 'clairvoyance' on this thread has been lacking...:oops:

    Best regards
    HP
     
  13. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,147
    204
    I would not even go that route. There are some hour meters that require a contact closure to ground to count or for that matter a transistor switch work. I'd get one of those and an appropriate phototransistor. Been there. Done that. Just not UV.
     
  14. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    Owing to the disparity of required precision, it would be helpful to know whether the intent is to merely 'track' lamp life -or- to meter exposure time (as for 'tanning', (UV)EPROM erasure, photochemical processes, etc...)
     
  15. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,147
    204
  16. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    My point exactly!:D 6 Min. is rather 'coarse' resolution for exposure timing (in any application I can imagine) moreover, inasmuch as discharge lamps seldom die unexpectedly, it seems pointless to track their 'age':confused:

    Still... 'tis a 'rum old world' hence I'm always prepared to hear 'all sides'!:D:D:D

    Best regards
    HP:)
     
    Last edited: Jun 29, 2015
  17. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    To this I would add:

    Inasmuch as the sensitivity of photo-semiconductors falls off precipitously below 400nm, I recommend application of (low 'persistence') fluorescent material to the 'lens'... Please note that many common substances exhibit strong UV-excited florescence -- especially toward the low energy (i.e. long wavelength) 'end' of the 'band'...

    Good Luck!:)
    HP
     
  18. trader007

    Thread Starter Active Member

    Feb 27, 2010
    222
    19
    lol, sorry guys, I do not know much about the circuit or meter! the circuit was made by some wahoo and and meters are from china. The inverters have all the model numbers scratched off the chips, and the meters do not have datasheets. So I am running in the dark too.

    Its a 12vdc power inverter, that outputs likely 300vac or so, but i do detect 26.5khz... now, again, I can measure the voltage and get back to you, but I don't think that matters at this point.... because the hour meter IS detecting the induction off the bulbs wire. The problem is that its apparently not detecting the osculations in the wire- just like a gas engine, the hour meter needs to see a HIGH/LOW pattern to determine the engine is "ON" and it needs to starting adding up the hours.

    Whats weird is the first hour meter we tried DID work great. But now we ran into the next dozen or so that don't. I am trying to get more data to determine if the power supply has changed, the bulbs have changed, or the hour meters have changed.

    I was just hoping someone would have some idea about these chinese hour meters, and how i might be able to modify something to get them to work with these bulbs. Again, I am guessing these meters are only designed to detect gas engines, which would be an A/C signal up to 15krpm. My A/C lightbulb "flashes" at 26khz, so the meter simply isnt thinking the "motor is running"
     
  19. trader007

    Thread Starter Active Member

    Feb 27, 2010
    222
    19
    problem is the meters we already custom fit for the product we produced, and there isnt room for a different hour meter. We ran multiple tests before we decided to use them, but we didnt run multiple tests with meters from different sellers... I have a feeling we got a few different batches of meters even though they all look the same. These are encased in epoxy, so its impossible to know whats inside them or how they really work other than they just detect an induction A/C signal to signal a counter...
     
  20. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    1,147
    204
    Aside: HP
    The discharge lamp (white anyway) spectrum changes with age. It looses a lot of the blue. The one I used required 22 V @ 45 A to run (about 1000 W) and a 40 kV pulse to start. It cost about $700 USD and needed replacing at 1000 hours. We always had one lamp "on hand". Our process was "spectrum sensitive". It took at least 2 hours (me) to change the bulb. The lamp is pressurized to about 15 Atmospheres, so it's like holding a small bomb.
     
Loading...