Wide Voltage Range LED Driver for TESTER

Discussion in 'The Projects Forum' started by rxtxau, Mar 20, 2011.

  1. rxtxau

    Thread Starter New Member

    Dec 13, 2003
    4
    0
    I'm looking to make up some "idiot proof" test devices for testing plug packs.

    What I had in mind was a single LED for Postive voltages & also a single LED for negative voltages. You just plug the plug pack in & get an immediate indication as to the polarity. AC would light both LEDS.

    The trouble is that I would like to make it as wide a voltage range as possible - from 1.2 or 1.5 volts up as high as I can.

    Some plug packs are as high as 24 & occasionally 30 volts so idiot proof would require as high a voltage as possible.

    The combination of LOW voltage operation & HIGH voltage operation seems to be the real kicker. I don't care if I have to use ICs or discretes just want to make something that will work. I took a look at various LED drivers & regulators but there are just so many of them that it is hard to find a suitable part.

    A device sold supposedly to do this with a voltage range of 1.5 - 12 volts just had 2 LEDs & 2 resistors apart from the sockets!!!

    I look forward to any suggestions.

    Thanks,
    Brian.
     
  2. nigelwright7557

    Senior Member

    May 10, 2008
    487
    71
    You need a constant current source for the LEDs.
     
  3. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    Maybe this would work.

    I've never heard the term "plug pack" before, so I had no idea what you were talking about. I've always heard them called "wall warts".
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    I agree, I didn't respond to the post because I haven't ever heard of a plug pack.

    It is worse than you know. Some wall warts are AC, there literally is no standardization with them.

    I would use a LM317 current regulator and a bridge rectifier. If you don't know what I'm talking about I'll draw a schematic.
     
  5. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    OK, I put a little thought into this, and KISS rules seem to apply. KISS = Keep It Simple Stupid.

    This circuit will work up to 40VDC (20ma), it will glow dimly at 5V(1ma), it will go down to almost 3VDC, and the glowing LED marks the positive lead. If both are glowing then the wall wart is an AC transformer (this is actually pretty common, modems seemed to like to use them back when).

    [​IMG]

    If R1 is 1800Ω it needs to be a 1W resistor for maximimum voltage. You could use qty 2 3.6KΩ ½W resistors in parallel.

    To adapt it to 12VDC max (LED full brightness) bump R1 down to 470Ω ¼W. Almost any diode will work for CR1 and CR2, but I would use a 1N4001 or better (1N4002, 1N4003, 1N4004, etc).
     
    Last edited: Mar 21, 2011
  6. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    I like it, Bill.
    I developed a Cadillac version, which has about a ±3V threshold. It illuminates the LEDs with 10mA (more or less, depending on the values of R2 and R4) from ±3V to as high as ±70V, and could go even higher with changes in the transistors and blocking diodes. Digikey has wall warts rated to 48VDC, 24VAC, and desktop units rated to 56VDC.
    I designed two versions. The Cadillac version uses resistors to bias the current sources, causing the LED currents to vary slightly over the range of input voltages. The Rolls Royce :rolleyes: version uses 1.1mA constant current diodes to bias the current sources, and its LED currents are absolutely constant over the range. It's mostly just a curiosity, just to see what is possible.
    There is another possibility using one current sink with a diode bridge in front of it, with two of the diodes in the bridge being LEDs. The LEDs must have reverse bias protection diodes in series with them, so the threshold goes up to about ±5V, and the current regulation with resistor biasing is worse that the version I posted.
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I threw this circuit together a year ago:

    [​IMG]

    Very similar to Bill Marsden's, but the 1N4148 diodes are eliminated, giving a lower threshold voltage by roughly ~0.5v at low current.

    If both LEDs are the old-style red, they will have a Vf of around 1.7mA @ 15mA. Newer high-brightness LEDs have a somewhat higher Vf, and can take somewhat more current.

    The resistor R1 would need to be increased for higher voltages, of course. As shown, 12v is the maximum. 2k Ohms would be ok for up to around 40v, but the LEDs would be very dim at low voltages.
     
  8. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    I don't think it can get any simpler than that.:)
     
  9. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    Attached is a picture of an LED I found in my junk box the other day. It fits into a 1/4" hole and gives both red and green colors, depending on the polarity. IIRC, it was about 2.1 V for green, 1.8 V for red with adequate brightness with currents in the 10-20 mA range. It would fit in a short probe with the resistor soldered to one lead. I wish I had more of them, as the usual places don't seem to stock them.
     
  10. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Here is a pic of the commercial gadget disassembled. For its cheap price it is a very handy little box to have in the toolbox.
    The basic unit accepts three common sizes of jacks, adding adapters as shown makes this more useful.

    If you want voltage ranging you could use several resistors and LEDs per polarity, so that you could also roughly gauge the supply voltage.
     
    Last edited: Mar 21, 2011
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
  12. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    ζκω∂κ∫

    :d
     
  13. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    It appears SgtWookie has promoted me to a vendor. Someonesdad is running a special on that Dialco LED right now, $20 to the first bidder. Quantity one. Shipping included to CONUS. :D

    Thanks, SgtWookie. I checked their distributor list and saw Newark and Allied, then checked Jameco and Digikey. I didn't see anything at the latter two and got distracted by something and didn't check any other places. My bad.

    Oh, BTW, studiot -- your response was Greek to me... :p But I like the device you posted a picture of. One of these days I'll make such a thing; the trick would be finding the different jacks. I've attached a picture of a handy set of plugs that came with a "universal" wall wart with outputs of 1.5, 3, 4.5, 6, 7.5, 9 and 12 volts at half an amp. Actually, the wall wart is pretty worthless, so I think I'll cut the cable off and put a dual banana plug on the end so I can use it with a power supply.

    While I'm on the subject of handy stuff, the other day I made the attached BNC to banana adapter from an old Pomona BNC to IC clips adapter that had the clips get broken off. Yes, I know you can buy such things, but the reason I'm pointing this out is that I like those banana plugs from Cal Test Electronics. They use two small set screws (and come with a 1.5 mm Allen wrench) to clamp to the wire -- no soldering. They let you make a nice cable fast. I like the rubbery material (Santoprene) rather than the typical nylon. An adapter like this is nice for those of you who like to use RF BNC cables for hooking things up and having to deal with a test instrument where the manufacturer didn't space the banana jacks at the standard 3/4 inch.
     
  14. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    To the OP:

    Don't forget, anytime 470Ω is shown this is for 12V, for higher voltages you will need a larger resistors (such as a 1800Ω). It might be worth putting a momentary contact switch for the two voltage ranges. 5V power supply will be very dim with a setup for 40V.
     
Loading...