Announcement

Collapse
No announcement yet.

OT electronics, zener diodes

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • OT electronics, zener diodes

    This is a follow-up to my post a few weeks ago regarding Zener diodes. To recap- I made a circuit which used 5.1 volt zeners where I needed about 3 volts. Because the current is so low at about 2 microamps, the 5.1v part was doing the job. Most of you said this wasn't the way to do it, suggesting much more complicated ways of achieving the goal- at least one of you said this would come back to bite me.

    I tested the full circuit over the next couple weeks and it always worked. I installed it yesterday, and guess what- it bit me! Of the six color controls, three of them worked perfectly- of the other three, one didn't work at all and the other two didn't have the full effect. The difference is because one of those circuits is loaded down by five 8.2 meg resistors in parallel, and the others by four 8.2 meg resistors in parallel. Not much loading, but enough to cause the problem.

    At the same time, I found that the digital voltmeter loaded the voltage down as well, and caused me to mis-understand what was happening. Once I realized what was going on, I came back to needing zeners that would work at those low currents.

    To the rescue- white leds. I breadboarded a circuit with some high value resistors and found the forward voltage to be very consistent at 2.45 volts. Adding a diode in series gave me the 3 volt 'turnover' point I was looking for. Now all channels work perfectly.

    So there I go again, using a part in a way which isn't intended- but getting exactly the action I need, and better than before. Now I hope this doesn't come back to bite me-
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  • #2
    I have used white LEDs as a voltage regulator which doubles as a power on light. It seems to work down to about 20 uA and up to 100 mA or so. There will likely be some variation with temperature as well as current, but probably adequate for many purposes, such as yours. I don't think this will "bite you".
    http://pauleschoen.com/pix/PM08_P76_P54.png
    Paul , P S Technology, Inc. and MrTibbs
    USA Maryland 21030

    Comment


    • #3
      Would I do that? Well, maybe.... NOT in a product to be made in quantity, where it just has to work the same all the time. But for a one-off such as you are making? Yeah, I admit it, I have done that sort of thing, and it worked OK.

      The GOOD news here, is that data sheet or not, you are at least using an intrinsic property of the parts.... forward voltage. Yes, it will likely vary a bit over temperature, but probably not enough to be a real issue.

      The difference is the use of a "secondary" (or worse) characteristic, in your case, apparently leakage current, something that is neither closely controlled for in testing, nor dependent on an intrinsic property. Those I have seen come back to bite many times.

      You found it quickly, you have a solution that should work acceptably because I do not think your circuit depends on extra-close control of that voltage. Good work.

      CNC machines only go through the motions

      Comment


      • #4
        It does sound like a better solution than your other, zener diode one.
        Paul A.
        SE Texas

        And if you look REAL close at an analog signal,
        You will find that it has discrete steps.

        Comment


        • #5
          Sorry, I dont know the whole Zenor Dia deal. Out of the loop.

          I do know what Zenors are good for.

          JR

          Comment


          • #6
            Thank you gentlemen. I keep learning-

            As part of this stage upgrade, I'm repairing some burnt out led fixtures. They call these chips 10 watts, and I'm running them at about 9 watts. I've had five failures over the last couple of years, out of 17 that were in use. Seems a high failure rate to me, or perhaps I'm being optimistic about the power handling capacity of these chips. At any rate, I have an operating lighting system again, yea!
            I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

            Comment


            • #7
              Originally posted by darryl View Post
              Because the current is so low at about 2 microamps, the 5.1v part was doing the job. Most of you said this wasn't the way to do it, suggesting much more complicated ways of achieving the goal- at least one of you said this would come back to bite me.

              I tested the full circuit over the next couple weeks and it always worked. I installed it yesterday, and guess what- it bit me!
              Yeah, I like you as well as the other EEs arounds here. Yeah, simple miss=spoke unless you meant it?

              "Because the current is so low at about 2 microamps, the 5.1v part was doing the job. Most of you said this wasn't the way to do it, suggesting much more complicated ways of achieving the goal- at least one of you said this would come back to bite me.

              I tested the full circuit over the next couple weeks and it always worked. I installed it yesterday, and guess what- it bit me!

              Hey! If you are not pushing then what is the point?

              I only had an issue with the English.

              "Because the current is so low at about "2 microamps"" we cant work. JR

              Comment


              • #8
                Strange things happen at microamps-
                I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

                Comment


                • #9
                  I bet the led is not very bright at 2 uA!
                  Location: Helsinki, Finland, Europe

                  Comment


                  • #10
                    The LED might be bright enough to see it in a darkened room. I had forgotten that the device was used as a series voltage dropping means, which might be more problematic than as a shunt regulator as I used, but it is likely that leakage current will be very low under threshold. It's difficult to find specifications or even typical values for LEDs so far from usual applications.

                    https://www.electronics-notes.com/ar...cteristics.php

                    The following research paper may "shed some light" on LED characteristics, but it may take greater mathematical proficiency than I have:

                    https://www.researchgate.net/publica...emitting_diode

                    .
                    http://pauleschoen.com/pix/PM08_P76_P54.png
                    Paul , P S Technology, Inc. and MrTibbs
                    USA Maryland 21030

                    Comment


                    • #11
                      Originally posted by PStechPaul View Post
                      The LED might be bright enough to see it in a darkened room. I had forgotten that the device was used as a series voltage dropping means, which might be more problematic than as a shunt regulator as I used, but it is likely that leakage current will be very low under threshold. It's difficult to find specifications or even typical values for LEDs so far from usual applications.


                      .
                      I can't remember exact numbers but based on my experimentation one AA lithium is enough to power white CREE XP-E power led for 30 years at brightness level that is enough to read a book at 2-3 alphabets at a time

                      Not sure if my eyes are still able to see with so dim lighting after 29 years tho.
                      Location: Helsinki, Finland, Europe

                      Comment


                      • #12
                        My test consisted of running 4 volts through a 130k resistor in series with a 390k resistor in series with the led. Since the voltage drop from the led is 2.45 volts, that leaves 1.55 volts to drop through the two resistors in series- which becomes 520k. According to ohms law, that means the test setup has- wait here while my calculator does the math- wait some more while I throw my calculator out and use the computer one- 2.9 microamps. And yes, with the room lights out and with the magnifier, I can see light emanating from the led. It actually has a blue cast- looks more like a uv led than a white one. With more current it goes white.

                        Ok, now my calculator is bugging me- I hate to throw things out, but this one has its digits going crazy. It's going outside to sit in the sun for awhile.
                        I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

                        Comment


                        • #13
                          Most white LEDs are actually blue, with phosphor that glows white. Then there are RGB LEDs which can produce many shades of color including white.
                          http://pauleschoen.com/pix/PM08_P76_P54.png
                          Paul , P S Technology, Inc. and MrTibbs
                          USA Maryland 21030

                          Comment


                          • #14
                            Originally posted by darryl View Post
                            My test consisted of running 4 volts through a 130k resistor in series with a 390k resistor in series with the led. Since the voltage drop from the led is 2.45 volts, that leaves 1.55 volts to drop through the two resistors in series- which becomes 520k. According to ohms law, that means the test setup has- wait here while my calculator does the math- wait some more while I throw my calculator out and use the computer one- 2.9 microamps. And yes, with the room lights out and with the magnifier, I can see light emanating from the led. It actually has a blue cast- looks more like a uv led than a white one. With more current it goes white.

                            Ok, now my calculator is bugging me- I hate to throw things out, but this one has its digits going crazy. It's going outside to sit in the sun for awhile.
                            Well thats odd. I came up with 520k also. Where is the loss? Always a loss, no Matter what? JR

                            Comment


                            • #15
                              Originally posted by PStechPaul View Post
                              Most white LEDs are actually blue, with phosphor that glows white. Then there are RGB LEDs which can produce many shades of color including white.
                              Nice. Do you know why they have to be coated? Hint. Florescent tube bulbs.

                              Take the internal coating away and the yuppies might buy them for Tanning bulbs! I Chit you Not. Idiots. Oh well.

                              You folks stay safe and look out for others. Keep them safe also. JR

                              Comment

                              Working...
                              X