Announcement

Collapse
No announcement yet.

Inexplicable voltage readings in LED circuits

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

    Inexplicable voltage readings in LED circuits

    Hi Guys, here is something I see often when working on LEDs in series wit ha 12V or 24V supply. I may see it other times as well but thinking about it, I only recall this in LED circuits.

    What basically happens is my meter reads reverse polarity voltages that seemingly can't exist.

    Have a look at this video - this is from a new Youtube channel I started a couple weeks ago.

    If you like, watch the whole video through, even let me know what you think of it , but if you don't have time just play it from 14min 30sec and you will see exactly what I mean, as this weird issue is a bit hard to explain. They say 'you just had to be there' LOL.

    https://www.youtube.com/watch?v=srDjOX1_MfE&t=887s
    Follow me on YouTube
    ------------------
    Learn Electronics Repair
    https://www.youtube.com/channel/UCFX...R8UZ2vg/videos

    #2
    Re: Inexplicable voltage readings in LED circuits

    usually when I see weird voltages it's because my reference is not what I think it is...

    Whoever made that sign should be made to fix it... oh boy... cant believe it worked from the get go?

    Comment


      #3
      Re: Inexplicable voltage readings in LED circuits

      lol
      is that matrix acting as an antenna???
      i see the reds are 5 series, how many on the blue - 3??

      Comment


        #4
        Re: Inexplicable voltage readings in LED circuits

        That was the complaint I had, whoever wired it seems to just haphazardly wired it.

        5 red 3 blue is actually somewhat expected trying to even out their voltage drops.

        Comment


          #5
          Re: Inexplicable voltage readings in LED circuits

          You will get weird readings after unplugging power because the bus voltage doesn't fall to zero.
          It's around a 12VDC PSU and at say 6V nothing is there as a load, the LED's are off... until a multimeter comes along and turns one on and the O/P cap discharges a little more.
          I didn't see anything odd about his measurements, guy's technique he's spazzing out because the Fluke reads nothing for Vf with blue LED's - totally normal.

          I have seen mystery ghost voltages, a strange -ve DC voltage and I attribute it to rectification of AM radio or AC hum. The LED sign is an antenna+detector so you could see a volt with nothing connected.
          Also like a 9V battery powered device reading -1.5V on a transistor that is not used kind of thing.

          Comment


            #6
            Re: Inexplicable voltage readings in LED circuits

            @redwire
            The guy was me making the video, didn't realize I was 'spazzing out'.... I also corrected that statement about the blue voltage drop later in the video if you watched it through - when i found the real cause of the problem.

            The weird voltage was with the sign turned on (before I fixed the 'wire in the wrong place' fault) was that I read 17V across the LEDs on that string when I only have 12V coming in, and if I reverse the meter leads I read around MINUS 7V across the same LED string with reference to ground!! That's impossible isn't it?

            I confirmed this on my other two DMMs, OK so I read a different minus voltage on those but it is still minus with respect to 0V on the PSU. That's why I can't figure out how it happens. And I have seen this effect regularly before on LED strings with an open circuit LEDs somewhere in the chain.

            The 'weird thing' after unplugging the PSU (yeah sure the PSU caps hold some charge, I do know that) is that I see MORE voltage across that wrongly wired LED string than I have coming in off the PSU! Surely that is just.... well.....weird.

            The rectification of RF or mains hum sounds reasonable - as I think I only ever see this in open circuit LED chains, which of course are diodes


            @stj - if you watched the whole video you will see there is 5x red and 3x blue in a chain.
            Last edited by dicky96; 04-08-2021, 11:58 AM.
            Follow me on YouTube
            ------------------
            Learn Electronics Repair
            https://www.youtube.com/channel/UCFX...R8UZ2vg/videos

            Comment


              #7
              Re: Inexplicable voltage readings in LED circuits

              AC and DC offsets also cause funny readings with meters too.

              Comment


                #8
                Re: Inexplicable voltage readings in LED circuits

                Sorry lol it was pressing the continuity button and the Fluke's timing 15:08 and yours were not going well, back and forth, I got confused as what was going to be measured next.

                With your testing I see you encountering two things that are common and can make you shake your head. It runs off 12V and you measured over 17V, free-energy is now yours.

                The mystery -ve voltage: Any weak source of AC (RF as in AM radio, or AC hum) and a floating diode (LED or transistor with one leg not connected) will rectify that and a multimeter is high impedance input so the sign-board and multimeter have a difference in potential that shows up. It's very low current but seeing -V does make you wonder what is going on.
                Part of the main cause is capacitive-coupling between the DMM and leads and bench or ESD mat, to the sign-board, getting rectified.
                Second, is the wallwart- it is not PE grounded and likely floating at a potential due to the Y-capacitor. You can measure ACV between the ESD mat and PE ground, and sign-board power to find the mystery source. It's only uA.

                I have seen LV power might even be 12VAC and the LED banks are wired to light on opposite AC cycles, ancient chinese secret to half the power needed, yet annoying flicker.

                With power on, a quick method is to briefly just bridge (short) each LED in the dead string with a multimeter probe to find the culprit, or use a DMM with a Lo Z function to stop ghost voltages.

                Comment


                  #9
                  Re: Inexplicable voltage readings in LED circuits

                  Cheers redwire I'm happy you watched the whole video now

                  Your explanation is a good one - come to think of it, I only started to come across this weird voltage reading effect once I put an ESD mat on the bench, which you can see I am using in the video.


                  I'm also powering the DUT on an isolation transformer (always) since around the same time.

                  Do you think either or both of these factors are coming in to play here?

                  I'm pretty sure the original PSU is giving DC not AC (though I understand your comments around that issue). I could always try power it off my bench PSU to be sure.

                  Is there any way you know to replicate this effect and prove the cause? I would be very interested in making a follow up video about that as I think a lot of 'learners' and more experienced folk would be interested in that

                  Cheers
                  Rich
                  Follow me on YouTube
                  ------------------
                  Learn Electronics Repair
                  https://www.youtube.com/channel/UCFX...R8UZ2vg/videos

                  Comment


                    #10
                    Re: Inexplicable voltage readings in LED circuits

                    Most ghost readings are due to capacitance between mains and the object. Either stray capacitance, or intentional Y-capacitor (which are added to pass EMC tests). 240VAC 50Hz into a 2,200pF cap is about 0.131mA small enough capacitance to not cause a shock but enough that a multimeter will pick it up.

                    You can prove/disprove what I'm saying by taking some voltage measurements. Put your multimeter on ACV with one probe connected to a known good PE ground, and with the other probe poke around.

                    In the case of two-prong wall adapters/laptop bricks, their Y-capacitor floats the output up to high voltage at tiny current. This is always big time troublesome, poor Arduino crowd wondering why their sensor is noisy.
                    In the case of three-prong wall adapters/laptop bricks, their Y-capacitor is between their output and PE and typically the output floats but has a 1MEG resistor to PE ground. You can verify with an ohmmeter.

                    Your ESD mat (if not grounded) will go up in voltage if you place a mains cord on it.

                    KSGER soldering station likely does NOT have the aluminium enclosure earth-grounded.
                    See what you measure for continuity from tip/case to PE GND. People take them apart and add a ground wire to the case (at the encoder tabs) and also connect a thicker PE wire to the 24VDC(-) there is a pad on the controller board for this provision, the PCB trace is skinny.

                    The majority of isolation transformers have a lot of capacitance between their primary and secondary windings. So the output is galvanically isolated but will also have an oddball common-mode AC voltage present due to the capacitance. Best to measure it.

                    For your sign-board, I think the AC adapter has a common-mode AC voltage present, it was not referenced to earth-ground like your ESD mat or soldering iron.

                    Comment


                      #11
                      Re: Inexplicable voltage readings in LED circuits

                      Thanks again redwire

                      As I mentioned the sign board PSU was on my isolation transformer - it's a big one too, 2KVA. I always have the DUT isolated.

                      My ESD mat is connected to one of the ground terminals on my Bench PSU (the green terminal between the red and black voltage output).

                      The bench PSU is a linear one dual, 0-30V 0-3A plus 5V @ 3A. I am sure that is grounded. It's quite an old one, I bought it second hand at a Ham Radio Rally in the UK some years ago. I can't see any pics of this now (it seems they are no longer sold) but you can see it here at 31mins: 5 secs and if you look close you can see the yellow banana plug from the ESD mat connecting to one of the ground terminals.

                      https://www.youtube.com/watch?v=lutL4wft6fQ&t=446s

                      I am pretty sure my KSGER is grounded - it's one of the ones with the 5-wire connector to the iron and I was aware of this earth problem when I bought it so looked for the grounded version. But to be honest I never checked with my meter.

                      But one thing I noticed since I started making videos and use a microphone headset I can hear a little click/crackle noise when I put the soldering iron onto the DUT. I can't hear this physically but I can hear this quite clearly on my recordings. This is worrying me now, especially soldering things like motherboards, though i don't know if i hear the crackle on those, I have to check..

                      Could this be damaging the device I am trying to repair? I have long believed every bench should (1) have a grounded ESD mat and (2) have isolation for the DUT. Now I am becoming not quite so sure.

                      Also watching many youtube vids etc it seems I am in the minority having a grounded ESD mat. Why is that? Is all this ESD considered by most to be just a scam to sell mats and wrist straps?

                      Even I don't wear the wrist strap though, but I am aware of keeping myself in contact with the mat, and to be quite honest, here because all year round is cotton shorts and tee-shirt weather, and carpets do not exist, it's not easy to build up a static charge.
                      Last edited by dicky96; 04-10-2021, 04:12 AM.
                      Follow me on YouTube
                      ------------------
                      Learn Electronics Repair
                      https://www.youtube.com/channel/UCFX...R8UZ2vg/videos

                      Comment


                        #12
                        Re: Inexplicable voltage readings in LED circuits

                        i would mod the ksger buy making sure the mains earth is connected to the iron via a 1meg or 2.2meg resistor.

                        that way your earthed but wont get a boom if you touch a live circuit or charged capacitor.
                        and trust me - that second option can blow the end of the tip off your iron - it happened to me in the past!
                        big 10,000uf+ caps hold a lot of current!

                        Comment


                          #13
                          Re: Inexplicable voltage readings in LED circuits

                          Yeah that makes good sense STJ
                          Follow me on YouTube
                          ------------------
                          Learn Electronics Repair
                          https://www.youtube.com/channel/UCFX...R8UZ2vg/videos

                          Comment


                            #14
                            Re: Inexplicable voltage readings in LED circuits

                            To be honest I don't recall picking up RF to be a large component of multimeter "noise" - maybe many millivolts, but not 5 volts.

                            Another thing worth trying is putting say a 10K ohm resistor in parallel with the meter and see if the reading still shows up, it'd at least give an idea of the impedance of the source of the anomalous source is - like it will snuff out RF noise as a potential source.

                            Likely it's something to do with the power supply...

                            Comment


                              #15
                              Re: Inexplicable voltage readings in LED circuits

                              I had a spat where new chips I soldered in died after running only a few minutes. Likely chinese semi's (SICSTOCK/liaoxiyuan) but I checked my workbench ESD mat and found it was not grounded. The connection at the snap was bad. The ungrounded mat measured around 50VAC because mains power cords were lying on it.
                              It's a very low leakage current but semi's have micron thick glass, MOSFET's and other ICs semi's are stressed with a 50V zap, so it's good to check your bench grounding, every year as well.
                              In Canadian winters the humidity gets so low, static electricity gets nasty just walking and sitting in an the lab chair makes many kV so I'm used to first touching the ESD mat to discharge. I rarely use a wrist-strap or ESD lab coat unless it's expensive and critical work.
                              I added LED's to my mat for fun, maybe try it out: Adding LED's to ESD mat and wrist-strap grounding

                              All my soldering stations are hard earth-grounded. It's part of electrical safety codes, they come that way (except KSGER supposedly double-insulation) so if the mains transformer fails pri-sec short that you won't get zapped.
                              If you want a soldering iron isolated, I'd use a 1MEG and 10-22nF cap in parallel, to earth ground. That cap swamps the KSGER's 2,200pF Y-cap Y1 in KSGER 24V PSU schematic, which can make a potential and EMI at the tip. It might be what your mic's are picking up is the SMPS in the soldering iron, assuming the mic's electronics are earth-grounded somewhere.
                              My bench power supplies are the same, floating with 1MEG and 10-22nF cap. to PE, most have that built in. Always measure them e.g. ACV from PSU output to their green binding post to see if they don't float up to something bad.

                              I'd just spend some time measuring ACV and resisitance between earth-gruond and some lab equipment to make sure all is well.

                              Comment

                              Working...
                              X