1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Offline LED driver doesnt work properly when scope probe is connected to it.

Discussion in 'General Electronics Chat' started by Flyback, Sep 8, 2017.

  1. Flyback

    Flyback Well-Known Member

    Joined:
    Jan 5, 2007
    Messages:
    1,990
    Likes:
    34
    Hello,
    Please can you tell why the attached offline LED driver test setup makes the LED driver not work properly whenever the coaxial probe is connected to the scope. As soon as we disconnect the BNC of the coaxial probe from the scope, the LED driver works properly again.

    Do you know what is happening?

    The strange thing is that this set-up was working fine a few weeks ago and giving nice scope shots of the current waveform via the sense resistor.

    The scope is a Tenma 72-8705A….
    http://uk.farnell.com/tenma/72-8705/oscilloscope-2ch-50mhz-1gsps/dp/1836059
     

    Attached Files:

  2. cowboybob

    cowboybob Well-Known Member Most Helpful Member

    Joined:
    Oct 22, 2011
    Messages:
    3,025
    Likes:
    477
    Location:
    James Island, SC
    Check your scope DIY coaxial probe cable for a short (or some other problem with the cable).

    Might also want to check the voltages of the circuit (with the scope probe in and out of the circuit).
     
    • Like Like x 1
  3. Reloadron

    Reloadron Well-Known Member Most Helpful Member

    Joined:
    Dec 23, 2009
    Messages:
    6,778
    Likes:
    281
    Location:
    Cleveland, Ohio USA
    Try disconnecting the shield lead and only the shield lead. If the problem goes away the system does not like being grounded. If the problem goes away with the coax shield lead removed then I would suggest using both vertical channels and only the center pins and no ground. Read the manual for the scope on using the math functions for the vertical input channels. That scope should allow you to algebraically Add the CH1 and CH2 inputs. See the: I. Operating Math Functions Math functions are displays of +, -, × ,/ and FFT mathematical results of CH1 and CH2 waveform section of the scope's operators manual. Setting things up this way will eliminate the ground you are introducing.

    Ron
     
    • Like Like x 2
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. Flyback

    Flyback Well-Known Member

    Joined:
    Jan 5, 2007
    Messages:
    1,990
    Likes:
    34

    Thanks,

    I will try the math method (but remember doing it some years ago and seem to remember it giving a noisy reading?)

    Things are strange……

    The method that I depict in the pdf of the top post, is a method which worked fine when we first started doing it, but then it started going more wrong gradually over the weeks..

    When problems first started, the LED driver would not start up with the scope probe connected to it, but then would start up when we removed the BNC of the scope probe from the scope, then when we connected the scope probe back in, the led driver would stay working…..but now, whenever the scope probe is connected up, the LED driver badly malfunctions. … its input current waveform gets very distorted (we saw this with a TA009 clamp on current probe), and the power goes well above what it should be , (by 100% )

    Recently, we replaced the 1W buck controller (in the LED driver product) with a LNK3202 (previously it was a LNK302)..and we wonder if maybe the LNK3202 has a much faster switching transition and is thus more noisy. Maybe this noise is manifesting itself through the coaxial probe in the manner described?

    We would use the TA009 clamp on current probe instead, but it does not give as nice images as the DIY coaxial probe method shown in the diagram of the top post. (at least when the coaxial probe method was working)

    TA009 current clamp:
    https://www.picotech.com/download/manuals/DO086-TA009-CurrentClampUsersGuide.pdf

    We also have a TA041 differential probe, but when we connect this across the sense resistors, we get alternate 10ms half sines of different amplitude due to the leakage current in the TA041 diff probe. Also, the TA041 diff probe waveform is more noisy than the DIY coaxial probe method (when that method worked)

    TA041 Differential probe:
    https://www.picotech.com/download/manuals/ta041-differential-probe-users-guide.pdf

    We are thinking that the only way to get good current wavefroms in an offline power supply is to use a really expensive clamp on current probe. Do you agree?

    I mean, we even put a heavy differential and common mode filter upstream of the LED driver and this didn’t help at all with the DIY coaxial probe method.

    Another strange thing is, that when I scope the 14V internal rail in the LED driver with a x10 scope probe, that doesn’t cause the led driver to malfunction...even though it again involves earthing the part of the led driver circuit to which the scope probe ground connects to.
     
    Last edited: Sep 8, 2017
  6. Reloadron

    Reloadron Well-Known Member Most Helpful Member

    Joined:
    Dec 23, 2009
    Messages:
    6,778
    Likes:
    281
    Location:
    Cleveland, Ohio USA
    Something else comes to mind. Typically a scope has a 1 Meg Ohm input impedance but newer scopes offer a 1 Meg or a 50 Ohm input impedance. If this is the case with your scope and it is set for 50 Ohms that may be an issue? Differential probes and Current clamps designed for scopes are indeed sweet but also costly. What I was looking to do using both channels was create a differential input but yes, they can get noisy. I would just like to see you with a simpler solution. Some how or some way I just tend to think the scope is loading things down and it shouldn't.

    Ron
     
    • Like Like x 1
  7. Flyback

    Flyback Well-Known Member

    Joined:
    Jan 5, 2007
    Messages:
    1,990
    Likes:
    34
    Thanks, i am wondering if when we put the scope to "x1" it makes the input impedance of the scope lower? I was setting the scope to x1 because the probe is just a piece of coaxial cable, not a "standard" scope probe with a 1meg or 10meg impedance.
     
  8. Flyback

    Flyback Well-Known Member

    Joined:
    Jan 5, 2007
    Messages:
    1,990
    Likes:
    34
    The reason why we wanted to do the DIY coaxial probe method of current waveform depiction was because the TA009 clamp on curent probe does such a poor job of depicting our input current waveform, as the attached shows.
    Do you know which is the cheapest clamp on current clamp that we could buy to give us good input current wave shape depiction?
     

    Attached Files:

  9. Reloadron

    Reloadron Well-Known Member Most Helpful Member

    Joined:
    Dec 23, 2009
    Messages:
    6,778
    Likes:
    281
    Location:
    Cleveland, Ohio USA
    The TA009 seems to not do well with input compensation to the scope. Been years since I used a current clamp on and during my working years I used Tektronix stuff which was far from inexpensive. Hopefully someone more active and knowledgeable will come along with a suggestion. The approach you are using should work. I can't figure out why the scope is loading things down other than the scope ground playing a roll.

    Ron
     
    • Like Like x 1
  10. Colin

    Colin Member

    Joined:
    Sep 25, 2003
    Messages:
    374
    Likes:
    22
    Location:
    Australia
    You can always put 10k on each line to the CRO
     
    • Like Like x 1
  11. Flyback

    Flyback Well-Known Member

    Joined:
    Jan 5, 2007
    Messages:
    1,990
    Likes:
    34
    Thanks, presumably you say 10k because its a small fraction of the scope's input impedance of 1Meg?
     
  12. Colin

    Colin Member

    Joined:
    Sep 25, 2003
    Messages:
    374
    Likes:
    22
    Location:
    Australia
    And it keeps your scope well-away from any sort of interference.
    Sometimes these current-sensing resistors can upset a circuit, depending on the current peaks.
     
    • Like Like x 1

Share This Page