Page 7 of 8 FirstFirst ... 5678 LastLast
Results 61 to 70 of 80

Thread: PSU heatsinks electrified?

  1. #61
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,196
    Thanks Thanks Given 
    544
    Thanks Thanks Received 
    303
    Thanked in
    219 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    Are you measuring neutral/hot at the switch or neutral/ground?
    Ah... I get where you're coming from. Neutral to hot.

  2. #62
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,027
    Thanks Thanks Given 
    115
    Thanks Thanks Received 
    63
    Thanked in
    56 Posts

    Default

    Quote Originally Posted by jonnyGURU View Post
    Ah... I get where you're coming from. Neutral to hot.
    That will be zero, since the switch cut the hot; but I assume the heatsink measurements were made against ground.

    Neutral/ground might be the small voltage you were seeing, and that is easily measured.
    Last edited by ashiekh; 03-23-2019 at 05:28 PM.

  3. #63
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    Quote Originally Posted by Per Hansson View Post
    It's also worth mentioning that both of these have a rocker switch that only interrupts one line.
    With the Schuko plug put in one way that means it interrupts the neutral line:
    The voltage between the heatsink and ground is then 230VAC.
    At first I thought this was only leakage current but I used my Fluke 113 that has a 3kΩ input impedance:
    Sure enough it creates a spark as I touch the heatsink, so there is allot of current there.
    Therefore that test was done in the basement not connected to a GFCI outlet on the Seasonic: otherwise that would have tripped the main breaker...
    Quote Originally Posted by Per Hansson View Post
    With the PSU off on the switch at the back I still have 235VAC on the primary side heatsink because the switch only interrupts one line.
    Quote Originally Posted by ashiekh View Post
    In theory it is the live line that should be interrupted, then there should be little voltage on the primary side (the neutral wire is not exactly at ground)
    As I already explained in the earlier quotes above the Schuko plug that is very common in Europe can be reversed.
    It's also worth mentioning that it is pretty common that the monkeys that wire up the IEC plug for the PSU put the cables in reversed.
    So even if you live in a country with polarized mains plugs the cable can be wired incorrectly.

    As for the small voltage anomaly recorded by JG as you mentioned it can be a difference between phase to neutral vs phase to ground.
    But it can also be capacitive coupling in the wires, normally a meter with low input impedance is used for measuring mains wiring:
    This is because a disconnected wire will capacitively couple from a nearby live wire, and present a ghost reading.
    Fluke has some good info on it here: https://www.fluke.com/en-us/learn/bl...al-multimeters

    For example: with the PSU in my previous post I read 235VAC between heatsink and ground as shown.
    If I reverse the Schuko plug and redo the same measurement then I read between 2.2VAC to 4.5VAC depending on if I measure to ground or neutral.
    However with a low impedance meter I read 0VAC as expected.
    It's also worth mentioning here that a meter is just a resistor you put in parallel with whatever you are measuring, so the act of connecting a meter can change the circuits behavior.
    As also is evident by my previous post and the scope readings, that issue is only worse the lower the meters input impedance is of course, to the point of being dangerous for sensitive circuits.
    "The one who says it cannot be done should never interrupt the one who is doing it."

  4. #64
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,027
    Thanks Thanks Given 
    115
    Thanks Thanks Received 
    63
    Thanked in
    56 Posts

    Default

    Quote Originally Posted by Per Hansson View Post
    At first I thought this was only leakage current but I used my Fluke 113 that has a 3kΩ input impedance:.
    Sure enough it creates a spark as I touch the heatsink, so there is allot of current there.
    Therefore that test was done in the basement not connected to a GFCI outlet on the Seasonic: otherwise that would have tripped the main breaker...

    My Fluke 110 has an input impedance > 5MΩ
    http://assets.fluke.com/manuals/11x_____umeng0200.pdf

    that is over a thousand time bigger than the Fluke 113
    https://www.fluke-direct.com/pdfs/ca...113-manual.pdf


    I wonder if this is the price paid for combining functions; even the basic Fluke 101 has input impedance in the MΩ range.

    Quote Originally Posted by Per Hansson View Post
    It's also worth mentioning here that a meter is just a resistor you put in parallel with whatever you are measuring
    If that were so your 3kΩ would be dissipating 120W when measuring 600V
    Attached Images Attached Images
    Last edited by ashiekh; 03-24-2019 at 12:27 PM.

  5. #65
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    Obviously the low input impedance has a function, which if you had cared to read the Fluke post I linked you would understand.
    Or if you would have read my messages you would have understood it too, that it is to eliminate ghost readings like JG reported.
    Of course they put a PTC in series with the resistor to limit the current incase the user does something stupid, how else do you think it's CAT-III rated?
    "The one who says it cannot be done should never interrupt the one who is doing it."

  6. #66
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,027
    Thanks Thanks Given 
    115
    Thanks Thanks Received 
    63
    Thanked in
    56 Posts

    Default

    Quote Originally Posted by Per Hansson View Post
    Obviously the low input impedance has a function, which if you had cared to read the Fluke post I linked you would understand.
    Or if you would have read my messages you would have understood it too, that it is to eliminate ghost readings like JG reported.
    Of course they put a PTC in series with the resistor to limit the current incase the user does something stupid, how else do you think it's CAT-III rated?

    I'm confused; I thought we were talking volt-meters in which case we need high impedance.

    The Fluke 113 doesn't measure current.
    Last edited by ashiekh; 03-24-2019 at 02:53 PM.

  7. #67
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    For the love of god just read the damn Fluke page: https://www.fluke.com/en-us/learn/bl...al-multimeters
    "The one who says it cannot be done should never interrupt the one who is doing it."

  8. #68
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,027
    Thanks Thanks Given 
    115
    Thanks Thanks Received 
    63
    Thanked in
    56 Posts

    Default

    Quote Originally Posted by Per Hansson View Post
    For the love of god just read the damn Fluke page: https://www.fluke.com/en-us/learn/bl...al-multimeters

    You are right, but as I understood it JG measured the stray voltage when the power supply was OFF.

    Quote Originally Posted by jonnyGURU View Post
    Even with the power switch OFF, the heatsinks are still live with about .5V (half a volt) until I unplug the PSU completely.

    This is even though I can probe the AC inlet while plugged in with the switch off and get 0V. So this is stored energy.
    so I supposed no capacitive coupling ghost voltages; that is why I wondered about the neutral ground voltage.

    But yes, the Fluke article was an education for me, thanks for pointing it out; but a feature that trips the GFCI seems a little strange to me. I can just imagine myself checking the ground in a place and tripping the main GFCI; now that would be embarrassing.

    What I wonder about is inductive pick-up which won't be solved by LoZ
    http://www.jonnyguru.com/forums/show...-s-voltage-law


    And thanks again for your persistence; low(ish) impedance for voltage measurements didn't make sense till now, and to some extent still doesn't.

    Apologies that I am a slow learner.
    Last edited by ashiekh; 03-24-2019 at 04:00 PM.

  9. #69
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    "but a feature that trips the GFCI seems a little strange to me."
    It's not meant as a feature: it is just a side-effect of how it works.
    And as long as you know what you are doing and don't probe to ground you wont trip the GFCI anyway.
    And in some cases it can be a good way to test GFCI functionality in the field.

    "You are right, but as I understood it JG measured the stray voltage when the power supply was OFF."
    The whole point of capacitive coupling is that the thing you are measuring is off but you still have a voltage reading:
    It is extremely common in industrial equipment: if the fuse fails on one phase you could still be reading close to the expected voltage on this wire due to capacitive coupling from the other two wires.
    A decent load on the circuit by a meter with low input impedance takes care of this.
    However as it is a bit dangerous with sensitive electronics I do it another way: via the Fluke SV225 stray voltage adapter.
    That way I'm sure not to confuse myself, however now that meters have both modes available on the rotary switch that is even more helpful.
    "The one who says it cannot be done should never interrupt the one who is doing it."

  10. #70
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,027
    Thanks Thanks Given 
    115
    Thanks Thanks Received 
    63
    Thanked in
    56 Posts

    Default

    Quote Originally Posted by Per Hansson View Post
    It is extremely common in industrial equipment: if the fuse fails on one phase you could still be reading close to the expected voltage on this wire due to capacitive coupling from the other two wires.
    Would be interesting if JG were to measure his neutral/ground voltage (since his equipment was off)

    I must confess that for some reason inductive coupling bothered me more than capacitive; guess that is now going to change. $70 for that stray voltage eliminator is way out of budget for me.

    Whenever you say 'low impedance' my mind is thinking milli-ohms, not kilo-ohms; a source of some confusion.

    It has been an education.

    much appreciated


    P.S. you should cook up some devilish claim for April 1st to drive the likes of me crazy; as if live heat sinks was not diabolical enough.
    Last edited by ashiekh; 03-25-2019 at 10:10 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •