User Tag List

Page 1 of 2 12 LastLast
Results 1 to 10 of 15

Thread: Xfmr Reverse Connected

  1. #1
    Join Date
    Oct 2004
    Location
    California Bay Area
    Posts
    12
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Xfmr Reverse Connected

    When a typical 150 kVA, 480V, 3PH delta to 208Y/120V, 3PH dry type xfmr is reverse connected with a 208V, 3PH, 3W primary, is there any negative impact on the xfmr because it is connected with a 3PH, 3W feed in lieu of a 3PH, 4W feed. Are there any engineering or NEC principles violated. Any thoughts.


    Michael

  2. #2
    Join Date
    Jul 2006
    Location
    Cape Coral, Fl.
    Posts
    299
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Don't do it.

    The transformer manufacturers design input impedence to limit the initial magnetizing current of the transformer. If it is wired backwards as a step-up, you will trip breakers most of the time the power is applied, not all of the time. I am sure someone else can explain why it doesn't trip every time, but my experience is... Don't do it.
    Living in the sunshine, staying contented most of the time.

  3. #3
    Join Date
    Nov 2004
    Posts
    3,172
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Don't do it!

    Quote Originally Posted by mjobee
    When a typical 150 kVA, 480V, 3PH delta to 208Y/120V, 3PH dry type xfmr is reverse connected with a 208V, 3PH, 3W primary, is there any negative impact on the xfmr because it is connected with a 3PH, 3W feed in lieu of a 3PH, 4W feed. Are there any engineering or NEC principles violated. Any thoughts.


    Michael
    Without a neutral on the primary, the wye phase voltages, and the delta phase voltages will vary wildly with unbalanced loads. Sort of like an open neutral on a 120/240V service.
    Don't mess with B+!
    (Signal Corps. Motto)

  4. #4
    Join Date
    Nov 2004
    Posts
    3,172
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Magnetizing current:

    Quote Originally Posted by boater bill
    Don't do it.

    The transformer manufacturers design input impedence to limit the initial magnetizing current of the transformer. If it is wired backwards as a step-up, you will trip breakers most of the time the power is applied, not all of the time. I am sure someone else can explain why it doesn't trip every time, but my experience is... Don't do it.
    In a 2:1 stepdown xfrmr, the magnetizing inductance of the secondary will be 1/4 that of the primary inductance, Lp. But it needs to be Lp/2. The upshot is that the magnetizing current is twice what it should be, and this saturates the core which causes a large increase in magetizing current. The breaker will trip unless the xfrmr is designed to be reversible.
    Don't mess with B+!
    (Signal Corps. Motto)

  5. #5
    Join Date
    Feb 2003
    Location
    Wisconsin
    Posts
    9,915
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by rattus
    Without a neutral on the primary, the wye phase voltages, and the delta phase voltages will vary wildly with unbalanced loads. Sort of like an open neutral on a 120/240V service.
    Many transformers manufacturers (including Square D, Hammond, and ACME) have cautions, but not prohibitions, about reverse connecting transformers. While high inrush current is one concern, voltage stability on the delta is not. In fact, the strongest warnings I have seen advise against any connection (including ground) to the X0 terminal.
    Just because you can, doesn't mean you should.

  6. #6
    Join Date
    Apr 2006
    Location
    Springfield, MA, USA
    Posts
    3,344
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by rattus
    In a 2:1 stepdown xfrmr, the magnetizing inductance of the secondary will be 1/4 that of the primary inductance, Lp. But it needs to be Lp/2.
    I don't get this. The lower voltage coil has half as many turns, and since inductance scales as N^2, the inductance on the low voltage side is 1/4 that of the high voltage side. You note this above. The applied voltage is half, so with 1/4 the inductance, the magnetizing current is doubled. But since the number of turns is half, the net ampere turns remains the same and core saturation remains the same.

    -Jon

  7. #7
    Join Date
    Nov 2004
    Posts
    3,172
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Uh-oh!

    Quote Originally Posted by winnie
    I don't get this. The lower voltage coil has half as many turns, and since inductance scales as N^2, the inductance on the low voltage side is 1/4 that of the high voltage side. You note this above. The applied voltage is half, so with 1/4 the inductance, the magnetizing current is doubled. But since the number of turns is half, the net ampere turns remains the same and core saturation remains the same.

    -Jon
    Winnie, yer right! Then, why can't we reverse feed the tranformer without tripping breakers?
    Don't mess with B+!
    (Signal Corps. Motto)

  8. #8
    Join Date
    Nov 2004
    Posts
    1,525
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by rattus
    Winnie, yer right! Then, why can't we reverse feed the tranformer without tripping breakers?
    First my disclaimer - I'm way out of my expertise here.

    The last paper I read was from GE - Can't find the number but I'm still looking.

    My translation was:
    1. The LV winding is generally next to the core, so the turns are shorter.
    2. The LV winding wire size is larger (ratio of Lv current to Hv current)
    3. And, of course, the LV winding has less turns
    4. For worse case inrush, the core saturates, inductance goes to zero, leaving only the resistance to limit the current during the inrush half-cycle peaks.
    5. The resistance of the Lv coils is significantly less (porportionally) than the Hv side. The Lv side is short length per turn, less turns, fatter wire

    So, backfeed inrush (fed from the Lv side) is significantly higher than forward feed

    Then again, could be my disclaimer applies.

    carl
    Using the code for a design guide is a sign of incompetance

  9. #9
    Join Date
    Nov 2004
    Posts
    1,525
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by rattus
    Winnie, yer right! Then, why can't we reverse feed the tranformer without tripping breakers?
    I think you can. Size the CB right up to the max allowed. Fatten the primary wire to fit. The few times I have done this, have not been a problem.

    carl
    Using the code for a design guide is a sign of incompetance

  10. #10
    Join Date
    Apr 2006
    Location
    Springfield, MA, USA
    Posts
    3,344
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Rattus,

    I think that coulter must be on to something here; I hope he can find that paper

    My point about inductance was a steady state issue, and for what it is worth I've successfully backfed a couple of smallish (50KVA or so) transformers with _undersized_ (relative to the rating of the xformer, not the load) breakers, without any problems of tripping on inrush.

    Another factor: transformers usually only have taps on their _primary_ coil in order to allow adjustment for proper core saturation. When a transformer is reverse fed you don't have any taps, and thus if the supply voltage is just a bit high, you will oversaturate the core, and get excess heating on a continuous basis, as well as greater inrush on startup

    -Jon

Posting Permissions

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