Page 1 of 3 123 LastLast
Results 1 to 10 of 24

Thread: a little OT- phase shifted transformer

  1. #1
    Join Date
    Jan 2003
    Location
    Chilliwack, B.C.
    Posts
    10,958

    Default a little OT- phase shifted transformer

    Had a thought today about rectifying ac voltages- this concerns single phase only. Normally you put a pair of rectifiers or a bridge rectifier on a secondary winding, then each half cycle you get a voltage peak and a drop to zero, repeat for alternate voltage, repeat again, etc. The result is a dc voltage on a filter cap. Maximum current flow is at the peak of the voltage, zero current flow at zero voltage. Not exactly true, but close enough depending on how many cycles it takes to bring the capacitor voltage up. The capacitor supplies the load current during the low to no voltage parts of the cycle- that's your typical transformer/rectifier/filter cap power supply. What if you added a second transformer, but fed it through an ac capacitor to provide a phase shift. It would have its own rectifiers, but would feed the same filter capacitor. Then you would have a second charging current source to feed the capacitor during the low to no voltage parts of the cycle of the first transformer. That should ease the requirements for all the parts involved and reduce the hum on the output during high current draws on the power supply.

    What if you added a third transformer and could wire them in such a way that you had pseudo three phase on the outputs- I don't know if this is even possible. The goal would be to have two or three 'power points' for each half cycle of the input voltage instead of just one- dropping the need for high capacitance on the output, easing the job for the rectifiers, and lowering the ripple voltage as well.

    Just an idea, which I know would work, but requiring a phase shift between transformers which I don't know how easily that could be done. I know that you can get 3 phase out from a vfd, but you have to start with single phase rectified anyway, so that's not the goal here. The goal is multi-phases coming out of rectifiers feeding a single output cap to give all the advantages listed above.

    Yes, it's more parts, but in many cases it's easier to find transformers for lower power levels than for higher power, and simplifying rectifier and filter requirements would be worth something-

    Comments?
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  2. #2
    Join Date
    Aug 2002
    Location
    Albuquerque
    Posts
    2,811

    Default

    All it will do is save on the cap. size at the expense of iron. And of course the cap. to cause the phase shift is still going to be required. Three phase power with full wave (ie. six legs ) is the real way to go, if the 3 phase is there to start with.
    ...lew...

  3. #3
    Join Date
    Apr 2009
    Location
    Ashburton, near Christchurch New Zealand
    Posts
    6,268

    Default

    This sort of question has exercised the brightest minds from the days of the first Niagara power station when Scott-T transformers were used to produce 3 phase from a 2 phase supply. I dont know if that method can produce 3 phase from single phase.

  4. #4
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    25,840

    Default

    The 2 phase and 3 phase are entirely equivalent.

    Single phase is not equivalent to either.

    It is, at least theoretically, possible to make single phase from 3 phase. I have done that. It is not particularly effective, but it is possible. The method did involve a Scott- T connection, and some other parts.

    The reverse is possible also. All the parts in the system are linear and bidirectional, therefore they will do the transformation either direction.

    It is not particularly effective, as it needs a lot of continuous "tuning". not worth it for a rectifier.

    IF you start with 3 phase, it can be effective for high power, as with the regular 3 phase you get 14% or so ripple, with no filter. With another delta-ẃye transformation, you can cut that roughly in half. There is even a scheme for 3 transformations that gives very pure DC indeed, at ashigh a power level as you can possibly want.
    1601

    Keep eye on ball.
    Hashim Khan

  5. #5
    Join Date
    Oct 2013
    Location
    USA MD 21030
    Posts
    3,874

    Default

    A better solution is to use a power factor correction circuit to create DC from an AC source. It is essentially a boost converter that provides maximum boost when the line voltage is low and minimal boost at the peak, so it draws nearly constant current over most of the waveform. Because the boost converter (PFC circuit) operates at high frequency, it can utilize relatively small inductors and capacitors.

    http://www.infineon.com/dgdl/Infineo...4a62c75a923b05

    http://www.newark.com/pdfs/techartic...onHandbook.pdf

    https://www.eeweb.com/design-article...tor-correction

  6. #6
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    25,840

    Default

    Actually, in the ideal case, it draws current proportional to voltage, so it presents a load like a resistor. You get maximum power capability from the line.

    But the rectified voltage MUST be larger than the peak of the AC so there are limits as to applications. some related topologies can produce other voltages, but they too have limits.
    1601

    Keep eye on ball.
    Hashim Khan

  7. #7
    Join Date
    Apr 2009
    Location
    Ashburton, near Christchurch New Zealand
    Posts
    6,268

    Default

    You could possibly achieve transformation of voltage, phase and frequency of whatever you like by building electrical machines such as motor generator sets to suit. There would be losses involved.

  8. #8
    Join Date
    Jan 2008
    Location
    Pittsburgh PA
    Posts
    81

    Default

    Quote Originally Posted by J Tiers View Post
    The 2 phase and 3 phase are entirely equivalent.
    Actually this is incorrect. In a two phase system the phases are shifted 90 degrees from each other.

    In a three phase system the phases are shifted 120 degrees from each other.

    So there are three animals: single phase, two phase and three phase. Neither is equivalent to the other.

  9. #9
    Join Date
    May 2016
    Location
    Michigan
    Posts
    309

    Default

    Ripple reduction by phase shifting single phase is not worthwhile.
    However it continues to be the preferred method for 3 phase where phase shifting to provide 6 phase and 12 phase is easy and desirable.

    For hobby job, I would suggest to use a " DC choke input" filter on the single phase rectifier.
    The choke passively stores energy and provides it in the form of a decaying current while the instantaneous rectifier voltage is low.

    I used to work on design of the 1 phase and 3 phase rectifiers up to 800A and 1600A for telepower (telephone exchanges.)
    The ripple requirements were very tight because nobody wants to hear hum and harmonics.
    "Choke input" filters were invariably used because of their advantages over the "capacitor input" types.
    To keep the choke cost down a bit, we used swinging DC chokes which had high inductance at low current and partially saturated at rated currents, where ripple was lower.

    Such a choke should be possible for the hobbiest using surplus or scrap cores ( provided the cores have not been welded)
    Actually lathes make good coil winders, and many old lathes are converted for this.

  10. #10
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    25,840

    Default

    Quote Originally Posted by Henro View Post
    Actually this is incorrect. In a two phase system the phases are shifted 90 degrees from each other.

    In a three phase system the phases are shifted 120 degrees from each other.

    So there are three animals: single phase, two phase and three phase. Neither is equivalent to the other.
    Sorry, but it IS perfectly correct. The two are equivalent

    The 2 phase can be made into 3 phase using only linear reversible components, in fact just two transformers.

    Therefore, they are equivalent.
    1601

    Keep eye on ball.
    Hashim Khan

Posting Permissions

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