Results 1 to 8 of 8

Thread: Switching frequency

  1. #1
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,612
    Thanks Thanks Given 
    178
    Thanks Thanks Received 
    98
    Thanked in
    89 Posts

    Default Switching frequency

    From the Guru's page (http://jongerow.com/)

    "
    The AX1600i was the first desktop power supply to go to market with GaN components. We replaced the entire silicon based bridgeless PFC circuit of our prior flagship product, the AX1500i, with a Transform GaN Totem Pole solution. This allowed us to increase efficiency, which helped reduce temperatures, making the PSU quieter at higher loads. And since the PSU used less power, we were able to output 100W more than its predecessor. And because of the higher switching frequency of the GaN components, we were able to reduce the size of other components reducing the overall size of the product by 20mm.
    "

    I thought the frequency was limited by the transformer core, not the switching transistors.

    My bad

  2. #2
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,918
    Thanks Thanks Given 
    571
    Thanks Thanks Received 
    524
    Thanked in
    388 Posts

    Default

    https://www.digikey.com/en/articles/...n-power-stages

    "GaN offers several advantages over silicon, including several related to the material’s higher electron mobility. Increased electron mobility bestows the semiconductor with a higher breakdown voltage (above 600 volts) and superior “current density” (amperes/centimeter2 (A/cm2)). Another advantage of GaN is that transistors constructed from the material don’t exhibit reverse recovery charge, a phenomenon that can lead to high switch current overshoot (ringing).

    But while these characteristics are important to the power supply designer, perhaps more critical is that high electron mobility allows a GaN transistor to switch in about one quarter of the time of a silicon MOSFET. Additionally, each time the GaN device switches, losses are around 10 to 30 percent those of a silicon transistor for a given switching frequency and current. As a result, GaN high electron mobility transistors (HEMTs) can be driven at much higher frequencies than silicon MOSFETs, IGBTs, or silicon carbide (SiC) devices."

    The impact to the transformer is that when the switching frequency is higher, the transformer can be smaller with less windings making it more efficient.

    But then you run into this: https://en.wikipedia.org/wiki/Proxim...ctromagnetism)

  3. The Following User Says Thank You to Jon Gerow For This Useful Post:

    ashiekh (3 Weeks Ago)

  4. #3
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,612
    Thanks Thanks Given 
    178
    Thanks Thanks Received 
    98
    Thanked in
    89 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post

    Which leads to the next question, does the AX1600i use Litz wire?
    Last edited by ashiekh; 3 Weeks Ago at 10:14 PM.

  5. #4
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,918
    Thanks Thanks Given 
    571
    Thanks Thanks Received 
    524
    Thanked in
    388 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    Which leads to the next question, does the AX1600i use Litz wire?
    Don't know.

  6. #5
    Join Date
    Feb 2015
    Location
    Germany
    Posts
    113
    Thanks Thanks Given 
    61
    Thanks Thanks Received 
    4
    Thanked in
    4 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    Which leads to the next question, does the AX1600i use Litz wire?
    If an inductor is wound with litz wire it doesn't necessarily mean that the inductor is suited for high switching frequencies. From a certain frequency upwards solid wire features lower skin effect and proximity losses than litz wire But with usual switching frequencies around 100 kHz litz wire is usually the better option.

  7. #6
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,612
    Thanks Thanks Given 
    178
    Thanks Thanks Received 
    98
    Thanked in
    89 Posts

    Default

    Quote Originally Posted by Microflop View Post
    From a certain frequency upwards solid wire features lower skin effect and proximity losses than litz wire

    How so?

  8. #7
    Join Date
    Nov 2010
    Location
    Serbia
    Posts
    2,248
    Thanks Thanks Given 
    128
    Thanks Thanks Received 
    16
    Thanked in
    8 Posts

    Default

    The skin effect depth depends on the effective surface area available to electrons. In a litz wire, due to the individual conductors' proximity, most of the additional surface area gained by splitting the material into thinner strands is lost, due to the electrons not having enough time to optimally re-arrange and "float to the surface" during voltage pulses (due to inertia). Having a greater bulk unified (more easily reachable) surface area for them to pass "over", as opposed to lots of lesser surfaces in close proximity and thus acting as parts of the same interior volume, is preferrable.
    Careful what you wish for... You just might get it.

  9. #8
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,612
    Thanks Thanks Given 
    178
    Thanks Thanks Received 
    98
    Thanked in
    89 Posts

    Default

    Quote Originally Posted by McSteel View Post
    ...due to the electrons not having enough time to optimally re-arrange and "float to the surface" during voltage pulses (due to inertia).

    Would that not be desirable? one does not want the electrons to be limited to the surface.


    Think I've found it: https://en.wikipedia.org/wiki/Litz_wire

    "At frequencies above about 1 MHz, the benefits become gradually offset by the effect of parasitic capacitance between the strands."

    "
    At microwave frequencies, the skin depth is much smaller than the diameter of the strands, and the current that is forced through the inner strands induces strong eddy currents in the outer strands, which negates the benefits of litz wire to the point where it performs much worse than solid wire of the same diameter"
    Last edited by ashiekh; 3 Weeks Ago at 12:55 PM.

Posting Permissions

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