Broadband high ratio transformer

Discussion in 'Wireless & RF Design' started by heliosh, Apr 12, 2015.

  1. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    Hi,

    I'm trying to build a broadband transformer (1-300kHz) to convert an input impedance of ~20mOhm to 100 Ohm, that's 1:5000.

    I thought I'd take a toroid and make 1:70 turns. But beyond 10 secondary turns there is no more gain in impedance/voltage.
    I'm using an epcos N30 (µ=4300) toroid.

    What am I doing wrong? Do I need a different core material? Do a need a different core type (binocular?)? Do i need more primary turns? is it even possible with a single transformer?


    Thanks
     
  2. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    I don't think what your doing is reasonable or possible. Your impedance ratio is:

    1E2 / 20E-3 = 5000
    and √5000 ≈ 70

    So you have a 1 turn primary and 10 tuns is all you get on the secondary. Do I have that correct?
     
    Last edited: Apr 12, 2015
  3. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    100/0.02 is 5000 which should be archieveable by 1:70 turns (70²=4900)

    I'm trying to match a 100 Ohm receiver to a 0.02 ohm antenna.

    That's how the 1:10 looks:
    [​IMG]
     
    Last edited: Apr 12, 2015
  4. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    How do you know that the antenna is 0.02 Ω and how do you know that the receiver is 100 Ω.
    The antenna is a short in RF terms. Can you measure both real and imaginary parts? What are you doing the measurement with?
     
  5. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    Yes, that's correct. If i use 1:20 (with a thinner wire though) i don't get more voltage/impedance on the secondary.
     
  6. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    That's theoretical, i haven't measured the impedance.
    The radiation resistance is almost infinite small at this size (2.4*10^-12 Ohm) but the wire resistance is 0.02 ohm.
    The receiver has a high impedance OpAmp, but it's terminated with a 100 ohm resistor.
     
  7. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    And since the antenna is not tuned, the radiation resistance is almost completely imaginary, but i don't know if this matters.
     
  8. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    You CANNOT measure the DC resistance of the wire and conclude that it measures the characteristic impedance of the antenna. You also CANNOT do this with the antenna on the ground. You need an antenna analyzer and the antenna needs to be in it's elevated position with the feedline in place. All of that however is mostly irrelevant for a receive antenna. For the Very Low Frequency range the primary requirement of the antenna is that it must be physically large. Don't waste your time with the transformer.

    In the Very Low Frequency band a single fixed antenna will not have a very wide bandwidth. You might get 10 kHz. if you're lucky. On the 160M Amateur Band (1.8 to 2.0 MHz.) most fixed antennas will give you about 50-75 kHz. of bandwidth.

    Radiation resistance is for transmit antennas and does not apply in this case, and I'm suspicious of your measurement methods and/or theoretical calculation of this number as well.
     
    Last edited: Apr 12, 2015
  9. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    I see, i just want to get the maximum voltage out of the antenna.
    It is a 10m (30ft) loop with thick wire, already in the air.

    The 1:10 transformer already gave an improvement, so i thought i could get even more.
     
  10. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    I'm not looking for a tuned antenna (bandwidth). I wouldn't have all these problems with a tuned antenna ;)


    It is neccessary that it is not tuned, the phase need to be the same over the whole frequency range for timing measurements.
     
  11. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    The 30 foot loop is nowhere near big enough to be an effective antenna in the 300 kHz. range. The wavelength is 1000 Meters, so a dipole, for example, would need to be 500 Meters long and be at least 250 Meters above the ground.

    You can certainly use smaller gauge wire to get more turns.

    And if it's working for you that's great!
     
  12. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    more turns gives capacity which gives a too low resonance frequency which has an influence on the phase.

    The size is not the problem for the efficiency, but the low impedance, which gives 99% loss already within the antenna.
    But if i can get the 1% out, i'd be happy.
     
  13. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    The problem is you don't know the actual impedance of the antenna. I can assure with 100% certainty that is NOT 0.02 Ω as you have stated.
     
  14. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    Right, that 0.02 Ohm is the loss.

    But AFAIK is the Impedance = R + jX
    R is 0.02 and jX is 2.4E-12, so it's basically 0.0200000000024 Ohms.
     
  15. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    NO NO NO. Impedance is an AC thing and it is frequency dependent. You CANNOT and MUSTNOT draw any conclusions about impedance from a DC measurement. How many times do I have to repeat myself.
     
  16. heliosh

    Thread Starter New Member

    Apr 12, 2015
    10
    0
    Wrong.
    Normally the ohmic resistance doesn't matter. If you make a lambda/2 dipole (R73,1 + j42,5) from a very thin wire where the ohmic resistance becomes significant, let's say 27 Ohms, the impedance would be R100+j42.5.

    Or are you trying to tell me that a dummy load doesn't have an impedance?
     
    Last edited: Apr 12, 2015
  17. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    I obviously can't tell you anything.
     
  18. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    @Papabravo
    I would think a point driven loop of 10 meters diameter would have an almost purely inductive reactance of around 90-100 ohms at 300kHz in free space. Not sure where the TS's 2.4E-12 ohms comes from. Looks like it was plucked out of the air without any basis.
    One would also think the TS would tweak to the point that you are an experienced RF operator and worthy of some credibility in this field.
     
  19. Papabravo

    Expert

    Feb 24, 2006
    10,178
    1,799
    Who worries about radiation resistance on a receive antenna. I'm waiting to see the three element yagi for 300 kHz.
     
    SgtWookie and t_n_k like this.
  20. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    0.02 Ohms <-> 30ft.
    #3 = 1:10 <-> 1:5000

    Result: TTYL.

    Question: why do the large toroids from a switchting power supply heat up considerably?
    when the resistance should be 0.02 Ohms?

    A) It isnt 0.02 in the first place
    B) The imaginary AC resistance (an equivalent to a supposed to be DC resistance) is much higher and goes up with frequency.

    Summary: G.L. (good luck).

    and btw. 1-300 KHz broadband = nonsense
     
Loading...