Ferrite for baluns

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KC0LWN, Feb 8, 2015.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: l-gcopper
ad: L-MFJ
ad: Left-2
ad: l-BCInc
ad: Left-3
ad: QSOToday-1
ad: abrind-2
  1. W5DXP

    W5DXP Ham Member QRZ Page

    Here's what I said in posting #16 of this thread. "It appears that there exist two different kinds of ferrite materials with the designation 'K' so caution is advised."

    I got my FT-240-K from Amidon. Testing with an AIM4170 indicates that it is slightly superior to #52 material for HF applications. In posting #18 of this thread, G3TXQ posted a graph comparing Amidon's K-material to 52-material. Here it is again:

    [​IMG]

    One would think that buying the FT-240-K toroid or the AB-240-250 balun kit from Amidon would ensure the u=250 K material. Although I cannot find any Fair Rite K-material technical graphs on the web, I believe (not certain) that Amidon's K material toroids were manufactured by Fair Rite.
     
    Last edited: Feb 22, 2015
  2. HEYJESSICA1987

    HEYJESSICA1987 QRZ Member

    Hey, welcome to this forum site. Here you can find lots of things. But right now i dont have the proper answer to your

    question. I would suggest you to consult an expert regarding your question.
    -----------------
    Top Ten classified website
     
  3. G3TXQ

    G3TXQ Ham Member QRZ Page

    I see that in a more recent blog Owen has commented that some of my measured values of Rs for 2 turns on an FT240-52 toroid look low compared to those you would predict from the FairRite u' and u'' data. I agree with him, and I've spent this morning trying to find the problem!

    I have several sets of data for 2 turns on an FT240-52 toroid, and all of them have significantly higher values for Rs than the set I used for that chart. I thought at first I may have had one "rogue" toroid, or one that had been mislabelled by the supplier, so this morning I re-ran measurements on all 5 of the FT240-52 toroids that I have; they all produced similar, significantly higher, values of Rs!

    I was puzzled until I realised that the unique feature of the data-set used for the chart was, unusually, it had been produced using my AIM4170 running v882 software. I say "unusually" because for the past couple of years all my choke measurements have been made using a VNA2180, but I had recently unpacked my AIM4170 to check some of the problems claimed for its 882 software, and must have run the FT240-52 measurement using it. So it looks like the AIM882 bug "bit me", although I haven't yet unravelled why it would have affected values of Rs quite so dramatically.

    Here's a revised comparison chart using the Type 52 data from the VNA2180:

    [​IMG]

    Steve G3TXQ
     
  4. KC0LWN

    KC0LWN Ham Member QRZ Page

    The K mix I own is from Bytemark CW who s a Fairrte distributor.
     
  5. KC0LWN

    KC0LWN Ham Member QRZ Page

    Amidon is also a Farrite distributer.
     
  6. KC0LWN

    KC0LWN Ham Member QRZ Page

    31b mix is great for chokes but has eddy currents at a lower level of energy than K. 61, and 67. I believe it is called hysteresis. When I contacted an engineer at Wurth lectronics he recommended only using NiZn mix for broadband transformers saying their 4w mix is ideal while the MnZn mixes are for transformers and common mode choke use.
     
  7. GM3SEK

    GM3SEK Ham Member QRZ Page

    All through this thread, I'm afraid you are still missing something fundamentally important: common-mode chokes are not 'transformers'. The two types of application are fundamentally different, so ferrites suitable for one application may not be optimum for the other.

    In 'transformer' applications the whole of the RF energy magnetizes the core, so losses in the core material must be avoided.

    But the choke application is different in two important ways:

    1. The main flow of RF energy does not pass through the core material. The transmission-line currents do not magnetize the core. The core is only magnetized by a much lower level of residual common-mode current.

    2. In choke applications at HF, it can be good to have losses in the core material. In a well designed choke, core losses help to increase the impedance of the choke. Contrary to what you might have expected, this will usually decrease the heat dissipation in the core itself.

    Therefore you should be applying a completely different set of criteria for 'choke' applications as opposed to 'transfomer' applications.

    Another reason why you may have become confused is that a material specified for transformer applications may also be suitable for choke applications but in a completely different frequency range.

    This kind of technical information isn't easy to communicate through the written word - but attempting to do it by phone is a whole new level of difficulty! I can quite imagine a conversation with the Wurth engineer that seemed to be making sense at the time, but there was actually a complete disconnect in understanding at both ends.

    Bottom line: understand the fundamental difference between transformer and choke applications, and then try to build a new understanding around the observed, known fact that #31 ferrite is suitable for choke applications.


    73 from Ian GM3SEK
     
  8. KC0LWN

    KC0LWN Ham Member QRZ Page

    I know. An ealier post sggested using 31 in a hybrid form 4:1. Common mode chkes atre not the issue. I have some K and will just use that. I thank everyone for their helpand apologize if there was a miscommunication.
     
  9. G3TXQ

    G3TXQ Ham Member QRZ Page

    If by "hybrid form 4:1" you mean a 4:1 Ruthroff (Voltage) balun combined with a 1:1 Guanella (Current) balun, that is a combination of Common-Mode chokes. See Section F in the paper linked here:
    http://www.karinya.net/g3txq/baluns/

    Steve G3TXQ
     
  10. VA3RR

    VA3RR Ham Member QRZ Page

    Excellent paper!
     
Thread Status:
Not open for further replies.

Share This Page

ad: elecraft