Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: What cores for 1.8~30Mhz?  (Read 2111 times)
G4COE
Member

Posts: 0




Ignore
« on: April 20, 2010, 11:46:56 AM »

Hi,
Is there anything special about selecting ferrite cores for HF broadband amplifier output matching?

What it is, I have some Fair-Rite 2643540402 cores (I got from R.S Components 467-2750), these are the 43 Material. I knocked a broadband transformer up using a pair of these and brass tubing for the primary with fibre glass pcb end pieces.

As a test so I could measure the inductance, I used standard ‘hook up’ wire winding four turns as the secondary this gave me around 100uH, the single turn brass tube was something like 10uH on a LCR meter, unloaded without any termination.

Kindly note, the 'hook up' wire is only for testing with the LCR meter, obviously no good for the finished product.

My intentions was to build a PA for a 1.8~30Mhz 100 Watt PA with a pair of 2SC2097 or a pair of MRF454 that I have to hand. I am aware that you can buy these transformers ready made from RF Parts, 2 questions-:

1) Does the inductance sound about right?

2) If this is not the right core would anyone know of a better one to use preferably that is easily obtainable in the UK from say Farnell or R.S. Components.

Thanks, Dave
Logged
AA4HA
Member

Posts: 1482




Ignore
« Reply #1 on: April 20, 2010, 12:07:25 PM »

There are a few really good reference works I use for balun core selection, winding, design and testing. Here is one;

http://www.filestube.com/836422ed2251dfae03e9,g/ham-radio-Transmission-Line-Transformers-Handbook.html

Another great reference is found at;

http://www.fair-rite.com/newfair/index.htm

On the top menu bar [technical], [use of ferrites in broadband transfomers] -and- [ferrite materials] This will download a document called broadband.pdf It has excellent details on different ferrite formulations and the frequencies they are good for.

The [materials] section has many listings for what ferrites are useful for RFI suppression and what other ferrites are useful for broadband transformers. You may have several choices for balun materials and '43 may not be the best.

Usually Nickel-Zinc formulations are better for baluns and Magnesium-Zinc is used for RFI or very low frequency transfomers.

Tisha Hayes, AA4HA
« Last Edit: April 20, 2010, 12:20:54 PM by Tisha Hayes » Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
G4COE
Member

Posts: 0




Ignore
« Reply #2 on: April 20, 2010, 12:36:40 PM »

Thanks for your time and the quick response Tisha.

I thought 43 material was the 'norm' for these applications, I'll make a note of the links and have a browse and do a bit more 'homework'.

Cheers and thanks again, Dave
Logged
AA4HA
Member

Posts: 1482




Ignore
« Reply #3 on: April 20, 2010, 02:40:06 PM »

I keep a bunch of antenna and balun documentation on my dropbox site. If you want to access the data (around 2.5 gig total for everything) you would need to install dropbox    https://www.dropbox.com/referrals/NTYyNDg4NzA5

Then shoot me an email and I will invite you into the dropbox so you can see, download and upload documents from the site. Much of the site is dedicated to the R-390A and SP-600 receivers but I have many directories on antennas (that includes baluns) and grounding and surge (lots of Polyphaser stuff, military EMP documents and other commercial protection design docs).

Some of the documents I keep in there include;
Amidon balun booklets
Fair-Rite design guides
transmission line guides
the Bryant balun documents
(around 50 other antenna and balun documents).

I collect data.

Tisha Hayes, AA4HA
Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
VK1OD
Member

Posts: 1697




Ignore
« Reply #4 on: April 20, 2010, 02:53:41 PM »

...
As a test so I could measure the inductance, I used standard ‘hook up’ wire winding four turns as the secondary this gave me around 100uH, the single turn brass tube was something like 10uH on a LCR meter, unloaded without any termination.

Viewing the thing as an ideal inductor (ie characterised as inductance only), and worse, based on measurement at a frequency outside the range of interest if fraught with problems.

See A method for estimating the impedance of a ferrite cored toroidal inductor at RF for more information.

Motorola published some good application notes on ferrite cored broadband transformers for RF power amplifiers.

Owen
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #5 on: April 20, 2010, 04:07:19 PM »

One method of analysing the transformer you envisage is that it is a transmission line formed by the tube and the n conductors inside the tube.

If the length of the turns is very small wrt wavelength, the current on the turns is approximately uniform, and the total of the currents on the turn conductors will equal the current flowing in the opposite direction on the inside of the tube.

The current on the outside of the tube is 'common mode' current, and determined by the common mode impedance of the struture... essentially the impedance of the single turn choke. This common mode impedance appears in shunt with the load seen by the transistors.

The fair rite data sheet contains an impedance plot for the cores you mention over HF.

From an impedance point of view, it is probably sufficiently high for the purpose, but I question whether just two cores can dissipate the heat produced in a 100W continuous rated PA. Doubling the number of cores roughly halves the total dissipation, and the dissipation of each core is reduced by about 75%.

My advice is to literally copy a known good design, copy it in every aspect... or design the transformer from the ground up and prototype it to confirm its performance.

Owen
Logged
G4COE
Member

Posts: 0




Ignore
« Reply #6 on: April 21, 2010, 03:39:56 AM »

I do seem to think it's a replacement for an obsolete one

I've inquired at Fair-Rite, I will post back with the 'yeah or neighs' for the benefit of others who may read this in future... just that I thought someone may know off-hand 'to be sure to be sure'.

VK10D has it about right in saying to copy a well known good design, this is what I tried to do, the ones we see here in the UK are either Philips cores which are not readily available or that it's ordering from abroad. 

I really don't want to go into the 'ins n' out's' of the physics, i.e. losses etc as LA7MI has done on his excellent web page http://www.noding.com/la8ak/12345/n12.htm, basically I just wanted confirmation if the cores in question where suitable for the job.

Thanks to everyone for the input, 73's and good luck, Dave.
Logged
G4COE
Member

Posts: 0




Ignore
« Reply #7 on: April 21, 2010, 03:40:39 AM »

Sorry clicked twice.... arghhh.

73's Dave
« Last Edit: April 21, 2010, 03:50:07 AM by David Smith » Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #8 on: April 21, 2010, 07:53:41 AM »

...
I really don't want to go into the 'ins n' out's' of the physics, i.e. losses etc as LA7MI has done on his excellent web page http://www.noding.com/la8ak/12345/n12.htm, basically I just wanted confirmation if the cores in question where suitable for the job.

Then take your prototype and parallel the high impedance winding with a 50 ohm dummy load, and drive the pair with 100W over the frequency range of interest. The VSWR isn't all that critical, but you would hope that it is below 1.5, but more importantly does the transformer get excessively hot?

Keep in mind that the ferrite core heats quite slowly, it might take more than half an hour to increase half of the otherwise would be temperature rise.

Owen
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!