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: 4:1 balun design  (Read 2468 times)
G4IJE
Member

Posts: 236




Ignore
« on: May 05, 2013, 01:32:04 PM »

I have just built an OCF dipole (AKA "Windom") and I need a 4:1 balun that covers 7 to 50MHz. I briefly used a Hari 4:1 balun (made in Germany) but I managed to fry it with 100 watts on 6 metres. After putting up the antenna this afternoon, I did a quick SWR check on the four bands it covers and was delighted to find everything exactly as per the EZNEC prediction. I found some Sporadic E on 6 metres and worked HA8EK on the first call. Great! It works! Unfortunately that quick exchange of 5/9 fried the poor balun (to be fair, it was only specified to work up to 30MHz), so now I need to build a better one.

I have some FT140-61 ferrite cores which I'd like to try. Searching the internet reveals many similar designs based on bifilar windings around a toroid. There seems to be some debate above whether it's worth reversing the winding. Also most designs are for 3.5 to 30MHz. Is it worth using a few less turns for a design intended for 7 to 50MHz? What wire gauge would be appropriate for 100 watts? Reverse the winding halfway or not bother? I plan to check the "match" by using a 200 ohm load and measuring SWR but some suggestions for a starting point (wire gauge, number of turns etc) would be appreciated.

73, Paul G4IJE.
Logged
K3VAT
Member

Posts: 699




Ignore
« Reply #1 on: May 05, 2013, 01:53:23 PM »

Paul,

Did you search the archives http://www.eham.net/ehamforum/smf/index.php?action=search there seems to be hundreds of postings on this popular topic and on 'ferrite cores'.

GL, 73, Rich, K3VAT
« Last Edit: May 05, 2013, 02:22:50 PM by K3VAT » Logged
G3TXQ
Member

Posts: 1432




Ignore
« Reply #2 on: May 06, 2013, 12:05:55 PM »

That's a very wide frequency range to try to cover with #61 material, which is relatively high Q. I just did a few test windings - 8 turns and 12 turns - and it still wasn't enough to get decent choking impedance at 7MHz.

Steve G3TXQ
Logged
G4IJE
Member

Posts: 236




Ignore
« Reply #3 on: May 06, 2013, 02:11:45 PM »

That's a very wide frequency range to try to cover with #61 material, which is relatively high Q. I just did a few test windings - 8 turns and 12 turns - and it still wasn't enough to get decent choking impedance at 7MHz.

Steve G3TXQ

Many thanks for that Steve - way beyond the call of duty :-)

Any suggestions where to go with this idea? The antenna looked excellent for the few minutes before I fried the Hari balun and I would really like to get it working properly.
Logged
G3TXQ
Member

Posts: 1432




Ignore
« Reply #4 on: May 07, 2013, 03:12:22 AM »

Is it a 4:1 Voltage balun or a 4:1 current balun you want? I'm sure you know that a voltage balun will drive CM current onto the coax feedline whereas a current balun will tend to prevent it.

Do you have any other cores available? If so what type?

What modes will you run at the 100W? It helps in doing the balun heating calculations!

73,
Steve G3TXQ
Logged
G4IJE
Member

Posts: 236




Ignore
« Reply #5 on: May 07, 2013, 09:13:51 AM »

Is it a 4:1 Voltage balun or a 4:1 current balun you want? I'm sure you know that a voltage balun will drive CM current onto the coax feedline whereas a current balun will tend to prevent it.

Do you have any other cores available? If so what type?

What modes will you run at the 100W? It helps in doing the balun heating calculations!

73,
Steve G3TXQ

I guess a current balun would be preferable. I don't have any other cores but I would be willing to invest in some. The 100 watts would be mostly SSB but I do like a bit of SSTV which I run at about 70 watts or so with 100% duty cycle for up to 2 minutes.

Any further assistance would be greatly appreciated.

Thanks and 73, Paul G4IJE.
Logged
K3VV
Member

Posts: 13




Ignore
« Reply #6 on: May 07, 2013, 07:38:21 PM »

I agree that you want a current balun.  I've made several 4:1 current baluns for OCF antennas using 43, 61, and 73 materials.

The crossed (Trask) winding style produces higher common-mode (CM) impedance at the upper end of the balun’s frequency range while producing lower CM impedance at the lower end of the range.  You should use the style that produces greater CM impedance at whichever end of the range is giving you the most trouble.  I always have trouble at one end or the other even with 80-10m designs.

Type 61 is probably the best choice to try if the desire is for a broadband balun. On paper type 61 material is good to 200 MHz, and it can handle power at 6 meters without overheating, so you're in good shape to start experimenting.  As Steve, G3TXQ pointed out, the problem with 61's relatively low permeability (v/v type 43) is that it takes many turns or tall stacks of cores to achieve adequate CM impedance. Tall stacks yield a better solution than many turns because they produce lower interwinding capacitance.   Type 61 is a better choice than type 43 because type 43 gets too hot running against an unbalanced load with an impedance mismatch.  A lot of the commercial designs use type 43 because it reduces the amount of ferrite required.

I use binocular designs formed of paired core stacks to keep more of the winding inside the cores.  This raises the common-mode-impedance to shunt-capacitance ratio, which is a good thing at the higher end of the balun’s frequency range.  The extra surface area helps if there's heating as well.

I usually put a 1:1 common-mode choke on the transmitter side of the balun to increase CM impedance further.  This is an ugly solution, but it works.

There are some photographs and comments on materials and construction here:

http://www.flickr.com/photos/flint-hill/sets/72157633429796001/

I would suggest dividing your stock of FT140-61 toroids into two stacks depending on how many you have.  I have had to use as many as 8 toroids per stack to get decent choking impedance on 80 meters while using few enough turns to work at 10 meters.  I don’t work 6 meters, so I have no direct experience there. Perhaps start with 4 per stack.

The best way to measure loss is to wire two of the 4:1 baluns back to back, take voltage measurements at the input and output and calculate the loss from the squared voltage ratio.

You can check for heating empirically by hooking the balun up at the transmitter and running a high-duty-cycle mode (SSTV in your case).  Check the balun temperature with your thumb when you’re receiving.  You'd like little or no heating even with extended operation.

There's really no substitute for testing the common-mode impedance once you've built the thing.  This is tough to do with a 4:1 balun unless you have an analyzer that allows you to null the text fixture before testing the balun.  It’s possible to make qualitative comparisons with a non-nullable analyzer like my MFJ-269. 

I've never used them, but local six-meter people speak well of Hy-Power's GU4-HF baluns.  They're rated for matched loads, so you'd want to derate them for real-world usage.   There's little enough ferrite even in their 5kW balun that I'd probably get that one rather than one of the lower-power versions.  I believe you can build a better balun yourself without too much trouble.
Logged
G4IJE
Member

Posts: 236




Ignore
« Reply #7 on: May 08, 2013, 12:06:54 AM »

Many thanks for the information. All very helpful indeed. It seems I need to buy a few more 61 cores - I have just two at the moment. It looks like designing a balun that will cover 7 to 50MHz with decent performance is not trivial.

Thanks again for your very detailed response.

73, Paul G4IJE.
Logged
G3TXQ
Member

Posts: 1432




Ignore
« Reply #8 on: May 08, 2013, 03:08:38 AM »

Paul,

I fear that trying to achieve good choking impedance across a 7:1 ratio frequency range will be near impossible using #61 material.

The best I've managed to achieve using the various cores I have to hand was >2200 Ohms from 7MHz to 50MHz; that was using 8 bifilar turns of "figure-8" twin on an FT240-43 core.

K3VV rightly points out that there can be power handling issues with #43 material because of its "lossy" characteristic; but that's the very characteristic that produces broadband choking performance.

A quick calculation shows that two of those chokes forming a 4:1 Guanella balun should be able to handle your 100W SSB or 70W SSTV.

If you decide to go that route, the cores are available from Amtools in UK.

73,
Steve G3TXQ
Logged
K3VV
Member

Posts: 13




Ignore
« Reply #9 on: May 08, 2013, 10:25:58 AM »

I think of Steve as knowing more about ferrite than I do, so I'd follow his advice if I were you.  Smiley

Here in the states, type 43 ferrite is easier to find and often quite inexpensive by comparison with type 61. I usually buy mine on ebay from a Canadian vendor.  Looking just now, see there are a number of vendors at ebay.co.uk that offer similar prices.

Four tallish stacks of type 43 toroids in 90 or 114 size arranged as two binocular cores could make a very effective 4:1 balun with a low turns count.  You'd need to check for heating when you build the prototype.

Logged
G4IJE
Member

Posts: 236




Ignore
« Reply #10 on: May 08, 2013, 12:39:53 PM »

Thanks for all the good ideas guys. I'll let you know how I get on.

73, Paul G4IJE.
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!