[TowerTalk] Snap-on Ferrites

[Grant Saviers]

I think there are two issues:

Is the core material (ferrite grade) the right choice for the job? Most 
commodity clamp ons sold at flea markets and on ebay aren't specified so 
may or may not provide the properties needed for the application. Is it 
31, 43, 61 or ??.

There are two air gaps in a clamp on magnetic circuit.  These reduce the 
effective permeability of the ferrite chosen and thus the inductance.  
No matter how well lapped the surface and how tightly held in contact 
the permeability is reduced because the gap can't be reduced to the zero 
of a solid toroid.  Here is a calculator:
http://www.encyclopedia-magnetica.com/doku.php/calculator/effective_permeability_from_air_gap

So for a #31 big clamp on, if the air gap is 0.002" total (0.001" each 
gap) and the average length of the magnetic circuit is 5"  then about 
25% is the permeability loss.  (approximate data).

So tight clamping is one lesson besides buying the correct grade. Tie 
wraps or #33 are holding all of my clamp ons tightly together as the 
plastic clips are iffy.

Is it a big penalty?  The #31 big clamps ons are not cheap but contain a 
lot of ferrite, work well, and are super convenient. Worth it IMO.  
Trust the test data from K9YC and others.  I've measured a number of "no 
brand or  spec" small clamp ons and many seem to be #43.  Some of the 
TDK ones and others on ebay have part numbers so the specs can be mined.

It is a trick in power magnetics to introduce an air gap to reduce the 
potential of saturating the magnetic material.

Grant KZ1W

On 6/18/2017 19:44 PM, john at kk9a.com wrote:
> I was reading the recent QST article RF Mayhem by WB8IMY. The author said to
> avoid using snap-on ferrites for problems caused by HF signals as they
> rarely work well at HF. It is best to stick with circular cores.
>
> Is this correct?  I was not aware that snap-ons were less effective.
>
> John KK9A
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>