[TowerTalk] New tool - toroidal ferrite core inductor calculator

[Jim Brown]

On 6/16/2021 8:33 AM, jim.thom jim.thom at telus.net wrote:
> It will  only handle a maximum of a 2.4 inch OD  type 31 core.  K9YCs
> latest designs all use  the much larger,  I think,  4 inch OD cores.

No, I studied both the larger cores and the more commonly used 2.4-in 
o.d. cores, and MOST of my designs are for the 2.4-in cores.

> As per number of turns, is this for wire, or coax,  or can be used for
> either ?   If coax used, what type of coax ?   I used  RG-303 and  RG-400
> on the smaller  CM chokes....and  RG-393 on the bigger CM chokes.

My designs are for RG400, #12 PTFL pairs, and #12 pairs from Romex.
> 
> Looking at the various attributes of a type 31  2.4 inch OD  toroid, by
> both  fair rite..and also amidon, I am not seeing any difference in
> properties between the 2 x manufacturers.

Amidon is NOT a mfr. They are a reseller of Fair-Rite, the inventor and 
ONLY mfr of #31 material, and they started making 2.4-in toroids with 
that material because I ASKED them for it in 2004 when I was beginning 
research on RFI to pro audio systems.
> 
> On a  side note, N3RR,  Bill Hider, a few yrs ago, bought  700  (seven
> hundred) of the type  31,  2.4 inch OD cores...from I think,  Fair
> rite..... and the tolerance between all of em  was strewn all  over the
> map,  like  +  or  -  20+ %.   Bill came up with a simple  1  turn test to
> evaluate,and then grade them into various groups.  Some time in the past,
> apparently  Fair rite  moved their production facilities to  CHINA, and
> after that, the  ± tolerances went all to hell.

There is no evidence that their production deteriorated when they 
established their facility in China. As part of my work to develop my 
Choke Cookbook, I measured 10-turn chokes wound with a single piece of 
RG58 on more than 200 cores; roughly half were 2.4-in o.d. cores, 
purchased over nearly 15 years from Fair-Rite directly and from 3-4 
industrial vendors, including the first five pre-production engineering 
samples that Fair-Rite sent me in 2004. The 4-in o.d. cores were a 
single group purchase for members of our contest club from a single 
vendor about five years ago.

My measurements of those 200 or so cores was intended specifically to 
address manufacturing tolerances, which have a profound effect on HF 
chokes wound on these cores. ANY competent engineer designing something 
for production must take these tolerances into account. To produce my 
cookbook, I put measurements of my reference winding on all of those 
200+ cores into spreadsheets, one for each cores size, and selected 
cores at the limits (i.e. worst case) on which to wind chokes to produce 
my designs. I then wound chokes on those "limits" cores with all of the 
winding materials (RG400, #12 PTFM and #12 NM), measured them starting 
with 18 turns (for the 2.4-in o.d. cores), snipping a turn, down to 
where their resonance was higher than 30 MHz, and remeasuring. For each 
measurement, leads had to be snipped to "zero length" so that lead 
inductance did not add measurement error (I was measuring a parallel 
resonant circuit).

All that data went into a spreadsheet, with data points at limits of 
160, 80, and 40M bands, centered in 20 and 15, and at around 28.5 MHz, 
and for each winding (turns, winding material), worst case of Rs (lowest 
value) was used to develop the Cookbook. The result is that a ham using 
the Cookbook will have a choke that is equal to or BETTER than those 
minimum values.

Before he died several years ago, G3TXQ published recommendations for 
chokes, some of them using #31 material, others using #43 and #52. After 
finishing my Cookbook, I bought 40 #52 cores from four US industrial 
vendors, 10 pieces from each, over a period of about two months, 
repeated my process of measuring my "reference" winding on each, and 
selecting limits cores. I then wound chokes following his 
recommendations on "limits" cores and measured them. I was unable to 
duplicate his results on any of those "limits" cores.

Steve was clearly a great scientist, and clearly knew how to measure 
correctly. What he apparently failed to understand was manufacturing 
tolerances -- I suspect that he measured no more than a few cores of 
each material. The shortcoming of NiZn materials like #43 and #52 for 
use on HF is that they are much higher Q, so manufacturing tolerances 
have far greater effect on the resonance of multi-turn chokes than #31, 
which is a VERY SPECIAL MnZn material, which has a dimensional resonance 
within the HF spectrum and lower Q winding resonance that makes it's 
impedance curve much broader.

To understand these concepts, I recommend a study first of 
k9yc.com/RFI-Ham.pdf and then the Choke Cookbook. Note that I said STUDY 
-- I'm an EE, and it took me a decade of serious study to understand 
this stuff!

73, Jim K9YC