Jim,
Ok, you made me fire up the network analyzer (HP8753B). The only cores
that I could find which were both from Fair-Rite, and in 31 and 43
materials were 2631102002 (1.025"OD x .510"ID x 1.14" long and a 43
matl. core (1.15"OD x .535"ID x 1.14" long) which are very close to the
same size. The 43 material core I found would be slightly better than a
43 material core of the size that I had for the 31 material core but
this was the closest comparison I could find.
Here are the measurements:
#43 core: 1 turn / 2 turns / 3turns
1.8MHz 6+j27 / 23+j114 / 60+j265
3.5MHz 23+j46 / 95+j191
10MHz 61+j60 / 240+j259
30MHz 117+j81 / 418+j385
100MHz 201+j36 / 524+j421
#31 core: 1 turn / 2 turns
1.8MHz 20+j39 / 81+j158
3.5MHz 44+j49 / 175+j200
10MHz 74+j58 / 290+j253
30MHz 131+j79 / 471+j383
100MHz 207+j43 / 534+j446
At 1.8MHz and with 2 turns the 31 material core was better than 2 turns
on the 43 material core but by adding another turn on the 43 core
changed the result. I wish I could have found identical sized cores but
the above 2 were quite close. Granted going from 2 to 3 turns should
increase the impedance by 1.5 squared = 2.25 but then again the 31
material core is about twice the permeability of the 43 core. Draw your
own conclusions.
My posted formula is indeed simplistic but I have used it for years when
I was at an EMC lab charging $200/hour and I needed a fix quickly to a
non-compliant piece of equipment.
73, Larry W0QE
Jim Brown wrote:
> On Sun, 11 Jan 2009 20:44:20 -0700, Larry Benko wrote:
>
>
>> Two stacked FT-50-43 cores would have just about the
>> same impedance as a single FT-50-31 (if that part was made).
>>
>
> Not below 5 MHz it wouldn't. #31 behaves very differently below
> about 5 MHz.
>
> Jim
>
>
>
> _______________________________________________
> RFI mailing list
> RFI@contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
>
>
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|