Charles, Yes you are correct. In my previous post I failed to take the square root so the turns are 3 instead of 11. Also, if you are building a choke balun for your coax, type 43 material is preferable to type 77 material because it will not saturate as easily as type 77 material. I have built choke baluns using an F-240-43 cores. I placed 16 turns of RG-58 on the core which gives a measured inductance of about 155 uh for a reactance of 1753 ohms at 1.8 MHz. I wound it in the split mode as you described. This core is large enough not to saturate even at full power as long as the current on the outside of the coax is not too large. The F-140-43 should be all right at 6 turns. The trick is to get the inductance as high as possible for maximum choking without saturating the core. Be careful when you do the crossover on the winding. In the process of this discussion, I took another look at one of my baluns, and I had it incorrect. I was getting 3.5 uh instead of 155 uh. Thank you, Charles, for causing me to catch that error. But back to the point of the F-114-77. I recommended that core for the USB cable not a coax choke balun. However, if you have FT-140-43 cores, by all means try them. They will probably work fine.
73 Tom W0IVJ Charles Greene wrote: > Tom, > > I always did have trouble with that formula. > > At 09:59 AM 3/8/2007, you wrote: > >> Charles, >> >> The equation is: L = (N^2*Al) / 1000000 where L is in mh or L = >> (N^2*Al) / 1000 where L is in uh. Using an Al = 1270 with one turn, >> then L = 1.27 uh. If N is doubled to two then L = 2^2*Al/1000 or >> 5.08 uh. You have two errors in your text. You switched from mh to >> uh and you failed to square the multiplier of the turns. Your >> calculation of 44 uh giving a reactance of about 1000 ohms at 3.5 MHz >> is correct., but normally the reactance is from 5 to 10 times the >> coax characteristic impedance which would be 250 to 500 ohms. Let's >> assume 300 ohms which would give 13.6 uh at 3.5 MHz.. Solving for >> the number of turns on an FT-114-77 core would be: N = >> SQRT(1000*13.6/1270) = 11 turns. > > > SQRT(13,600/1270) = SQRT(10.7) = 3.27 TURNS. Correct? That's the > kind of trouble I get into with this formula. > > Working the other way, L = (3.27^2 * 1270)/1000 = 13.6 uh > > Now the Al for a F-140-43 core is 952. Solving for the number of > turns on a F-140-43 core would be N = SQRT(1000*13.6/952) = > SQRT(14.28) = 3.78 or approx 4 turns. > > To build, wind two turns on the core spaced so that the finish is 180 > deg from the start. The next turn goes through the center and comes > out near the start, winding in the same direction and away from the > start. Space two turns so that the end is 180 deg from the start. It > makes a nicer job if the cable is tied to the core at the start and > finish. If this is not clear, look at page 7 of Sevick's Building and > Using Baluns and Ununs. That's what I was trying to find out, if the > F-140 could be used instead of the F-114-77. > > Is there any problem if I use 6 turns constructed in a similar manner? > L = (6^2 * 952)/1000 = 35 uh. At 1.8 MHz, Xl = 6.28 * 1.8 * 35 = 396 > ohms. I think 6 turns will fit. Like I said, I have a whole bunch of > F-140-43 cores left over from a kit. > > Chas > > > L + > >> However, we are not building a choke to go over the coax here. We >> are building a choke for the USB cable. The difference is the common >> mode current on the USB cable is much smaller that the common mode >> current on the coax shield. >> One has to be careful with high-u ferrites not to saturate the core. >> That is why current baluns are built with low-u ferrites or powdered >> iron cores that have large cross sectional areas. Typically an >> FT-240-43 core will make a good choke balun for coax. For the USB >> cable the common mode currents are small enough, so you can put a lot >> of turns on, and use an FT-114-77 core. >> >> I hope this has helped. >> >> 73 Tom W0IVJ >> >> >> Charles Greene wrote: >> >>> Tom, >>> >>> I know that in a coil L is proportional to N^2. How do you >>> reconcile the fact that, in a F-114-77 core, Al is 1270 mh/1000 >>> turns, or 1270 uh/turn. That would make one turn produce 1270 uhy >>> and two turns produce 2540 uhy. What is wrong with my logic? On >>> 1:1 baluns 44 uhy produces approximately 1000 ohms at 3.5 MHz (44 * >>> 6.28 * 3.5) ohms or 20X the coax impedance at 3.5 MHz, so it seems >>> the type 77 material doesn't need many turns to be effective. Also >>> I read up on the -77 material, and it states "Extensively used for >>> frequency attenuation at .5 to 50 MHz." I have a bunch of F-140-43 >>> cores and I was wondering how effective they would be The price ea >>> for a F-114-77 at CWS Bytemark.com is $2.25. Do you have a better >>> source. I have bought F-140-43s from Dan's Small Parts for $1.00 ea. >>> >>> Chas >>> >>> At 11:28 PM 3/7/2007, you wrote: >>> >>>> Charles, >>>> >>>> You are correct about the F-140 because of its increased size, but the >>>> 77 material has an initial permeability of 1800 as opposed to the >>>> initial permeability of 850 for 43 material. The 1800 initial >>>> permeability will yield more inductance per aturn , and a core with >>>> multiple turns yields inductance as a function of turns squared >>>> whereas >>>> a ferrite clamp increases inductance linearly with respect to the >>>> number >>>> of clamps. This difference is due to the mutual inductance effect >>>> that >>>> the multiple turn core has that does not occur with the multiple >>>> clamp. >>>> >>>> 73 Tom W0IVJ >>>> >>>> >>>> Charles Greene wrote: >>>> >>>> > That sounds like a good EMI filter. Wouldn't a F-140-43 core be >>>> > better? (more turns, and better performance at HF? I don't know, >>>> > just asking). >>>> > >>>> > C >>>> > >>>> > At 10:19 AM 3/7/2007, Tom Thompson wrote: >>>> > >>>> >> Charles, >>>> >> >>>> >> Get an FT-114-77 ferrite toroid from Palomar Engineering and >>>> wrap as >>>> >> many turns of the USB cable as you can on the core. >>>> >> >>>> >> Tom W0IVJ >>>> >> >>>> >> >>>> >> Charles Greene wrote: >>>> >> >>>> >>> Hi, >>>> >>> >>>> >>> I am getting some RF feedback when I use a Linear amp that causes >>>> >>> the USB to Parallel connection to occasionally disconnect. The >>>> >>> system ground is about as good as I can reasonably get it >>>> except for >>>> >>> the Laptop itself. It doesn't happen with the SDR-1000 100 watt >>>> >>> amp. There doesn't appear any good place to ground the >>>> Laptop. Any >>>> >>> suggestions? Also, should I use an EMI suppressor on the >>>> >>> USB/parallel cable, and if so on which end, computer or >>>> SDR-1000, or >>>> >>> both? >>>> >>> >>>> >>> Tnx, W1CG. >>>> >>> >>>> >>> >>>> >>> _______________________________________________ >>>> >>> FlexRadio mailing list >>>> >>> FlexRadio@flex-radio.biz >>>> >>> >>>> http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz >>>> >>> Archive Link: >>>> http://www.mail-archive.com/flexradio%40flex-radio.biz/ >>>> >>> FlexRadio Homepage: http://www.flex-radio.com/ >>>> >>> >>>> >>> FlexRadio Knowledge Base: http://kb.flex-radio.com/ >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >> >>>> > >>>> > >>>> > >>>> > >>>> >>>> >>>> >>>> _______________________________________________ >>>> FlexRadio mailing list >>>> FlexRadio@flex-radio.biz >>>> http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz >>>> Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ >>>> FlexRadio Homepage: http://www.flex-radio.com/ >>>> >>>> FlexRadio Knowledge Base: http://kb.flex-radio.com/ >>> >>> >>> >>> >>> >> > > > > _______________________________________________ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com/ FlexRadio Knowledge Base: http://kb.flex-radio.com/
