Re: LF: Re: My first ferrite experiments

[Daniele Tincani <danieletincani@yahoo.com>]

Hello Wolf, thank you for sharing this informstion. Yes I fully agree with you that patch cords with crocodile clips could be an issue. I plan to re-do some measurements as soon as I will have given a more suitable "mechanical" look to the set.

Best 73

Daniele

Da: wolf_dl4yhf <dl4yhf@freenet.de>
A: rsgb_lf_group@blacksheep.org
Inviato: Mercoledì 28 Settembre 2011 22:46
Oggetto: Re: LF: Re: My first ferrite experiments

Hi Daniele,

Looking at those alligator clips: Did you ever remove the rubber insulation, and check the connection between clip and cable, and check the 'copper diameter' of these cables ?
I did, and found that...

a) they didn't even solder them, but used a trashy loose "crimp" (well, not really crimped but crippled) connection, the copper conductor folded back over the wire's insulation, sometimes only slightly touching the chrome surface of the clip,

b) there's a ridiculously low amount of copper in those wire !  (lot of PVC which makes you *think* the wire is thick, but it's crap)

Cure: Replaced the wire with other colourful cabling salvaged from a PC power supply, and SOLDERED the wire to the clips.

Effect: DC resistance "from clip to clip" droppe to less than 25 % of the original, for the same cable length.

Maybe, if those cables are part of your parallel resonant circuit on the test bench, they are a show stopper - besides the close aluminium shield, which should really be a bit closer away from the rod / coil, or be slotted, as already suggested by Peter.

Cheers,
   Wolf .

Am 28.09.2011 21:05, schrieb Daniele Tincani:

Hello Stefan,

 

this morning I received the litz wire and tried another build, using a 16 mm PVC tube to enclose the 3 x ferrite rods. I wound 70 turns of litz (as good as I could) and went to the test bench again to measure Q (please see the attached picture). In fact, things have improved a bit. With the air varcap I'm using for these tests, tuning can range from about 235 KHz up to about 978 KHz. Using nominal min/max valued of capacitor, this gives approx L = 670uH which is very good for my purposes.

Unfortunately Q - although improved - is still quite low: about 104 @ 235 KHz, about 60 @ 978 KHz.

As I don't plan to try another build of the ferrite-cored coil, I can only hope that Q will improve further with shorter and better connections with the capacitor.

Many thanks to you, Stefan, for the very helpful suggestions!

 

Best regards

Daniele

 

Da: Stefan Schäfer <Stefan.Schaefer@iup.uni-heidelberg.de>
A: rsgb_lf_group@blacksheep.org
Inviato: Martedì 27 Settembre 2011 22:42
Oggetto: Re: LF: Re: My first ferrite experiments

Hi Daniele,

Am 27.09.2011 17:15, schrieb Daniele Tincani:

Hello Stefan,

 

> but maybe the Q is reduced by the aluminium shield!

Do you think the aluminium could be the problem or the distance between the coil and the core?

The wire shouldn't wound directly on the rod. I have a PVC tube above the rod and so the wire is in a distance of about 3 mm to the rod. I used litz wire and got a Q of 485...

I could replace the aluminium tube with a plastic tube, if shielding is not necessary, but this would not improve the Q if the problem is the gap between the coil and the core.

Just experiment what is the best solution. But the aluminium shield - if used - should be thin and can be applied above the winding in some distance. I know you have much QRM at home so probably an E field shield makes sense here. Anyway you will be bothered by the local magnetic QRM of course.
I personally think that a shield is not necessary if you are /p and far away from man made noise. Then, a higher Q helps more than the shield and a Q of 30!

 

> Are the rods galvanically separated from each other by a thin plastic tape? What type is the ferrite material?

Yes I put two layers of plastic tape on tips of each rod. The ferrite material is russian MU400HH.

Fine! I also use the same ferrite type and 55 turns. Your cap is surely the best what you can use. So shorten the wires between L and C and measure the Q on 137 kHz. If it is still in the range of 30, i assume the shield causes high losses.
Another idea: There is some conducting spray available. Maybe this can help here? No experience with that, just a spontaneous idea...
http://it.farnell.com/_/kontakt-chemie-emv-35-200ml/rivestimento-conductive-emv-35/dp/220115

73, Stefan/DK7FC