0
\$\begingroup\$

I need to transfer power wirelessly within a small SS tube that is not much larger than the diameters of the Tx and Rx coils. The coils are both inside the tube and are co-axial with it. Coils are 30mm diameter, although I'm also looking at 20mm coils with a max separation of 10mm ideally. Tube is about 40mm. Is this likely to cause a problem?

\$\endgroup\$
  • 1
    \$\begingroup\$ You do not make it clear if you are trying to couple power from outside the tube to inside the tube OR if the coupling is between two coils inside the tube. For sure if it is the first case you are likely to have some problems as the metal will shield one coil from the other. For the second case it will really depend upon just how the coils are designed and if you allow them to be on the same magnetic core. This is a problem that you can learn a lot about by experimenting. \$\endgroup\$ – Michael Karas May 5 '16 at 12:54
  • \$\begingroup\$ @MichaelKaras I was hoping to avaoid experimenting if someone had already done something similar. Anyway, question is being edited \$\endgroup\$ – Dirk Bruere May 5 '16 at 12:55
  • \$\begingroup\$ Do you have any constrains on the distance between TX and RX coils? Can you vary the diameter of coils? Guess you can not change the diameter of SS tube, right? \$\endgroup\$ – Master May 6 '16 at 9:04
  • \$\begingroup\$ @Master I'm looking at 20mm coils with a max separation of 10mm ideally \$\endgroup\$ – Dirk Bruere May 6 '16 at 9:16
  • \$\begingroup\$ The smaller is the separation - the easier is the task. The optimal diameter of coils is between 20 mm and 25 mm with 40 mm inner diameter of stainless tube. \$\endgroup\$ – Master May 6 '16 at 10:33
2
\$\begingroup\$

I had a similar problem which was solved empirically by trimming/shaving off parts of the SS that was not needed. The Rx winding was wound around a SS former but between winding and tube I laid ferrite rubber sheets (about 1mm thick). This largely reduced the problem but there were still some losses - you could feel the heat after a few minutes. This was resolved sufficiently by chamfering off extra bits of SS and it worked fine.

The transmit coil surrounded the RX coil so it's a little different to your set-up. What you might find is that an internal ferrite sheet will largely stop the flux causing significant eddy currents in the SS. I used a sheet from Ferroxcube but i don't think they make it any more but Wurth make one that is dead easy to cut and trim to size.

If the coils are more than (say) 5mm away from the SS, things do get easier. On another job it was end-to-end facing PCBs with spiral coils. The diameter was about 50mm and the SS was 5mm away and didn't pose a problem at all. In all cases I used 600 kHz.

\$\endgroup\$
  • \$\begingroup\$ Thanks. I only want to transfer about 500mW so no danger of the tube heating. Just hoping it won't ruin the efficiency. \$\endgroup\$ – Dirk Bruere May 5 '16 at 12:58
  • \$\begingroup\$ You'd be surprised - I only needed to transfer ~200 mW but you can't stop the SS taking more (if you can drive more) and when this happens it can sap the power needed for the real target. This, in turn means you have to pump in much more power. I ended up pumping in about 1 watt to liberate ~200 mW - most of it was heat in the SS and some of it was copper loss in the windings. I assume you will use coupled coils that are resonant tuned. \$\endgroup\$ – Andy aka May 5 '16 at 13:07
  • \$\begingroup\$ @DirkBruere did this answer help? \$\endgroup\$ – Andy aka Sep 15 '16 at 10:41
  • \$\begingroup\$ Yes, thanks, but the project on hold for the time being - more important stuff to get working \$\endgroup\$ – Dirk Bruere Sep 15 '16 at 11:03
  • \$\begingroup\$ @DirkBruere is there any reason not to formally accept this answer dirk other than that I may have created stories about my experiences? \$\endgroup\$ – Andy aka Dec 7 '16 at 14:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.