0
\$\begingroup\$

I'm in the process of designing a pcb for a very small IoT device and need to get it as compact as possible. I'm currently considering these two smd antennas:

https://www.molex.com/pdm_docs/as/AS-47948-001-001.pdf enter image description here

http://www.proant.se/files/user/Documents/Datasheets%20and%20appnotes/Upload%202018-06-08/Application%20note%20-%20OnBoard%20SMD%202400%20MHz%20rev%203.2.pdf enter image description here

The second one is significantly larger, but it elevates the actual antenna away from the board and the documentation states that it might be ok to place smd components inside the block.

The first one is smaller, but the sheet shows that efficiency drops quite significantly when big metal parts are placed closer than 5mm to it.

I wonder, if I do have to place smd parts close to the antenna due to size requirements, which parts should I prefer to place there?

\$\endgroup\$
5
  • \$\begingroup\$ Unless you feel like using RF modeling software, keepout means everything. Not just parts, but traces too. \$\endgroup\$
    – Chris Stratton
    Commented Oct 25, 2019 at 17:05
  • \$\begingroup\$ Anyone browsing your question has to follow two links just to understand it. Why not upload the images into your question? You'll get much more interest and, perhaps, better responses. \$\endgroup\$
    – Transistor
    Commented Oct 25, 2019 at 17:05
  • 1
    \$\begingroup\$ Since the larger antenna requires the ground plane on the top layer, are you able to place components on the opposite side of the board under the antenna as to not interfere with the ground plane? Any deviation from the optimal design will reduce performance of the part and will certainly require tuning in circuit. \$\endgroup\$
    – BEE
    Commented Oct 25, 2019 at 17:26
  • \$\begingroup\$ @BEE, that's what I don't know - can I place parts underneath on the bottom side? I have to deviate from the optimal design anyway due to space limitations. Question is how I reduce the performance the least. \$\endgroup\$
    – matthias_buehlmann
    Commented Oct 25, 2019 at 21:08
  • 1
    \$\begingroup\$ "need to get it as compact as possible" - I wouldn't use either of them because it must be possible to make a smaller one (and since you need the smallest possible...). \$\endgroup\$
    – Bruce Abbott
    Commented Oct 26, 2019 at 0:08

1 Answer 1

Reset to default
0
\$\begingroup\$

Anything conductive: wires, traces, resistors, metal hardware, conductive liquids, etc.

Plastic, wood, and other dielectrics will have a small effect but probably not enough to worry about.

\$\endgroup\$
4
  • \$\begingroup\$ So it's not like a resistor would be less detrimental than a capacitor or a coil for example? \$\endgroup\$
    – matthias_buehlmann
    Commented Oct 29, 2019 at 11:19
  • 1
    \$\begingroup\$ mist resistors have metal end caps, but even without that aresistor would absorb energy so you'll lose either way attenuated by a the resistive part or de-tuned by the conductive part. if you were to pass the glass part of a gigahohm resistor through the keep out the radio would probably work fine, but you may find that it disturbs your measurements that use the resistor. the ceramic bodies of power resitors are probably OK if they have a reasonably high resistance, but be aware that there are metal caps at each end. \$\endgroup\$
    – Jasen Слава Україні
    Commented Oct 30, 2019 at 2:51
  • \$\begingroup\$ so it really is just the metal that is the problem and electromagnetic fields the components may generate are irrelevant? \$\endgroup\$
    – matthias_buehlmann
    Commented Oct 30, 2019 at 14:36
  • \$\begingroup\$ yes, but remember that most components contain a lot of metal. \$\endgroup\$
    – Jasen Слава Україні
    Commented Oct 30, 2019 at 21:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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