3
\$\begingroup\$

I am working on an VHF down-converter using the SA602. The RF input impedance to this chip is 1.5 kΩ. I understand that hooking up to most antennas will require a 50 Ω matching network. However, for this case I am wondering if I need a matching network at all.

  1. Matching is to avoid losses from reflection. I am Using the equation line length/wavelenth > 0.01, so any line length ratio to wavelength ratio larger than 0.01 will have reflections.
    At 140 MHz, if I keep the trace length to the antenna connection shorter than 1 inch, then transmission line effects can be ignored, correct?

  2. Even assuming a short transmission line, do I still need to design the antenna to have 1.5 kΩ impedance to satisfy the maximum power transfer theorem? Or would any antenna (receive only) be fine so long as it is tuned for 140 MHz?

Edit: To sum this up, I can design the antenna with 1.5 kΩ to match the IC input impedance of 1.5 kΩ. I understand 50 Ω would be needed if I were using a 50 coax or stripline, but the trace is around 1 inch. The antenna connection would be direct, no SMA or BNC (since those are 50 Ω).

I wanted to bounce this idea of off some RF gurus out there.

\$\endgroup\$
5
  • 1
    \$\begingroup\$ How far is the antenna from the board? \$\endgroup\$
    – user319836
    Commented Sep 12, 2022 at 1:12
  • \$\begingroup\$ I am considering using off center fed loaded half-wave dipole. The connection from radiating element (wire) to the chip is less than 1 inch. I was planning on testing feed points to get the1.5k match along the wire antenna. \$\endgroup\$ Commented Sep 12, 2022 at 1:31
  • 1
    \$\begingroup\$ Check with the HAM radio folks. OR the ARRL manual. \$\endgroup\$
    – user319836
    Commented Sep 12, 2022 at 1:50
  • 1
    \$\begingroup\$ Having very wide bandwidth at the SA602 input is asking for trouble with interfering signals at other frequencies...this chip handles strong signals poorly. A shunt inductor at the input (pins 1 to 2) is a very good idea. It would be used to compensate for the 3pf input capacitance added to stray capacitance added to capacitance of the short (1") transmission line. I'd actually add some capacitance, so that this tuned-input has some frequency-selectivity. \$\endgroup\$
    – glen_geek
    Commented Sep 12, 2022 at 3:18
  • \$\begingroup\$ @glen_geek excellent point, thank u \$\endgroup\$ Commented Sep 12, 2022 at 3:50

2 Answers 2

3
\$\begingroup\$

To sum this up, I can design the antenna with 1.5 kΩ to match the IC input impedance of 1.5 kΩ. I understand 50 Ω would be needed if I were using a 50 coax or stripline, but the trace is around 1 inch. The antenna connection would be direct, no SMA or BNC (since those are 50 Ω).

You are doing it backwards. First design the antenna the way you want it, then match its impedance to the IC for best performance (ie. highest gain and selectivity). So the question you should be asking is, "What is the impedance of the antenna I want to use?"

Generally you also want a tuned circuit to help reject out of band interference and image frequencies. This provides an easy way to match any antenna impedance via a tapped coil, extra winding or series capacitors.

\$\endgroup\$
0
\$\begingroup\$

If you place a 50 resistor across the connection point or across the input to the chip, you will be fine. You definitely need the 50 ohm termination for maximum power transfer.

Edit: Impedance of free space is 300 ohms. I haven't tried to match to 1.5k.

I guess to answer your question, "Yes you still need a matching network." The antenna is the matching network to free space. An appropriate tap may be possible. I would lower the impedance to 300 ohms or less. It may be easier to get a match,

\$\endgroup\$
6
  • 8
    \$\begingroup\$ Adding a 50 ohm resistor is just going to lower the signal into the ,mixer. You might get maximum power transfer but most of the power will be transferred to the resistor. If you're worried about maximum power transfer you use a transformer or matching network. \$\endgroup\$
    – GodJihyo
    Commented Sep 12, 2022 at 1:36
  • \$\begingroup\$ @GodJihyo, so if the antenna was fed at a 1.5k point, and the pcb trace was short it would be an acceptable match into the IC correct? \$\endgroup\$ Commented Sep 12, 2022 at 1:38
  • 1
    \$\begingroup\$ Are you connecting the antenna elements directly to the input with no feedline? \$\endgroup\$
    – GodJihyo
    Commented Sep 12, 2022 at 1:44
  • 1
    \$\begingroup\$ @GodJihyo Yes, off-center fed to get 1.5k ohm impedance on the rx antenna \$\endgroup\$ Commented Sep 12, 2022 at 1:53
  • 1
    \$\begingroup\$ @user3425949: You should try it. Try several feed points with the appropriate matching resistor. Choose what gives the best performance. \$\endgroup\$
    – user319836
    Commented Sep 12, 2022 at 2:41

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.