Antenna matching at 2 MHz

[vittovitto]

Hi all,

I am designing a transmitter-receiver system working at 2 MHz. This is not for radio communication, but for environment sensing. There is no bandwidth to worry about and the power is low (<1W). My antennas are small loops (approx. 4 inch) diameter and are each connected to a tiny tx and rx boards respectively.

I was given the schematics of these boards, which are very simple. The transmitter consists of a simple matching network, to match the impedance of the antenna to the 50 ohm feed line and to bring resonance at the desired frequency (2 MHz). The surprise was when I looked at the receiver: there is no matching network whatsoever. The antenna is fed directly into a differential amplifier and that's it. When I asked them why they told me "we tune the transmitter, then the receiver just picks up whatever".

Does anybody have suggestions on how to design a simple matching network to match my receiving antenna to the receiver amplifier? I don't have much experience working in this frequency band. Any example circuit that could help get me get started?

Thanks!

 

[VEBill]

They're correct that it's not critical to match the receiver to the antenna at these relatively low frequency ranges. Here's why:

The Signal to Noise Ratio (SNR) at 2 MHz is dominated by external (atmospheric or man-made) noise. The internal noise figure of the receiver is not critical. The efficiency of the receive antenna system is also not critical (to a large extent). Matching the antenna is typically not critical. The SNR is set *external* to your entire receiving system (antenna and receiver).

The net result of all this is that you could use a receiver with a much lower noise figure; you could carefully match the receiving antenna system for maximum efficiency; and it would not make any difference whatsoever.

This only applies at relatively low frequencies. Somewhere in the higher end of the HF band (perhaps 25 MHz) this rule about external noise becomes untrue and noise figure and antenna matching start to become important. These factors become critical at VHF and above.

To be clear, if the 2 MHz receiver system is a complete botch it can still be made to not work.

 

[vittovitto]

Hi VE1BLL,

Thanks for the answer. What you say makes sense. However, don't you think that SNR and matching cannot be treated the same way?

The environment where the system will be operating will be relatively noise free. Everything will operate in a lab setting, a lot of care is put into shielding and we are interested in measuring mV if not uV at the receiver side. So, provided we have a decent SNR obtained by carefully designing the receiver enclosure, then antenna matching should still matter to a certain extent. My small loop will have an impedance with a fairly small real part and an almost completely inductive imaginary part. I connect that to an amplifier system with an input impedance of about 50 kOhm purely real. The mismatching will be very big and I won't have very good power transfer between antenna and op-amp.

For example here:

http://sidstation.loudet.org/antenna-theory-en.xhtml.

These people worry about antenna matching at 20 kHz!

 

[VEBill]

"These people worry about antenna matching at 20 kHz!"

They're not matching. They're resonating loops (using math instead of a variable capacitor like everyone else) and also exploring the effective Q with various different (unmatched) loads.

"...antenna matching should still matter to a certain extent."

Only to the extent that your desired signal and any external noise rise above the noise floor generated internally in your receiver. At 2 MHz the receiver noise floor should be very low, so matching as such is typically a non-issue.

As I mentioned, yes you could bork it up if the system is really badly designed.

If your transmitters are at least close to 100 mW ("<1W") and you're in a quiet shielded enclosed environment, then I don't see how you could have weak signals unless the lab is the size of a football pitch.

 

[vittovitto]

VE1BLL,

OK, I was confusing between resonating and matching, thanks for making it obvious.

Regarding the signal level just ignore it: my signals are really tiny, I wrote <1 W to make it clear that I'm not operating a devices that needs to move a lot of power.

Now, getting back to the original question: the receiver board right now feeds directly the loop into a differential amplifier with an input impedance of 50 kOhm. Assuming that I don't care about matching, I would still want to adjust the resonance point of my receiving loop to tune it to the right frequency, right? I would ballpark the value of the capacitor it with the math and then put a variable capacitor to fine tune it.

 

[VEBill]

Resonance can help improve the selectivity of the antenna system. Selectivity will reduce the amplitude of noise and unwanted signals on other frequencies.

To keep the Q high, the receiver input should be high Z input.

Beware extreme Q because 1) the resonance peak might drift off frequency, and 2) if your signal is not narrowband it might impact the sidebands (if any) [this is not likely].


Keep in mind that there are several digital techniques to add coding gain. Barker codes and various Spread Spectrum techniques can provide ten or more dB coding gain if required. This might be useful if your project involves micro-controllers at each end.

If you want to worry about antenna matching, review the Transmit side. Small loops (at low frequency) are often difficult to feed; the impedance is typically very low, and the current and resistive loss can be relatively high.

 

[Higgler]

You are essentially operating at AM Radio frequency-ish (1.6 Mhz is high end of AM). If your antenna is a simple loop, with no ferrite core, substitute an AM radio antenna for comparison to your current setup. Might be better.