Author Topic: Strange behaviour of RF circuit  (Read 1544 times)

0 Members and 1 Guest are viewing this topic.

Offline VulpeculaTopic starter

  • Newbie
  • Posts: 8
  • Country: de
Strange behaviour of RF circuit
« on: October 01, 2021, 09:13:53 am »
Hi there!

I have an RF circuit which behaves strangely and since I am pretty new to these kind of circuits I need some advice on how to solve or at least narrow down that problem. I have attached a brief overview of the circuit.

The circuit is supposed to receive AIS Data on VHF channels 87B (161.975 MHz) and 88B (162.025 MHz). It does to that, however, only under certain circumstances. It works whenever I attach a simulation device (QK-AT011 AIS Transmission Simulator) directly to the antenna input. In that case I do have a signal and I am able to decode the simulated AIS sentences.

Here is the strange part: I do not receive any signals OTA when attaching a real AIS antenna. But once I start probing around with a spectrum analyzer, I am able to decode incoming messages for as long as my probe is touching the signal path anywhere on the board. Also, it does not matter if the ground of the probe is connected to my circuit or not. (I am using a rather cheap TinySA handheld spectrum analyzer here.)

I tried eliminating different parts of the circuit (e.g. removing the Chebyshev filter or the SAW filter or both and so on) but the problem stays the same. As I said: I am pretty new to RF circuitries and I don't really know what could cause this problem or where I should start looking. Maybe someone out there has some ideas and is able to point me into the right direction.

Thanks and greetings
Vulpecula
 

Offline TheMG

  • Frequent Contributor
  • **
  • Posts: 866
  • Country: ca
Re: Strange behaviour of RF circuit
« Reply #1 on: October 01, 2021, 01:39:59 pm »
Could be the receiver front end is being overloaded by a strong signal off-frequency, and the extra loading from the spectrum analyzer is attenuating that down just enough for the receiver to no longer be overloaded.

Have a look up and down the full band and see if there is the presence of any excessively strong signals. It could even be from a nearby FM broadcast depending how good the front end filter of the receiver is at attenuating out-of-band signals.

Another possibility, the antenna (or its cable) that you are using is faulty and the probe itself is acting as an antenna.
« Last Edit: October 01, 2021, 01:42:02 pm by TheMG »
 

Offline Joel_Dunsmore

  • Regular Contributor
  • *
  • Posts: 168
  • Country: us
Re: Strange behaviour of RF circuit
« Reply #2 on: October 01, 2021, 03:32:57 pm »
Another possibility, the antenna (or its cable) that you are using is faulty and the probe itself is acting as an antenna.
Very likely this second, but I would say differently, with antennas, size matters.  Especially at VHF.  You probe is a BIG antenna and your ground is the earth (the SA is grounded through the mains).  It is similar to the magic engineer's figure where the circuit works when you touch "here" but stops when you take your finger off.

You might consider trying to connect your antenna to your SA and see if you can see a signal on the SA. And, just for fun hook a 1 foot wire to your receiver instead of your antenna. if it works under that condition, well then you know.
 

Offline VulpeculaTopic starter

  • Newbie
  • Posts: 8
  • Country: de
Re: Strange behaviour of RF circuit
« Reply #3 on: October 01, 2021, 06:58:57 pm »
First off...: I tried different pigtail adapters (the antenna is connected through an adapter to a SMT U.FL socket). I also tried different AIS antennas I had on hand. Just to be sure I tried the adapters and antennas with a (known working) commercial AIS receiver where everything works as expected.

What I also did was trying a different DC blocking cap: The original circuit wanted a 10pF C0G(NP0) 0603 which I did not have. Instead I soldered a 3.3pF cap to the board at first. Now I've replaced that cap with a rather disadvantageous combination of three 3.3pF caps in parallel which reside on another PCB since I had no room to put three of those caps on the actual board. That seems so have fixed the problem somehow since I am receiving data without having to throw the SA probe at it plus it looks like I even gained 5dB in signal strength. Also, I tink that the extra PCB delivers more capacitance than just the three times 3.3pF.

Well... I've ordered some 10pF caps now and we will see how far that will get me.  :-// Anyway... I am trying to understand how and why the original cap of 10pF was chosen. I do not know how to apply some of the formulas and what factors are important to that. Does the cap (which is in series) act as some kind of high pass filter in this case?
 

Offline VulpeculaTopic starter

  • Newbie
  • Posts: 8
  • Country: de
Re: Strange behaviour of RF circuit
« Reply #4 on: October 01, 2021, 07:06:05 pm »
Especially at VHF.  You probe is a BIG antenna and your ground is the earth (the SA is grounded through the mains).

I have some pretty big antennas here. Actually different sizes of GLOMEX antennas. Also: My SA is a cheap one, which uses battery power, not mains power.
 

Offline shadow.dark

  • Contributor
  • Posts: 46
  • Country: cn
Re: Strange behaviour of RF circuit
« Reply #5 on: October 02, 2021, 12:47:47 pm »
Is the antenna directly connected to the SAW filter? In my impression, there will be some matching problems. The impedance change of the antenna will affect the pass band of the SAW filter. I usually use a low-gain amplifier to separate the two.
 

Online Wolfram

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: no
Re: Strange behaviour of RF circuit
« Reply #6 on: October 06, 2021, 11:25:40 am »
Your 3.3 pF coupling cap is -j300 ohms at 160 MHz, so it will for sure have some impact on the signal level. Even the original 10p is significantly higher than the circuit impedance, is it intended as some sort of matching network for the antenna or SAW filter? In that case, the value could be critical and might have to be tuned for your particular parts.
 

Online iMo

  • Super Contributor
  • ***
  • Posts: 4673
  • Country: nr
  • It's important to try new things..
Re: Strange behaviour of RF circuit
« Reply #7 on: October 06, 2021, 08:01:56 pm »
Would be better to show the schematics (at least the input stage)..
 

Online radiolistener

  • Super Contributor
  • ***
  • Posts: 3281
  • Country: ua
Re: Strange behaviour of RF circuit
« Reply #8 on: October 06, 2021, 09:01:06 pm »
Here is the strange part: I do not receive any signals OTA when attaching a real AIS antenna. But once I start probing around with a spectrum analyzer, I am able to decode incoming messages for as long as my probe is touching the signal path anywhere on the board. Also, it does not matter if the ground of the probe is connected to my circuit or not. (I am using a rather cheap TinySA handheld spectrum analyzer here.)

I see the following possible issues:
- bad untuned antenna
- bad umatched transmission line
- bad design input RF frontend on your circuit (unmatched transmission lines, reflections and other issues due to bad PCB layout or mistake in RF circuit)
- burned out input port of your device

Since you can receive signals when connect some metal thing to the transmission line, it looks that your input port is not burned out. But it worth to check it. In order to do it, try to connect your signal simulator through RF attenuator to get low signal level, just about 10-20 dB higher than guaranteed device sensitivity. If it works, then port is not burned out.

For other issues, you're needs Vector Network Analyzer in order to test your antenna, transmission line and device input impedance.

Could you show the photo of your connections between device and antenna? It can help to reject simple mistakes.

What I also did was trying a different DC blocking cap: The original circuit wanted a 10pF C0G(NP0) 0603 which I did not have.

that part of your circuit looks very suspicious, it may be the root of cause of your issue.
Could you show it's schematic and the photo of PCB?
« Last Edit: October 06, 2021, 09:11:08 pm by radiolistener »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf