1
$\begingroup$

I am using a USRP B205 to detect electromagnetic pulses. Sometimes the signal power bursts to 100 dB, average power is around 10 dB. What can be used to protect the SDR receiver against bursts?

$\endgroup$
5
  • $\begingroup$ Not an answer because this is the first time I've heard of this device, but: Check the Github in case others have a software solution via the gain API. github.com/EttusResearch/uhd/issues The datasheet for the AD9364 mentions it has auto-gain on the front end, so check to see if the product API exposes this and uses it correctly. There are steps for calibration that might be necessary first? $\endgroup$
    – user21417
    Feb 1, 2022 at 13:08
  • 4
    $\begingroup$ 100dBm? 100dBuV? 100dB dynamic range? $\endgroup$ Feb 1, 2022 at 15:40
  • $\begingroup$ @jdv,yes.I am not is there any method to calibrate SDR without power meter? $\endgroup$
    – kittygirl
    Feb 2, 2022 at 1:29
  • $\begingroup$ The USRP Hardware Driver and USRP Manual has a self-calibration section. I don't know if applies to your case, though. There is a lot in those manuals and datasheets, so I guess I'd make sure I wasn't missing something obvious. $\endgroup$
    – user21417
    Feb 2, 2022 at 3:04
  • 1
    $\begingroup$ As hobbs says, "100 dB" is not a power. $\endgroup$ Feb 2, 2022 at 10:58

3 Answers 3

3
$\begingroup$

There are a few types of overload you might be suffering from. Here are some possible solutions:

1. An in-band signal, stronger than your wanted signal

This may drive the AGC and make the wanted signal look smaller for a while.
It may also overload the receiver and cause wide band splattering and interference.

It might help to set the gain of the receiver lower.
Turn off AGC if you don't need it on.
You could also add an external attenuator to reduce the total input power. These start at about $20. enter image description here

2. A strong out-of-band signal

This can cause desense, cross-modulation and intermodulation, which will make the wanted signals look smaller, distorted, or produce other false signals on top of them.

Again, you can try reducing the gain and adding an external attenuator.

3. Very strong signals that can damage the receiver

Very strong signals can actually damage the receiver. This is usually caused by local radio transmitters like cellphones, aircraft transponders or Radar. There should be a specification for no-damage power. The USRP says "RX power -15 dBm max" but that's quite small, probably a maximum useful signal and not a damage level. Check with the manufacturer! Typical no-damage power is about +10 dBm.

Keep radio transmitters like cellphones away from the receive antenna.

Attenuators can help but these signals can be so strong that reducing them to a safe level will push the wanted signal below the noise.
In this case you need a limiter, an active device incorporating a diode (rectifier) that responds to the strong signal by attenuating everything only when the total input power is large. You may need to adjust the level with an attenuator after the limiter.

Here's one limiter from Pasternack:
enter image description here

Example specifications:

  • maximum input power: 100 watt peak
  • maximum output power: 18 dBm
  • no-attenuation power : 9 dBm
  • Cost from $500. ITAR-controlled so difficult to ship internationally.

You can also make a limiter with a diode or two ($2 each) and an inductor. This is from the MACOM datasheet:
enter image description here
The inductor carries the DC return, the rectified RF, generated by the diode, that keeps the diode conducting. For narrowband designs you could use a quarter wave line, for wide band you need an inductor with an SRF higher than the frequencies of interest. These can be small and expensive.

$\endgroup$
1
  • $\begingroup$ Does out-of-band signal go through ADC? $\endgroup$
    – WhiteGirl
    Feb 11, 2022 at 12:58
2
$\begingroup$

If your SDR has over 100 dB of dynamic range, I would attenuate the input path (and all parasitic paths!) down to where the average power is 10 dB above the MDS (minimum detectable level) of the SDR. Then a 90 dB PAPR might stay within the SDRs input range. Maybe a string of shielded in-line attenuators in a nesting of shielded boxes. A bigger issue might be how to shield and/or isolate the computer, and all the computer and power cables. They might have to all go inside a nest of shielded boxes as well (e.g. battery powered and monitored over optical fiber.)

$\endgroup$
2
  • $\begingroup$ Could you pls tell me some methods to protect SMA cable? $\endgroup$
    – kittygirl
    Feb 2, 2022 at 1:27
  • 2
    $\begingroup$ @kittygirl he literally did. $\endgroup$ Feb 2, 2022 at 10:59
1
$\begingroup$

Your receiver should be inside an enclosure for shielding, if it isn't already.

A common trick to prevent damage to a receiver from nearby transmitters is to put back-to-back diodes in the received signal path. See a snapshot of part of the Elecraft K2 schematic below. When the signal voltage exceeds the forward voltage of the diode, the diode conducts and shorts the signal, hopefully before the receiver is damaged. You would want to select the diode type to switch as fast as possible and have a maximum forward voltage that is low enough to prevent damage to your receiver.

excerpt from the schematic of the Elecraft K2, showing back-to-back 1N4148 switching diodes in the receive signal path

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .