Observation: Airspy Mini and HackRF One under my local RFI and non-RFI conditions

[Ayushman Tripathi]

Test environment: SDRangel and VIRGO running on Ubuntu.

I was testing my local setup, observing near Capella (see attached Stellarium screenshot). I noticed that when my monitor is on, the H-line becomes very difficult to distinguish with the Airspy Mini in SDRangel, while it remains more clearly visible with the HackRF One. I have observed this several times over the last month.

I’ve also observed that if the RFI increases beyond this level, I stop getting usable results from the Airspy Mini and then even the RFI is not visible in the spectrum, at that point the spectrum becomes flat and then develops wavy small lines (I'll try to post the images of that soon) but the HackRF still gives the result.

Both SDRs were tested in SDRangel with same pointing and max gain settings. With the monitor off (reduced RFI), both perform similarly.

Attached:

• Today's SDRangel screenshots from both SDRs, with and without RFI

• VIRGO plot from HackRF with RFI, captured at the same time and pointing

I'll run longer observations and then compare again. Tried various Airspy Mini settings but couldn't improve results in the high-RFI condition when it becomes wavy (not shown in these images). Can anyone suggest why I'm not getting good results with the Airspy Mini? As I believe maybe I'm doing something wrong.

Thanks

HackRF without Monitor RFI: 

Airspy without Monitor RFI: 

HackRF with RFI: (Monitor ON)

Airspy Mini with RFI:

[Alex P]

The config for the ASMini sets the signal power level @ 4X that of the HRF   so you would expect a Higher RFI signal ?

[b alex pettit jr]

Due to the higher gain , perhaps out of band RFI is causing clipping/saturation = higher apparent RFI ?

[Ayushman Tripathi]

Hi Alex,

Yes, possibly. I've observed this several times with my setup, the airspymini sometimes doesn't even show the RFI, just a flat spectrum with small wavy repeating structures. Once the RFI is reduced, it starts working again. The HackRF, however, keeps working fine and clearly detects the H-line. I'll test more. 

I'll share that wavy spectrum image soon.

Thanks. 

[b alex pettit jr]

The Hack has that mid band notch that I prefer to Not have in my data 

Remove the source of  RFI 

it also causes saturation/blocking  non linearity in the LNA amps ( that is not apparent in a spectral display

[Ayushman Tripathi]

Hi Alex,

Yes, for sure the effort should be in reducing the RFI because once it saturates the LNA front-end, no amount of software processing can recover the signal, but I wanted to share this as an observation because sometimes RFI can't be controlled (e.g., from neighbors).

The HackRF's DC spike can be fixed using calibration (e.g., like VIRGO does in the above plot) or by offsetting the center frequency away from the target, but yes it's always better to not have it in the first place.

I'll work on reducing the RFI, but as a test I'll post longer observations using both SDRs with and without Monitor/PowerStation RFI.

Thanks

[b alex pettit jr]

The overall background level has shifted upward : much more with the ASmini

Less with the Hack

Reduce the Gain of the AS mini so the non RFI levels of it matches the Hack ( from -23 dB to -28 dB ) and ReTest

[Stephen Arbogast]

All....  I finally  solved my major RFI problem  by placing my older  iMac  from 2012  between my  Dell  monitor  and my  dish.

See...  photo

I  still occasionally  get  RFI  from neighbors.  Oh well.

[andrew....@googlemail.com]

Is the older Mac turned on or acting just as barrier?

 

From: 'Stephen Arbogast' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Sent: 26 January 2026 23:31
To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Observation: Airspy Mini and HackRF One under my local RFI and non-RFI conditions

 

All....  I finally  solved my major RFI problem  by placing my older  iMac  from 2012  between my  Dell  monitor  and my  dish.

See...  photo

I  still occasionally  get  RFI  from neighbors.  Oh well.

 

 

On Monday, January 26, 2026 at 11:27:50 AM UTC-7 b alex pettit jr wrote:

The overall background level has shifted upward : much more with the ASmini

 

 

 

Less with the Hack

 

Reduce the Gain of the AS mini so the non RFI levels of it matches the Hack ( from -23 dB to -28 dB ) and ReTest

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/39f68435-c54a-48b3-9aeb-01444e7f8c4dn%40googlegroups.com.

[Stephen Arbogast]

My older iMac  is  not  turned on  and not  plugged in,  not  even  grounded.

If one  does not have  an old iMac  I think  some  wire mesh screen  would  work just  as well.

[Stephen Arbogast]

I  have  had  the  same problem  with my HackRF One, turn the  gain   down  to avoid  distortion.

[Stephen Arbogast]

HackRF  One  was not meant to be a weak  signal  device  even  with a  low  noise figure   LNA  in front.

It works  fine  for  exploring  and  experimenting  with strong terrestrial  signals.  

[Stephen Arbogast]

Thus  the name  HackRF

[Stephen Arbogast]

As  I recall  HackRF  was not meant  to be used with  an  LNA, too much  gain causes distortion.  Great  if you are  a Rock  guitar  player!

[Stephen Arbogast]

If  you  want  overdrive  and  distortion  try  this  circuit ..   I  built  one  and it sounds  great!

https://www.google.com/url?sa=t&source=web&rct=j&url=https%3A%2F%2Frobrobinette.com%2FHow_Amps_Work.htm&ved=0CBYQjRxqFwoTCNiuy_zNqpIDFQAAAAAdAAAAABAH&opi=89978449

[Stephen Arbogast]

[Stephen Arbogast]

Mine  is  the  Fender  5C1

[Stephen Arbogast]

A sample....   https://www.youtube.com/watch?v=a3mqDspcgtM

[Stephen Arbogast]

Sound sample..   https://www.youtube.com/watch?v=bf7SmWjWXPQ

[Stephen Arbogast]

In summary...   do not  use an LNA  with  HackRF... lower  the  gain  in  your  HackRF.

[Stephen Arbogast]

HackRF  is not  the  best  choice  for   noise figure.

[Stephen Arbogast]

What is  a  parametric  amplifier?   How  dos  it  work?

[Stephen Arbogast]

If  anyone  has  too much  snow  please send   some to us in Colorado.....

[Alex P]

IF You Can See The Feed, It Can See You & Your Surroundings

                                                     Surround The Dish With Window Screen to act as a Faraday Shield against RFI Reception

[Stephen Arbogast]

In my  case  not  necessary....     the second reflector  is   at  the focal point  of  my   dish   which   reflects  into  the   dipole   which   feeds  into   the     LNA....

[Stephen Arbogast]

90 %  of  the  time  no   problem...     

[b alex pettit jr]

incorrect

The "Reflector" may enhance the intended signal, but it Does Not Prevent Extraneous Signals from Being Received by the Feed Pattern

And being they are 10000X stronger than H Line .. they don't need to be Focused to Overwhelm the data

(   there is lessened reception in-line with the dipole length ... but not much )

[Stephen Arbogast]

The only  time   have   an  RFI problem is   about   10  %   some neighbor  runs  an appliance that   causes  problems.

[Stephen Arbogast]

Totally   disagree   with  you.... 

[Stephen Arbogast]

Problem  solved.. neighbor    running    cheapo  Bluetooth    device....

[Ayushman Tripathi]

Hi Stephen, Alex

Even though the HackRF is less sensitive, I'm observing (for my setup) that the Airspy Mini often struggles to clearly detect the H-line, even when slight RFI is present (at max gain - I'll test with reduced gain soon). The HackRF, however, consistently detects it well. The attached plot is from my 6-hour observation last night using the HackRF One.

Over the past month, I've noticed several instances where the Airspy Mini couldn't detect even a faint H-line, while the HackRF detected it consistently (as shown in my previous message). Sometimes when RFI levels get very high, the Airspy Mini stops showing even the RFI itself, yet the HackRF still detects the H-line.

I'm not sure why this happens, but my guess is that the Airspy Mini's higher sensitivity makes it more susceptible to RFI saturation, whereas the "deafer" HackRF handles it better.

By the way, I'm using the latest official HackRF One. Which HackRF do you have ? You can try testing same by turning your monitor on or off ( OR by removing that iMac :) )

ayushman@ayushman-i914900KS:/mnt/extra$ hackrf_info
hackrf_info version: 2024.02.1
libhackrf version: 2024.02.1 (0.9)
Found HackRF
Index: 0
Serial number:
Board ID Number: 2 (HackRF One)
Firmware Version: 2024.02.1 (API:1.08)
Part ID Number:
Hardware Revision: r10
Hardware appears to have been manufactured by Great Scott Gadgets.
Hardware supported by installed firmware:
    HackRF One

Thanks.
Attached are the plots from tonight's HackRF observation: 

Key Components:

Tuner
HackRF One: RFFC5071/2 + MAX5864
Airspy Mini: R820T2
ADC
HackRF One: MAX5864 (8-bit)
Airspy Mini: LPC4370 (12-bit)
Front-end LNA
HackRF One: MGA-81563 (switchable)
Airspy Mini: R820T2 internal

[Ayushman Tripathi]

And to fix the HackRF drift issues, I'm getting one of these OCXOs, as I can't find a cheap GPSDO.

[b alex pettit jr]

Hello Ayushman,

An Idea :

Run the AS Mini under  SDR#  and observe the real time spectra w/ w/o RFI  to validate the  Virgo Software .

 You should see the same background shifts / spikes 

( setup here )

HLine3D/21cm_Step2C_Software_SDR#_Setup_01_13Jul25.pdf at main · AP-HLine-3D/HLine3D

[b alex pettit jr]

w/ SDR#,   there are 3 gain settings available using the "Free" mode

[Stephen Arbogast]

I cannot  be  of  any   help   here    I  run   Air Spy  on   Linux.......

[Ayushman Tripathi]

Hi Alex,

Sure, I already have SDR# + IF Average set up. I’ll retest the Airspy Mini on that and share the results, though last time I tried I had similar issues. Also, it’s not only with VIRGO - I’ve seen the same issue with SDRangel sometimes. The Airspy Mini can’t detect the H-line, while the HackRF keeps seeing it even when pointing at very faint source.

Thanks

[Ayushman Tripathi]

Hi Stephen,

Yes, I also use it on Linux as it’s easier to automate for long-term observations and can run on remote setups like the Raspberry Pi 5. Also, it’s much easier to see and modify the source code - for example, I modified (Open Source) VIRGO to save the median-filtered data separately. I can try the .grc you made soon.

I think in my local environment the Airspy Mini is getting oversaturated, while the HackRF’s lower sensitivity prevents that. The RTL-SDR showed similar results to the HackRF.
Thanks

On Tuesday, January 27, 2026 at 6:06:31 PM UTC+5:30 Stephen Arbogast wrote:

[b alex pettit jr]

Hi Ayushman,

You may have identified the problem .. being the dynamic range of the Mini  ( OR the gain set by the software )

FYI,

I did an analysis of the AS Mini using a band limited RF radiated noise source into the Antenna > LNA > SDR.

The amplitude of the noise source was adjusted  into soft clipping at about +6dB above 290K Earth equivalent ( from tests )

and a series of nooelec SMA attenuators added to the input of the ASMini

Mini set to Max Gain as shown on previous post

results

[Stephen Arbogast]

HackRF      not for   radio  astronomy...   

[Stephen Arbogast]

You  must     adjust  the  gain    carefully  ..    will  be  spurs  that    that  you    remove   carefully  using ezCon.