Why hardware filters are effective and necessary with SDRs for radio astronomy in slite of software filtering

[Andrew Thornett]

Here are the results of my research into this subject. (This has NOT come from AI.)

The following explains why RF filters (preselectors and band-stop filters) are important when using wideband SDRs such as the RTL-SDR.

Main points:

Wideband SDRs receive from about 24–1700 MHz, making them vulnerable to strong unwanted signals.

A preselector (band-pass filter) allows only the frequencies you want to reach the receiver, reducing interference and improving the signal-to-noise ratio.

Unlike many traditional radios, the RTL-SDR and many other cheaper SDRs do not have built-in preselectors, so external filters are often needed.

If your application is fixed (e.g. the 1420 MHz hydrogen line), a dedicated preselector is definitely worthwhile.

Why hardware filters make an important difference?

This is due to ADC overload, which can be explained by considering the scenarios below:

First scenario: Gain is set low. Strong FM, pager and GSM signals are present but below the ADC overload limit. The desired weak signal is visible, although weak. Note the FM and other signals are not at the desired frequency.

Second scenario: Gain is increased. The strong broadcast signals now exceed the ADC overload threshold and overload the SDR's analogue-to-digital converter. As a result, false image signals appear and the receiver becomes less sensitive. The desired weak signal is not improved, despite the higher gain setting on the SDR. This is in spite of the FM signal etc not being at the same frequency as the desired weak signal. Overloading the ADC at these out of band frequencies stops the SDR from being able to improve the gain of the desired weak signal.

Third scenario: Adding in a band stop filter removes the out of band signals so that, when gain is increased, the ADC is not overloaded and the desired weak signals gain is able to be increased, improving the detectability of the desired weak signal.

Key lesson:

For radio astronomy—especially 1420 MHz hydrogen-line observations—this is highly relevant. Rather than simply increasing the RTL-SDR gain, it is usually better to:

use a 1420 MHz band-pass filter (preselector),

reject strong out-of-band signals (FM, mobile phone, paging, etc.),

then apply gain with a low-noise amplifier (LNA).

This approach gives a cleaner spectrum and allows weak astronomical signals to be detected without overloading the SDR.

[Stephen Arbogast]

In what  form should that  analog filter be?

[Andrew Thornett]

I only got as far, in my research this time, as to why a hardware filter was a good idea, because it has been suggested that the ability to filter in software means a hardware filter wasn't needed - but that turns out to be incorrect.

In terms of what hardware filter, well, it ks a band pass filter of some type.

From personal experience, and from extensive discussuin in this forum and in H-Line group, I would recommend, for H-Line, Nooelec SAWBird H1 LNA/filters which work very well.

Other alternatives exist and can work well, but tend to be more complicated to use in practice and more expensive - e.g. interdigital filters - which I have discussed previously.

Andy

Sent from Outlook for Android

---

From: 'Stephen Arbogast' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Sent: Thursday, 09 July 2026 03:39:20
To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: [SARA] Re: Why hardware filters are effective and necessary with SDRs for radio astronomy in slite of software filtering

 

--
--
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/6e355e62-17d0-45e7-a8aa-f7871d4bcea7n%40googlegroups.com.

[Stephen Arbogast]

In my  humble opinion....  we don't  want  any  digital  stuff  near our  antennae so we use  analog filters.  Back in  the  1960's  I was lucky to pick up  a  used Collins   S-Line......the receiver   had an  analog  mechanical   filter  ...  Collins  Radio   pioneered this ... it was very  good!   I was WA0TIU  Advanced Class.

Another   subject I would like  to say a  few words about...  Some  one  mentioned in the recent  BAA  Hydrogen  Line  online meeting  that  ezRA  is a lot  to learn...  yes it  is  but  well worth it.   There is not much  post processing  software available...   I  think  ezRA  is  wonderful  thanks to Ted Cline..   

I am still  learning ezCon     ...  Doing  a study  where  I  ran  ezCol at   10 Meg samples per  second  with  ten samples between  reference samples. on  AirSpy-R2    I ran  the plots...   Then I  removed all  of the  reference  samples and ran the   plots again.  By  removing   all of  the  outliers   I am  getting some  good plots..  sort  of  good... I want to do  more  before publishing  both sets of data.

Stephen

[fasleitung3]

A couple of years back I published a series of articles which, among other things, also covers the topic of filters for the 1420 MHz region. The relevant part is section 4 of https://astropeiler.de/wp-content/uploads/2020/01/Hydrogen_2.pdf

At that time, the Noolec SAWBird was not availabe, so it is not covered. However, the Noolec SAWBird uses a SAW filter (hence the name) which is covered in section 4.5 of the article.

Beyond this, I highly recommed a cavity filter designed by Matjaz Vidmar which is coved in section 4.7. This is vey easy to make as it does not require a milling machine.

Best regards,

Wolfgang

[Stephen Arbogast]

Wolfgang,

I  am using  the Discovery Dish   and  Discovery  Feed with  LNA  which  has a SAW filter...    with  AirSpy-R2

The problems    I  think  most people have  are  with   post  processing   the  raw data.......

The  hard part   is to sort out the  data.....   throw away  the  outliers .....  many different    ways   to   process   the raw  data.  In my opinion   there is no  single recipe  to do this...  It  depends   on  one's  situation.....

Stephen

[Robert Hamers]

Thanks, Wolfgang-

I looked for the paper by Vidman but was not able to find it. I know the dimensions of the filter are critical to its operation. Is there somewhere where the actual dimensions are listed?  If so, perhaps you could post that to the group?

Best wishes,

Bob H.

Get Outlook for Mac

From: 'fasleitung3' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Date: Thursday, July 9, 2026 at 2:10 AM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Why hardware filters are effective and necessary with SDRs for radio astronomy in slite of software filtering

--

--
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/10e6617fe013253dac495288a371d02cdad7fcef.camel%40googlemail.com.