Sawbird + H1 LNA

690 views
Skip to first unread message

Ron Coffey

unread,
Sep 17, 2021, 12:40:48 PM9/17/21
to Society of Amateur Radio Astronomers
All, I've been using the scope in the box for a few months now and seem to be having an issue with the LNA. Sometimes the LED power light is on indicating the LNA is working and other times its off when it should be on. This happens regardless of whether or no I'm using the SDR bias tee or the external 3-5 Volt port.  My 1st question, which will appear stupid is when the LED power light is off, is that impacting the results of the scan. My second question, maybe not so bad--- do I actually need it for the 1.4 GHZ signal. And thirdly, has anyone else see this and how did you fix it.

PS, I thought it might be a connection issue between the SDR and the LNA so I ordered new connectors that will arrive this weekend.

Thx

Ron 

Marcus D Leech

unread,
Sep 17, 2021, 2:20:45 PM9/17/21
to sara...@googlegroups.com
If the LNA is losing power, then it will be acting like a 15-20dB attenuator. So, you can’t just “live with it”. 

Sent from my iPhone

On Sep 17, 2021, at 12:40 PM, 'Ron Coffey' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:

All, I've been using the scope in the box for a few months now and seem to be having an issue with the LNA. Sometimes the LED power light is on indicating the LNA is working and other times its off when it should be on. This happens regardless of whether or no I'm using the SDR bias tee or the external 3-5 Volt port.  My 1st question, which will appear stupid is when the LED power light is off, is that impacting the results of the scan. My second question, maybe not so bad--- do I actually need it for the 1.4 GHZ signal. And thirdly, has anyone else see this and how did you fix it.

PS, I thought it might be a connection issue between the SDR and the LNA so I ordered new connectors that will arrive this weekend.

Thx

Ron 

--
--
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 on the web visit https://groups.google.com/d/msgid/sara-list/974ccfd2-9bdf-4be5-8d28-667b94a13500n%40googlegroups.com.

Hamish Barker

unread,
Sep 17, 2021, 5:29:27 PM9/17/21
to sara...@googlegroups.com
I have had sawbirds fail, but only because of getting wet. my weatherproofing technique has improved in direct proportion to the number of sawbirds I have purchased as I get more and more motivated. having the LNA exactly on the underside of the feedhorn isn't perfect because water runs around the circular feed then drips down the sma connector as it is the lowest point. Now I use a stout plastic bag over the feedhorn , plus a few holes at the underside of the bag to stop it from accumulating water.

--

Anthony

unread,
Sep 17, 2021, 5:40:34 PM9/17/21
to sara...@googlegroups.com
I've had one Sawbird H1+, LED light would glow dim. I tested other SAW LNAs with the same voltage and thier LED were bright. 

I believe my earlier learning I introduced more than 5volts to my first Sawbird LNA. 

Marcus D Leech

unread,
Sep 17, 2021, 5:54:41 PM9/17/21
to sara...@googlegroups.com
The Sawbird use a 90deg light pipe which can sometimes be out of alignment. 

Sent from my iPhone

On Sep 17, 2021, at 5:40 PM, Anthony <itpart...@gmail.com> wrote:



Anthony

unread,
Sep 17, 2021, 6:05:22 PM9/17/21
to sara...@googlegroups.com
Good to know. I'll test that particular Sawbird later this weekend.

Ron Coffey

unread,
Sep 17, 2021, 8:37:40 PM9/17/21
to Society of Amateur Radio Astronomers
I noticed the LED would come on sometimes when I was plugging or unplugging the USB power supply but generally would not stay on. . So, I took the cover off the Sawbird and applied 5 V directly to the Vin and ground terminals. Low and behold, the LED came on and stayed on. So my guess is there is something wrong with the power connection. However, I don't understand how the bias tee supplies power and if that has anything to do with the power connection. In any event, I just ordered a new Sawbird--- that seemed like the easiest solution.

Thanks for all the posts and help!!!!

Ron

Marcus D. Leech

unread,
Sep 17, 2021, 8:43:27 PM9/17/21
to sara...@googlegroups.com
On 2021-09-17 8:37 p.m., 'Ron Coffey' via Society of Amateur Radio Astronomers wrote:
I noticed the LED would come on sometimes when I was plugging or unplugging the USB power supply but generally would not stay on. . So, I took the cover off the Sawbird and applied 5 V directly to the Vin and ground terminals. Low and behold, the LED came on and stayed on. So my guess is there is something wrong with the power connection. However, I don't understand how the bias tee supplies power and if that has anything to do with the power connection. In any event, I just ordered a new Sawbird--- that seemed like the easiest solution.

Thanks for all the posts and help!!!!

Ron
The SAWBIRD LNAs are willing to take power from:

  o The output connector--since many RTL-SDR variants (including the NooElec SMartTee variants) will provide +5V up the input coax--using a so-called
     Bias-T in the receiver.  Satellite receivers do this as well, although THEY put out up to +20V(!!!!!!!).

  o The Micro-USB connector next to the RF input

  o The Vin/Gnd terminals on the circuit board.

Check for a DC-short somewhere in your RF wiring--the output of the LNA has an "inverse" bias-T, so it "sees" anything you do with either of the two other
  DC inputs, unless you have a DC-block (capacitor) on the output of the LNA (which would render any DC coming "up" the coax useless, but would isolate the
  rest of your RF chain from the two other DC inputs).  If *I* were designing this, I might put a low-forward-drop diode in place to isolate all the various points
  where DC can be injected.


Pablo Lewin

unread,
Sep 29, 2021, 2:07:03 PM9/29/21
to Society of Amateur Radio Astronomers
I'm having the same problem even when trying to power several LNAs in series... apparently it's a blocking diode problem on either the LNA or the SDRI just ordered a DC block from Nooelec " Certain SDRs, like previous versions of the NESDR SMArt (not including the current production version, v4), may have their ESD protection diode tripped by the LNA when powered externally. You can still use the LNA with these SDRs of course, but you would need to install a DC block between the LNA and SDR. This can be as simple as a low-distortion, high-quality capacitor in series with the RF path, should you prefer a DIY solution. We also sell an affordable DC block in our webstore. Alternatively, you can power the LNA with a bias tee capable SDR like the NESDR SMArTee or NESDR SMArTee XTR."

Anthony

unread,
Sep 29, 2021, 2:30:50 PM9/29/21
to sara...@googlegroups.com
I'm my case I ran multiple Nooelec H+1 LNA and bandpass filters in this way.

Nooelec LNA + DC Block + Nooelec + DC Block + Hydrogen bandpass filter (no power required, but) + DC Block + RTL SDR, but I'm also using, two 10' low loss coaxial extension cables, their not top notch I don't believe, Low-Loss-S-LMR240.

The chained DC Blocks in my configurations, I got the idea from Marcus.

This setup worked for my portable feedhorn, now for the 3-meter dishes, that maybe entirely a different setup.


Anthony

unread,
Sep 29, 2021, 2:33:32 PM9/29/21
to sara...@googlegroups.com
The RTL-SDRs diode would trip on LNAs powered by external sources without a DC Block between the LNA and  TRL-SDR blog v3

On Wed, Sep 29, 2021, 2:07 PM Pablo Lewin <pabl...@gmail.com> wrote:

Hamish Barker

unread,
Sep 29, 2021, 3:30:45 PM9/29/21
to sara...@googlegroups.com
I think the bandpass filter should go as close to the first LNA as possible, then the second LNA can come downstream after the filter.

Marcus D. Leech

unread,
Sep 29, 2021, 3:31:12 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 2:33 p.m., Anthony wrote:
The RTL-SDRs diode would trip on LNAs powered by external sources without a DC Block between the LNA and  TRL-SDR blog v3
A few random comments.

RF bits and pieces aren't automatically prepared for DC currents.  Understanding a bit of electronics, and having schematics available
  really helps you to understand your overall system architecture.

A TVS diode will absolutely trip when you apply more than about 0.3V of forward bias to it, and will likely conduct significant current before
  self-destructing, depending on input voltage.  AGAIN: understanding the schematic and a bit of electronics circuit theory really helps here.

The SawBird LNAs don't have "DC wrap-around", so you cannot "chain" them together sending DC up the coax--the one closest to the bias injector
  (whether that's external or your SDR) will be powered, but anything beyond it will NOT be.   This is in contrast to CATV/SATTV "line amps" that
  have a "DC wrap-around" circuit in them--but NOTE CAREFULLY--these "line amplifiers" required at least 12V on their inputs to work satisfactorily,
  which will instantly make your 5V-powered LNA go "poof".

Also, unless you're driving very long coax runs (50-60m), you should NOT need more than the single SawBird LNA.

At the CCERA observatory, we inject bias on our RF cables of about +15.5V, and then use a bias-T in "inverted" mode to pick off the +15.5V, filter and
  regulate it down to +5V for our SawBird LNAs.   We have a control panel in the observatory trailer that controls DC injection on each of the cables:



This power-control array is powered by an external Hammond 13.8V power supply (custom built for a small group of Hams back in the 1980s
  by Fred Hammond himself), which I've adjusted to 15.5V.     Now that our cable runs are much shorter, I should probably crank it down to
  below 12V, since that will make the regulators run a bit cooler, and we don't have anything at the moment that *NEEDS* that higher
  bias voltage (like satellite LNBFs, etc).

Inside those panels there are a bunch of these, along with the switches, etc:

ASKA PI-10 Power Inserter Up to 60 VDC 1 Amp 2.2 GHz Coaxial
      Inline Power Pass Signal Insertion Loss 0.5dB Max 5-2150 MHz, Part
      # PI10

That's an Aska PI-10.  These are available on Amazon, from SummitSource, eBay, etc.  Holland electronics also makes them, as I believe
  does Channel Master.  They're likely all built in the same factory in China :)

At the LNA end, we have another of these PI-10, but operated so the "TV" port goes to the LNA *output*, the "to AMP" goes to the coax,
  and the DC port goes to our regulator assembly which then powers a micro-USB cable that plugs into the SawBird micro-USB power port.
  Since the "To TV" port is already DC blocked, I don't have to worry about a separate DC-block on the output of the SawBird LNA.

Similar arrangements can also be made for things like the SPF-5189Z modules that are available "everywhere"--they again WILL NOT TOLERATE
  more than +5V in, and even +5V they sometimes don't like, so I sometimes put a couple of series 1N4004 diodes between the regulator
  output and the SPF-5189Z module--this will drop the +5V to around 3.6V or so.  The SawBirds of course are perfectly happy with
  +5V, but NO MORE THAN +5V!!!!




Marcus D. Leech

unread,
Sep 29, 2021, 3:34:32 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 3:30 p.m., Hamish Barker wrote:
I think the bandpass filter should go as close to the first LNA as possible, then the second LNA can come downstream after the filter.
Indeed, the filter should be between the first and 2nd LNA, if you have a 2nd LNA.  The SawBird H1 LNA modules already have a filter
  on the output, so they (if needed at all) "chain" reasonably well.   If would be nice if NooElec sold a "naked" filter module for 21cm
  for use right in front of the receivers.   My habit has been to have a "last chance" filter right in front of the receivers, although to be
  honest, it's not needed most of the time, except in situations where you have long cable runs and high ambient RF fields--those cables
  *will* pick up some noise, and it's nice to band-limit that noise before your receiver can see it.


Anthony

unread,
Sep 29, 2021, 4:16:54 PM9/29/21
to sara...@googlegroups.com
I wondered about that, but thought the bandpass filter would help to reduce the excessive LNA artifacts? 

Anthony

unread,
Sep 29, 2021, 4:19:10 PM9/29/21
to sara...@googlegroups.com
There is quite a bit of RF interface in the lower part of my back-yard. 

I'll have test that out this weekend. 

Anthony

unread,
Sep 29, 2021, 4:24:33 PM9/29/21
to sara...@googlegroups.com
But when using one Sawbird LNA, it seemed the H1 signal was incredibly weak, even using on

Anthony

unread,
Sep 29, 2021, 4:27:05 PM9/29/21
to sara...@googlegroups.com
But when using one Sawbird LNA and the bandpass filter it seemed the H1 signal was incredibly weak, even when reducing the coax to only 10 feet. 


Anthony

unread,
Sep 29, 2021, 4:28:38 PM9/29/21
to sara...@googlegroups.com
Any good electronic books out there?

On Wed, Sep 29, 2021 at 3:31 PM Marcus D. Leech <patchv...@gmail.com> wrote:

Marcus D Leech

unread,
Sep 29, 2021, 4:29:16 PM9/29/21
to sara...@googlegroups.com
What band pass filter?

Some of them (particularly SAW based) can be 
Very lossy. 

Sent from my iPhone

On Sep 29, 2021, at 4:27 PM, Anthony <itpart...@gmail.com> wrote:



Marcus D Leech

unread,
Sep 29, 2021, 4:30:43 PM9/29/21
to sara...@googlegroups.com
The first few chapters of the ARRL handbook. Any edition published in the last 20 years or so should suffice. 

Sent from my iPhone

On Sep 29, 2021, at 4:29 PM, Anthony <itpart...@gmail.com> wrote:



Anthony

unread,
Sep 29, 2021, 4:35:07 PM9/29/21
to sara...@googlegroups.com
I thought that may be the case for either a bad Sawbird or the bandpass filter not working. Here is the 1420 MHz (attachment) band pass filter I'm using. I have a few Sawbirds, 
I may swap out the one I'm currently using. 

20210918_172712.jpg

Anthony

unread,
Sep 29, 2021, 4:40:22 PM9/29/21
to sara...@googlegroups.com

Anthony

unread,
Sep 29, 2021, 4:55:35 PM9/29/21
to sara...@googlegroups.com
Maybe this LNA, not specific for neutral Hydrogen emissions but may work along with the Sawbirds H+1


LNA.jpg

Marcus D. Leech

unread,
Sep 29, 2021, 4:58:45 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 4:34 p.m., Anthony wrote:
I thought that may be the case for either a bad Sawbird or the bandpass filter not working. Here is the 1420 MHz (attachment) band pass filter I'm using. I have a few Sawbirds, 
I may swap out the one I'm currently using.
Those GPIO labs filter modules are actually not bad--only about 2.5dB insertion loss, which makes them entirely-reasonable as an interstitial filter.

Keep firmly in mind that the strongest H1 response you'll see with your ~10" aperture will be in the vicinity of about 0.5-0.6dB above the notional
  noise floor.

The field-of-view-averaged noise temperature during galactic-plane transit is likely somewhere around 30K (perhaps lower).   Your setup will
  "see" half of that (due to random polarization, and a single-polarization reception setup), so about 15K.   Your Tsys is perhaps 100K possibly
  slightly less, possibly slightly more.

So, the ratio between 115K and 100K is not that much--around 0.6dB.  Don't be expecting very large "bumps" due to hydrogen.  You can make it
  stand-out more by doing baseline subtraction, and making sure you have adequate integration time.


Marcus D. Leech

unread,
Sep 29, 2021, 4:59:42 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 4:55 p.m., Anthony wrote:
Maybe this LNA, not specific for neutral Hydrogen emissions but may work along with the Sawbirds H+1
Sure, I have a prototype of one of these and they work adequately.

But, AGAIN.  You shouldn't need one.


Anthony

unread,
Sep 29, 2021, 5:20:00 PM9/29/21
to sara...@googlegroups.com
That makes sense, I'm working on how to perform, well understanding how to do baseline subtractions. Using the 50 Ohm termination caps doesn't do the best job. I'll review my integration time as well and I'm reading on antenna temperature, "This temperature has nothing to do with the physical temperature of the antenna itself but is related to the temperature of distant objects that the antenna is looking at. Looks like my reference book is talking about radiation resistance R, can't make those symbols in email. :-D Is that the temperature you are referring to

Reflection mechanisms, Radiation mechanisms, Radiation resistance, and patterns. HPBW and FNBW, all very interesting along with the math, definitely a challenge doing the example calculations. 

On Wed, Sep 29, 2021 at 4:58 PM Marcus D. Leech <patchv...@gmail.com> wrote:
On 2021-09-29 4:34 p.m., Anthony wrote:
I thought that may be the case for either a bad Sawbird or the bandpass filter not working. Here is the 1420 MHz (attachment) band pass filter I'm using. I have a few Sawbirds, 
I may swap out the one I'm currently using.
Those GPIO labs filter modules are actually not bad--only about 2.5dB insertion loss, which makes them entirely-reasonable as an interstitial filter.

Keep firmly in mind that the strongest H1 response you'll see with your ~10" aperture will be in the vicinity of about 0.5-0.6dB above the notional
  noise floor.

The field-of-view-averaged noise temperature during galactic-plane transit is likely somewhere around 30K (perhaps lower).   Your setup will
  "see" half of that (due to random polarization, and a single-polarization reception setup), so about 15K.   Your Tsys is perhaps 100K possibly
  slightly less, possibly slightly more.

So, the ratio between 115K and 100K is not that much--around 0.6dB.  Don't be expecting very large "bumps" due to hydrogen.  You can make it
  stand-out more by doing baseline subtraction, and making sure you have adequate integration time.


On Wed, Sep 29, 2021 at 4:29 PM Marcus D Leech <patchv...@gmail.com> wrote:
What band pass filter?

Some of them (particularly SAW based) can be 
Very lossy. 

Sent from my iPhone

On Sep 29, 2021, at 4:27 PM, Anthony <itpart...@gmail.com> wrote:


But when using one Sawbird LNA and the bandpass filter it seemed the H1 signal was incredibly weak, even when reducing the coax to only 10 feet. 

Marcus D. Leech

unread,
Sep 29, 2021, 5:27:30 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 5:19 p.m., Anthony wrote:
That makes sense, I'm working on how to perform, well understanding how to do baseline subtractions. Using the 50 Ohm termination caps doesn't do the best job. I'll review my integration time as well and I'm reading on antenna temperature, "This temperature has nothing to do with the physical temperature of the antenna itself but is related to the temperature of distant objects that the antenna is looking at. Looks like my reference book is talking about radiation resistance R, can't make those symbols in email. :-D Is that the temperature you are referring to

Reflection mechanisms, Radiation mechanisms, Radiation resistance, and patterns. HPBW and FNBW, all very interesting along with the math, definitely a challenge doing the example calculations.
When we talk about temperature, we're talking about equivalent blackbody temperature.    A termination resistor at ambient will have a noise temperature of
  whatever the actual temperature is (297K at 25C).  An object in space isn't necessarily a pure "black body", but using the black-body temperature model
  is a useful abstraction for doing comparisons.

An amplifier will have a noise temperature as well, that is expressed either in dB ("Noise Figure") or Noise Temperature (degrees K).  Now, the noise temperature
  of your entire *system* will be *dominated* by the noise temperature of your LNA (See Friis Noise-chain analysis).   But there will be other factors contributing
  to your system temperature (Tsys).  Including things like ambient in-band noise levels as seen by the antenna, loss in front of the LNA, etc, etc.   A good starting
  point to estimate this when doing calculations is about 100K--actual systems will vary somewhat, but an amateur station achieving an overall Tsys below about
  50K is fairly difficult, and a Tsys above 200K usually means, these days, that something is wrong, probably fairly-early in your receive chain.



On Wed, Sep 29, 2021 at 4:58 PM Marcus D. Leech <patchv...@gmail.com> wrote:
On 2021-09-29 4:34 p.m., Anthony wrote:
I thought that may be the case for either a bad Sawbird or the bandpass filter not working. Here is the 1420 MHz (attachment) band pass filter I'm using. I have a few Sawbirds, 
I may swap out the one I'm currently using.
Those GPIO labs filter modules are actually not bad--only about 2.5dB insertion loss, which makes them entirely-reasonable as an interstitial filter.

Keep firmly in mind that the strongest H1 response you'll see with your ~10" aperture will be in the vicinity of about 0.5-0.6dB above the notional
  noise floor.

The field-of-view-averaged noise temperature during galactic-plane transit is likely somewhere around 30K (perhaps lower).   Your setup will
  "see" half of that (due to random polarization, and a single-polarization reception setup), so about 15K.   Your Tsys is perhaps 100K possibly
  slightly less, possibly slightly more.

So, the ratio between 115K and 100K is not that much--around 0.6dB.  Don't be expecting very large "bumps" due to hydrogen.  You can make it
  stand-out more by doing baseline subtraction, and making sure you have adequate integration time.



On Wed, Sep 29, 2021 at 4:29 PM Marcus D Leech <patchv...@gmail.com> wrote:
What band pass filter?

Some of them (particularly SAW based) can be 
Very lossy. 

Sent from my iPhone

On Sep 29, 2021, at 4:27 PM, Anthony <itpart...@gmail.com> wrote:


But when using one Sawbird LNA and the bandpass filter it seemed the H1 signal was incredibly weak, even when reducing the coax to only 10 feet. 
--
--
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.

Anthony

unread,
Sep 29, 2021, 5:29:18 PM9/29/21
to sara...@googlegroups.com

Interesting antenna temperature... attachments. 
20210929_172112.jpg
20210929_172131.jpg
20210929_172220.jpg
20210929_172711.jpg

Anthony

unread,
Sep 29, 2021, 5:30:46 PM9/29/21
to sara...@googlegroups.com
Wouldn't the sun, and moon be considered blackbody radiation or temperature sources? 

Marcus D. Leech

unread,
Sep 29, 2021, 5:33:02 PM9/29/21
to sara...@googlegroups.com
On 2021-09-29 5:30 p.m., Anthony wrote:
Wouldn't the sun, and moon be considered blackbody radiation or temperature sources?
It depends on wavelength...

Lunar radiation is basically a pure black-body phenomenon, since it arises strictly from the heating of the lunar surface by the Sun.

The Sun itself?  More complicated.  A tangled miasma of synchrotron and thermal emissions, the distribution of which depends on wavelength.

Anthony

unread,
Sep 29, 2021, 5:35:20 PM9/29/21
to sara...@googlegroups.com
Let me review Friis Noise-chain analysis, I understand this more!

Thank you! :)

bsn...@gmail.com

unread,
Sep 29, 2021, 7:02:14 PM9/29/21
to Society of Amateur Radio Astronomers
The latest edition of Horowitz & Hill, "The Art of Electronics" or "Practical Electronics for Inventors" by Scherz & Monk.
Either one will keep you occupied for months through a long winter and neither one is aimed at those wishing to become an EE.

Anthony

unread,
Sep 29, 2021, 7:23:47 PM9/29/21
to sara...@googlegroups.com

Anthony

unread,
Sep 29, 2021, 8:12:55 PM9/29/21
to sara...@googlegroups.com
 I like this book of the two choices,

Practical Electronics for Inventors, Fourth Edition 4th Edition

Thanks again, I purchased this book as well. 

Reply all
Reply to author
Forward
0 new messages