Low(er) Noise Amplifier ?

[Alex P]

I tried a few of the GPIO Labs LNAs and the nooelec LaNA ahead of

the nooelec SAWBird H1 ( at the antenna ) to pre-amp the signal .

What I found : the noise was amplified equal to the signal such that

although the AGC in the SDR  did 'less work', the S/N  ratio remained constant.

Is  there some other LNA which can enhance the system's S/N  ?

Thanks,

Alex

[Jeff Kruth]

No.  Whatever the S/N is at the antenna terminals, thats it. The following electronics cannot make it better (unless it is a deterministic signal, then some math tricks like spread spectrum can help). But for noise, thats it. This is from basic comm theory and is based on signal out/noise out developments. The receiver electronics (+LNA) only make it worse.

Regards, Jeff Kruth

--
--
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/1d329683-eac8-4e37-9a7b-dbb65dd36d9fn%40googlegroups.com.

[Paul Oxley]

Jeff

True, but a lousy LNA will add significant noise. The noise added in dB = Noise in LNA / LNA gain.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/2142150778.1313966.1633193501566%40mail.yahoo.com.

[Paul Oxley]

Jeff

The equation in dB is LNA Noise Factor - LNA gain. The equation shown previously would work if the factors were not converted to dB.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/659243624.1303279.1633194220153%40mail.yahoo.com.

[b alex pettit jr]

These are $50 LNAs .. are there amps in the $100s which have significantly lower noise ?

Alex

===================

[Marcus D. Leech]

On 2021-10-02 1:19 p.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

These are $50 LNAs .. are there amps in the $100s which have significantly lower noise ?

Yes, for example:

https://www.radioastronomysupplies.com/store/p61/1420_MHz._HIGH_PEFORMANCE_LNA%2C_NF_0.29dB%2C__GAIN_37dB.html


And HB9BBD can make you a cavity-input LNA:

http://www.hb9bbd.ch/article.php?key=34

But it really depends on how you define "significant".

The "signal" arrives with a certain SNR, and your goal as a system designer is to minimize the extent to which your
  system makes it *worse*.  The H1 "signal" arrives having already been contaminated with the non-H1 broad-spectrum
  continuum noise along the same path, as well as contributions from the atmosphere (perhaps 1-2K), and whatever
  ground-noise (300K diminished by whatever your spillover profile is) is making it into your feed.

To do a comparison between similar systems, you could compare with the antenna work that I did.  I got about 0.6dB of H1 peak
  during transit around Deneb using the para-grid and "stock" dipole with a SawBird H1 LNA.

You have a similar setup, except that you have a loop feed.  If your results are considerably worse than the above, then something
  else is wrong with your system.

I would say that spending lotsa $$$ on an LNA for an antenna setup that is only intended as a "get your feet wet" type of scheme might
  not be a good spend, unless the intention is to move it to a larger antenna later.

Alex

===================

On Saturday, October 2, 2021, 01:04:19 PM EDT, Paul Oxley <oxl...@att.net> wrote:

Jeff

True, but a lousy LNA will add significant noise. The noise added in dB = Noise in LNA / LNA gain.

Paul

--
--
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/1359039226.763211.1633195152958%40mail.yahoo.com.

[Paul Oxley]

Alex via Marcus

Yes there are affordable LNAs that will do the job.

After the signal is received, the integration over multiple samples enhances the desired signal and minimizes the noise. For example if looking for a pulse from a pulsar, the pulse repeats and will be enhanced by the integration.

With poorer LNAs, the integration time would increase to find the pulse.

Paul

https://groups.google.com/d/msgid/sara-list/5c5a7b8b-9b63-35dc-ddb3-f39d5a25a3c0%40gmail.com

.

[Marcus D. Leech]

On 2021-10-02 3:21 p.m., Paul Oxley wrote:

Alex via Marcus

Yes there are affordable LNAs that will do the job.

After the signal is received, the integration over multiple samples enhances the desired signal and minimizes the noise. For example if looking for a pulse from a pulsar, the pulse repeats and will be enhanced by the integration.

Although for that special case, you have to use synchronous integration--so-called "folding".  But the overall statistical methods are the same.


The minimum detectable noise increase (as from a celestial object) is, roughly:


S ~=  Tsys/sqrt(B * Tau)

Where B is the bandwidth and Tau is the integration time.

It's easy to see that there are several dimensions along which one can optimize the system.  But since the effect of "B" and "Tau" go with the *square-root*,
  changes in Tsys (such as improving your LNA's performance) improves things more-quickly.

Now, in the case of hydrogen spectral observations, "B" is inherently "fixed" by the spectral distribution of the object itself, and increasing your input bandwidth
  does nothing to improve things, which means that Tsys and Tau are the only "free variables" in the specific case of spectral observations, such as H1.

[b alex pettit jr]

Hi Marcus,

Is the  "  0.6 dB "  value the relative amplitude above the Background Noise Level ?

As this ??  The actual numeric data in the table is in the - 92 dB range. I added an offset for the plot.

 And  the red_dot  level  equates to Cosmic & Other noise rather than 99% Amplifier noise ?

Thanks for Your Patience .. this is all new.

Regards,

Alex

[Marcus D. Leech]

On 2021-10-02 3:57 p.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

Hi Marcus,

Is the  "  0.6 dB "  value the relative amplitude above the Background Noise Level ?

As this ??  The actual numeric data in the table is in the - 92 dB range. I added an offset for the plot.

 And  the red_dot  level  equates to Cosmic & Other noise rather than 99% Amplifier noise ?

Thanks for Your Patience .. this is all new.

Regards,

Alex

Yes, I was referring to "above the notional noise floor".

The noise floor is a combination of "other cosmic noise" and contributions from the instrumentation and
  things like spillover and side-lobe noise.

What SDR are you using, and what gain settings?  This is with just a single SawBird?

To do a comparison between similar systems, you could compare with the antenna work that I did.  I got about 0.6dB of H1 peak
  during transit around Deneb using the para-grid and "stock" dipole with a SawBird H1 LNA.

--
--
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/722690939.770233.1633204645632%40mail.yahoo.com.

[b alex pettit jr]

Yes one SB. The SMArTee

I just set the gain to RTL= AGC .

That might be "RA 7:00"   +- 1/2 hr. I was more intent on comparing W  &  W/O pre-amp than the particular part of the sky .

I had the mount running in RA to keep the same field of view over the same testing period..

That was one 3 minute average.

Regards,

Alex

===================================================

[b alex pettit jr]

I have the 1.0 m x 0.6 m ParaGrid ..

.

[Anthony]

Hi Alex,

I'm new to this as well, but from my understanding setting the RTL-SDR to  RTL AGC actually degrades reception, even for the Nooelec E4000 tuners (SMArTee) or FC0012/13 Zero IF tuners.

According to the SDR# user guide. Even if you're not using SDR#, you may want to manually configure your gain. 

https://www.rtl-sdr.com/sdrsharp-users-guide/

I normally use manual and set my gain to maximum. As a side note, I had one of those SMArTee's and returned it. The reviews were correct, it was unstable and kept crashing! I either use the AirSpy mini dongle or RTL-SDR v3 blog, but make sure they actually are made by RTL-SDR v3 Blog. I had a few from overseas (no country name-calling) and they too were very unstable and promptly returned them, even after trying USB 3 or USB 2.0 cables, just kept getting excessively hot and freezing up, even with several fans cooling them off. 

Regards,

Anthony 

--
--
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/494387969.782216.1633208990506%40mail.yahoo.com.

[Marcus D. Leech]

On 2021-10-02 5:34 p.m., Anthony wrote:

Hi Alex,

I'm new to this as well, but from my understanding setting the RTL-SDR to  RTL AGC actually degrades reception, even for the Nooelec E4000 tuners (SMArTee) or FC0012/13 Zero IF tuners.

According to the SDR# user guide. Even if you're not using SDR#, you may want to manually configure your gain. 

https://www.rtl-sdr.com/sdrsharp-users-guide/

I normally use manual and set my gain to maximum. As a side note, I had one of those SMArTee's and returned it. The reviews were correct, it was unstable and kept crashing! I either use the AirSpy mini dongle or RTL-SDR v3 blog, but make sure they actually are made by RTL-SDR v3 Blog. I had a few from overseas (no country name-calling) and they too were very unstable and promptly returned them, even after trying USB 3 or USB 2.0 cables, just kept getting excessively hot and freezing up, even with several fans cooling them off. 

Regards,

Anthony 

I use the NeSDR SMART (without the bias-T, using the R820T2 tuner) for many of my experiments.  Never had a problem.  I use 3 of them in my riometer design,
  which just runs and runs and runs.

Some USB2 ports are "marginal" for power, and the RTL-SDRs often drawn close to the limts--then having it supply an LNA can push it over the edge sometimes,
  causing the controller to kick it off the bus.

Conventional wisdom for radio astronomy is to NOT use AGC, although it depends very much on the details of the AGC behavior.  Your mileage may vary, etc, etc.

So, I've gone over this before, but let's review.

Your Tsys will be somewhere in the range of 100K (100-150K, let's say) with these setups.   The beam-averaged Tant (antenna temperature) may be about 20-30K
  when a "good" part of the hydrogen sky occupies some part of the beam, and perhaps 10K when there's "nothing particular" in the beam.

So, the ratio between:

(Tsys + Tsky_interesting ) / (Tsys+Tsky_boring)

=

130 / 110

=

Is about 0.7dB

BUUUUUUUUUUUUUUUUUTTTTTTTTTTTTTTTTTTTTTTT

You're only going to "see" half of those Tsky temperatures, because of polarization loss (source is randomly polarized, instrument is fixed).

So, it then becomes:

115/105

approximately 0.4dB

You result is somewhere in between, at about 0.6dB.   You're doing fairly well.

If we drop Tsys by, let's say, 10K, we get (keeping the other very-rough approximations constant):

105/95

=

0.43dB

The main thing that happens with very-small apertures is that the "thing of interest" only occupies a smallish part of your beam area.  Even though H1 clouds are
  "extended" they generally *will not* fill your entire beam/FOV for these very-small aperture antennas.  Which means your beam is "diluted" with a lot of sky
  that is quite a bit colder (in the radio brightness sense) than the H1 clouds you're trying to measure.  Once your Tsys is "reasonable", at this scale, more
  aperture is what is called for, IMHO.

[b alex pettit jr]

Hello Anthony,

GREAT INFO .. Thank You. I'll try the manual gain option.

I've had no issues with stability , but  I have the units on a fan cooled heat sink ,. ( attached w/ thermal tape )

Regards,

Alex

[Anthony]

That's a nice setup, Alex!!

I like how you have the fans cooling off the SDR & LNA!

--
--
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/1860292094.779653.1633212394824%40mail.yahoo.com.

[b alex pettit jr]

This is a set of  8 ea, 10 minute drift averages  taken at Dec 0 Deg ,. RA  6:30 - 7:30 Hrs. ( or thereabout )

( Vert Axis : dB ref to Background )

An area of the sky was used for calibration . It had some H1 content which created the dip at 1420.5 MHz

.

.

[Anthony]

That's interesting Marcus on the SmarTee, I even used three different laptops and computers to prove it was my system. Guess I'm unlucky with vendors... :) 

--
--
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/4f28236e-54a0-0af2-4bc5-f2c27bc525ea%40gmail.com.

[Anthony]

By the way, this was the Nooelec unit I was using and returned

https://www.amazon.com/dp/B06Y1D7P48?psc=1&ref=ppx_yo2_dt_b_product_details

The screenshot is also attached. 

--
--
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/1860292094.779653.1633212394824%40mail.yahoo.com.

[Marcus D. Leech]

On 2021-10-02 6:18 p.m., Anthony wrote:

By the way, this was the Nooelec unit I was using and returned

https://www.amazon.com/dp/B06Y1D7P48?psc=1&ref=ppx_yo2_dt_b_product_details

I use these:

https://www.amazon.com/dp/B01HA642SW/ref=cm_sw_em_r_mt_dp_470788DGP4R7X35A5F2Q?_encoding=UTF8&psc=1

They have had problems in the past with the USB connectors being slightly wobbly.  The USB parts industry is a festering
  swamp, and you can have entire batches with crappy connectors, etc.

The Odroid XU4 had USB connector issues also--USB3 connectors acting super-flaky until there had been several insert/remove cycles.
  Turns out that factory had wax-coated the connector internals, making them not a very good conductor until the mating connector
  had worn off the wax.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TzsOnHk7mpYcYShSGOOakmU32d1aOcpEL498F3oWTbDGw%40mail.gmail.com.

[b alex pettit jr]

Hi Marcus,

You beat me to this comment : YES, the USB connector is really finicky . I install it and leave it alone .. and seems to work ,..

but barely move the cable and a connection glitch stops the acquisition. At some point in time, I plan to try shimming it up.

Alex

[Marcus D. Leech]

I've had the same issue with AirSpy as well.  Even when the connector is as mechanically sound as its ever going to get, jostling them can
  *frequently* create enough noise on the bus to cause the controller to kick it off the bus.

The entire USB ecosystem was designed by notional "children".  The connectors, the signalling, the protocol state machine, etc, etc.
  BUT, building your "thing" with USB is VASTLY cheaper than building it with (for example 1GiGe), because 1GiGe "stacks" just aren't
  cheap.  Someone needs to address that.  Because then these flakies would just go away.

[Anthony]

Yes, I have that one also but have not used it yet. 

Good to know about the USB connectors on the RTL's dongle! I noticed on the overseas RTL-SDRs they were much thicker and the USBs didn't sit well on any of my laptops or PCs. The manufacture didn't say RTL-SDR Blog but the country of origin instead. 

https://www.amazon.com/dp/B01HA642SW/ref=cm_sw_em_r_mt_dp_470788DGP4R7X35A5F2Q?_encoding=UTF8&psc=1

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/86742525-0689-163b-8254-809c81abe46d%40gmail.com.

[Marcus D. Leech]

On 2021-10-02 6:43 p.m., Anthony wrote:

Yes, I have that one also but have not used it yet. 

Good to know about the USB connectors on the RTL's dongle! I noticed on the overseas RTL-SDRs they were much thicker and the USBs didn't sit well on any of my laptops or PCs. The manufacture didn't say RTL-SDR Blog but the country of origin instead. 

https://www.amazon.com/dp/B01HA642SW/ref=cm_sw_em_r_mt_dp_470788DGP4R7X35A5F2Q?_encoding=UTF8&psc=1

I have a *whack* of RTL-SDRs of the cheap-plastic-case variety--I started hoarding them several years ago, when they were often available for about C$10.00 apiece.
  They work fine and have similar USB connector problems to everyone else.  They have poor thermal management and crappy crystals.  But for a lot of things,
  that doesn't matter that much.  One of the reasons the metal-cased RTL-SDRs get so warm is that the manufacturers/designers have ACTUALLY PAID ATTENTION
  to thermal issues, and arranged for that heat to be conducted into the metal case.  The plastic ones?  Not so much.  The R820T2, in particular, since it's CMOS,
  and operating internally with 4GHz VFOs, tends to dissipate a lot more heat than you'd expect, Soooooo, NooElec and the RTL Blog folks figured out a while
  ago to have the heat conducted into the metal case.  Which means the case gets hot.  The plastic ones?  You don't notice nearly as much because the
  heat is staying inside, making the thermal pad under the R820T2 a nano frying pan... :)

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TwNQE_f1jLPs49QYXd%2BV%3DDAjf7wa2Qppxm1tjLLBRWFZA%40mail.gmail.com.

[bsn...@gmail.com]

I pass along for whatever it's worth....

I've had good luck using a D-link 4-port hub (DUB-1340)...SDR and everything downstream to the antenna feeds the hub which in turn feeds an older HP Laptop.  

If nothing else, the SDR sits flat (even the NESDR) and the mechanical stability seems to help.....YMMV

I just acquired another LNA from GPIO Labs but haven't had a chance to use it yet. Normally use a SAWBird but my first at-the-antenna amps were 2 SPF 5189Z's following Wolfgang's example in the SARA journal a few issues back. The only problems I've had with either the SAWBird or the 5189Zs is rain!

...bill....

[Anthony]

I'm trying to solve the issue of rain with my LNAs instead of using plastic sandwich bags. I may try facet covers (triangle style), two inverted, and cut find holes for the LNA to thread to the Chassis N Femail jack 4 hole panel mount.

Don't know if that will work, haven't tried it yet. Thinking of going to Home Depot to look around. I have a few small plastic containers that are lightweight...

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/019c69a2-4855-4e86-bced-4b7c27a73686n%40googlegroups.com.

[b alex pettit jr]

A few questions ..

1) From Marcus' calculations, this  " 0.6 dB "  data with my current setup is reasonably Ok ?

2) The noise floor is mostly Cosmic and Earth background rather than LNA Noise ?

3) Then, What might this hardware provide over the current setup ?

https://www.radioastronomysupplies.com/store/p61/1420_MHz._HIGH_PEFORMANCE_LNA%2C_NF_0.29dB%2C__GAIN_37dB.html

Thanks,

Alex

This is a set of  8 ea, 10 minute drift averages  taken at Dec 0 Deg ,. RA  6:30 - 7:30 Hrs. ( or thereabout )

( Vert Axis : dB ref to Background )

.

[Marcus D. Leech]

On 2021-10-02 8:17 p.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

A few questions ..

1) From Marcus' calculations, this  " 0.6 dB "  data with my current setup is reasonably Ok ?

Yup, quite reasonable for your small antenna.

A larger antenna, where the H1 cloud fills the aperture can produce up to 2dB above the
  noise floor.

2) The noise floor is mostly Cosmic and Earth background rather than LNA Noise ?

Actually, from the equations I provided, a goodly portion of the total noise at any given time
  *IS* from the system itself.   If your Tsys is 100K (typical for an amateur effort), then the
  sky is only perhaps 15-50K, depending on aperture and where you look.

3) Then, What might this hardware provide over the current setup ?

https://www.radioastronomysupplies.com/store/p61/1420_MHz._HIGH_PEFORMANCE_LNA%2C_NF_0.29dB%2C__GAIN_37dB.html

The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB
  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the
  Tant from before -- 15K for "interesting" and 5K for "boring":

(15+70)/(5+70)

=

About 0.5dB



A little bit better than with the SawBird H1.  Whether that's "worth it" for a small antenna?
  That's up to you and what your budget will bear.

Thanks,

Alex

This is a set of  8 ea, 10 minute drift averages  taken at Dec 0 Deg ,. RA  6:30 - 7:30 Hrs. ( or thereabout )

( Vert Axis : dB ref to Background )

.

--
--
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/1802066057.799123.1633220246621%40mail.yahoo.com.

[Marcus D. Leech]

On 2021-10-02 7:39 p.m., Anthony wrote:

I'm trying to solve the issue of rain with my LNAs instead of using plastic sandwich bags. I may try facet covers (triangle style), two inverted, and cut find holes for the LNA to thread to the Chassis N Femail jack 4 hole panel mount.

Don't know if that will work, haven't tried it yet. Thinking of going to Home Depot to look around. I have a few small plastic containers that are lightweight...

That's what I tend to do.  The type "N" is an N-to-SMA bulkhead connector--in this case with the 4-hole flange but other styles are available.  That goes via a
  SMA-M---SMA--M connector to the input side of the SawBird H1+.  Then there's a 15cm sma jumper cable, and I use type-F on the output, with a type-F to
  SMA adapter.  The N connector base is liberally goobered with marine silicone.

The type-F end is the notional "bottom" so has a few small holes drilled in it for equalization, and I mount it so that the holes are pointing to the ground(ish).

[b alex pettit jr]

POWER    or    VOLTAGE ??

For Rx, would the amplitude comparisons not be 20 Log ?

( If you note from the plot, "20" was my Y axis scaling factor )

A Power comparison drops my dB values in half ..

Alex

==========================================

From Marcus's Previous Posts

 If your Tsys is 100K (typical for an amateur effort), then the
  sky is only perhaps 15-50K, depending on aperture and where you look.
 

The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB
  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the
  Tant from before -- 15K for "interesting" and 5K for "boring":

(70 + 15 /(70 + 5 ) = 115 / 105
= about 0.5 dB     ( Power 10*Log(x) )

vs ( the Nooelec LNA )

( 100 + 15 ) / ( 100 + 5 )  = 115/105
= 0.4 dB      ( Power 10* Log(x) )

So, I've gone over this before, but let's review.

Your Tsys will be somewhere in the range of 100K (100-150K, let's say) with these setups.   The beam-averaged Tant (antenna temperature) may be about 20-30K    when a "good" part of the hydrogen sky occupies some part of the beam, and perhaps 10K when there's "nothing particular" in the beam.

So, the ratio between:

(Tsys + Tsky_interesting ) / (Tsys+Tsky_boring)

=  130 / 110

=  Is about 0.7dB  Power,   1.4 dB Voltage  ?

BUUUUTT

You're only going to "see" half of those Tsky temperatures, because of polarization loss (source is randomly polarized, instrument is fixed).

So, it then becomes:  115/105  approximately 0.4dB

You result is somewhere in between, at about 0.6dB.   You're doing fairly well.

If we drop Tsys by, let's say, 10K, we get (keeping the other very-rough approximations constant):

105/95  = 0.43dB

The main thing that happens with very-small apertures is that the "thing of interest" only occupies a smallish part of your beam area.  Even though H1 clouds are   "extended" they generally *will not* fill your entire beam/FOV for these very-small aperture antennas.  Which means your beam is "diluted" with a lot of sky   that is quite a bit colder (in the radio brightness sense) than the H1 clouds you're trying to measure.  Once your Tsys is "reasonable", at this scale, more   aperture is what is called for, IMHO.

-

[b alex pettit jr]

Calc's Re Posted ( bad copy/paste )

.The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB

  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the

  Tant from before -- 15K for "interesting" and 5K for "boring":

(70 + 15 ) / (70 + 5 ) = 85  / 75
= about 0.5 dB

[Anthony]

Good stuff to know Marcu, thanks for that information!

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/fc9d31ba-7710-2c58-15c3-704de4b2962f%40gmail.com.

[Anthony]

Where did you get that tube from, looks good! How heavy is it, I need something to protect the LNA when it's mounted to the waveguide element and chassis connector. 

--
--
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/5c2cf1e0-7cfa-d1e3-e6d3-9ded5334cec3%40gmail.com.

[Marcus D. Leech]

On 2021-10-03 10:00 a.m., Anthony wrote:

Where did you get that tube from, looks good! How heavy is it, I need something to protect the LNA when it's mounted to the waveguide element and chassis connector.

It's just hardware store 1-1/2" ABS drain pipe.  I think in many parts of the USA, the plumbing code specifies PVC rather than ABS, but you can get the same parts in
  PVC.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TwvoU2Ad%2BQOvdb97%2BYd9QP5NtjO%2BVrDVsFzmZiOps0-_g%40mail.gmail.com.

[Marcus D. Leech]

On 2021-10-03 4:59 a.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

POWER    or    VOLTAGE ??

For Rx, would the amplitude comparisons not be 20 Log ?

( If you note from the plot, "20" was my Y axis scaling factor )

A Power comparison drops my dB values in half ..

Alex

==========================================

From Marcus's Previous Posts

 If your Tsys is 100K (typical for an amateur effort), then the
  sky is only perhaps 15-50K, depending on aperture and where you look.
 

The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB
  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the
  Tant from before -- 15K for "interesting" and 5K for "boring":

(70 + 15 /(70 + 5 ) = 115 / 105
= about 0.5 dB     ( Power 10*Log(x) )

vs ( the Nooelec LNA )

( 100 + 15 ) / ( 100 + 5 )  = 115/105
= 0.4 dB      ( Power 10* Log(x) )

You use 20Log conversion when you're converting *voltage magnitudes* to *power*.  Otherwise, when you're comparing *power* or "power ratios*,
  it's 10Log.

Noise temperatures represent a noise *power*, and the output of an FFT in most apps is converted to *power* using
  a complex-to-mag**2.

-

--
--
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/859230100.686654.1633251566015%40mail.yahoo.com.

[Anthony]

Excellent, don't know why that didn't dawn on me, I'm heading to Home Depot after my coffee. :) 

The other piece I need to consider may start another thread, is do I go as far as motorizing the two actuators on either dish, one is vertical and the other is horizontal. I would have preferred to have Azimuth and Horizontal or Declination I believe for tracking. I'm not sure if I can use a PR1200 actuator mover for useful RA observations. Job, has a pretty nice setup with what looks like Azimuth and Dec motors for controlling movement. 

What I'm trying to figure out is, can actuators with only vertical and horizontal movement be used. Can I maybe attach a small Polar scope to use for plate solving to determine my position in the sky? Am making sense of what I'm trying to accomplish?

I may want to do more than a drift scan and I have decided to move one dish west, with a baseline between both dishes of 30 feet for a future interferometer. 

Thanks, Marcus!

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/91b5c9f2-d934-baa8-0abf-ee9856497c6e%40gmail.com.

[Marcus D. Leech]

On 2021-10-03 10:32 a.m., Anthony wrote:

Excellent, don't know why that didn't dawn on me, I'm heading to Home Depot after my coffee. :) 

The other piece I need to consider may start another thread, is do I go as far as motorizing the two actuators on either dish, one is vertical and the other is horizontal. I would have preferred to have Azimuth and Horizontal or Declination I believe for tracking. I'm not sure if I can use a PR1200 actuator mover for useful RA observations. Job, has a pretty nice setup with what looks like Azimuth and Dec motors for controlling movement. 

What I'm trying to figure out is, can actuators with only vertical and horizontal movement be used. Can I maybe attach a small Polar scope to use for plate solving to determine my position in the sky? Am making sense of what I'm trying to accomplish?

I may want to do more than a drift scan and I have decided to move one dish west, with a baseline between both dishes of 30 feet for a future interferometer. 

Thanks, Marcus!

You can achieve great things with only a single-axis of motion--the elevation axis, and let the earth rotate you through RA.  This doesn't work for very-weak
  objects that you have to track to get decent real-time integration.   But one can always use day-over-day RA-aligned "stacking" to improve SNR.

Elevation angle is just:

90 + local_latitude - declination

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65Ty22A7DKcmJN7cter5ui-D-CTwkLBEsfpMnkp1s9v_HCg%40mail.gmail.com.

[Anthony]

Awesome, thank you for that knowledge!!

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/ab530fd8-f035-1803-e04e-c9e6fab5d5ff%40gmail.com.

[b alex pettit jr]

Hi Marcus,

Thanks.

In doing some reading,  Noise is expressed in Power ..

SO, that puts my data more in line with what may be expected with this configuration

in this region of the Milky Way

Lot to learn.

Cheers,

Alex

==========================================

[Marcus D. Leech]

On 2021-10-03 10:44 a.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

Hi Marcus,

Thanks.

In doing some reading,  Noise is expressed in Power ..

SO, that puts my data more in line with what may be expected with this configuration

in this region of the Milky Way

What software are you using to produce the data?

Most software that logs spectra already does the conversion into power (via squaring) and then scaling into dB.

[Marcus D. Leech]

On 2021-10-03 10:44 a.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

Hi Marcus,

Thanks.

In doing some reading,  Noise is expressed in Power ..

Annoyingly, some devices specify noise *voltage* (such as ADCs).  Which you then have to
  convert to power, but only after understanding the characteristic impedance of the device.
  Grumble.

[b alex pettit jr]

The freq/ampl pairs from the IF_Avg file

1419.21    0.000024333
1419.212344    0.000024324
1419.214688    0.000024289
1419.217031    0.00002427
1419.219375    0.000024282
1419.221719    0.000024275
1419.224063    0.000024338
1419.226406    0.000024271
1419.22875    0.000024305
1419.231094    0.000024275
1419.233438    0.000024315
1419.235781    0.00002426
1419.238125    0.00002431
1419.240469    0.000024293
1419.242813    0.000024306
1419.245156    0.00002

[Anthony]

FYI

I was curious on the pricing and weight of the Cavity LNA (link Marcus provided) and protecting them from the weather. 

Below is the response from the vendor. 

From: Anthony Fuller <alfu...@itsecuritypartners.com>
Sent: Sunday, October 3, 2021 11:29 AM
To: dfaessler <dfae...@bluewin.ch>
Subject: Re: 1420 MHz Cavity LNA for sale

 

Thank you so much Dominique for responding.

Your LNA's look is fabulous and I'm sure they perform as well as they look!  

I was provided the information from Marcus Leech of the Society of Amature Radio Astronomers for high-end LNAs.

My only fear is, I won't be able to protect your LNAs from humidity. I'm located in Roswell, GA and it's extremely humid here! 

The Cavity LNAs are expensive and I would just die if they become damaged from humidity. I can protect them from the rain, but humidity is a challenge here in the USA, Atlanta, GA. 

I'll keep your Cavity LNA in mind if I can ever build a containment that can protect them from the humidity! 

Sincerely,

Anthony  Fuller 

Anthony Fuller

IT Security Partners, LLC

 

---

From: dfaessler <dfae...@bluewin.ch>
Sent: Sunday, October 3, 2021 11:07 AM
To: Anthony Fuller <alfu...@itsecuritypartners.com>
Subject: RE: 1420 MHz Cavity LNA for sale

 

Good morning Anthony

It seems to me that your system is  RECEIVE-only.

Therefore you do not face any sequencing issues.

Your question re protection seems to address environmental issues.

You need to have an enclosure, ideally made in aluminum or similar. Water is to be kept outside and cabling to be fed watertight into enclosure.

The LNA shall be AS CLOSE TO THE ANTENNA as mechanically possible. No need for isolation by a relay.

The feedhorn ideally is covered by a low-loss foil which prevents damage by humidity.

The weight of the LNA is approx.250 grams.

Please feel free to clear any further questions if there are.

Thank you for your inquiry and kind regards

Dominique Fässler

Von meinem/meiner Galaxy gesendet

-------- Ursprüngliche Nachricht --------

Von: Anthony Fuller <alfu...@itsecuritypartners.com>

Datum: 02.10.21 22:44 (GMT+01:00)

An: dfae...@bluewin.ch

Betreff: 1420 MHz Cavity LNA for sale

Hello!

My name is Anthony Fuller, and I'm interested in purchasing your 1420 MHz Cavity LNA for radio astronomy observations at 21 cm for neutral Hydrogen emissions.

http://www.hb9bbd.ch/article.php?key=26

What is the weight of the unit? What recommendations do you have for protecting this device, attached to a 3-meter dish?

Sincerely,

Anthony Fuller

Roswell, GA 

On Sat, Oct 2, 2021 at 3:10 PM Marcus D. Leech <patchv...@gmail.com> wrote:

On 2021-10-02 1:19 p.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

These are $50 LNAs .. are there amps in the $100s which have significantly lower noise ?

Yes, for example:

https://www.radioastronomysupplies.com/store/p61/1420_MHz._HIGH_PEFORMANCE_LNA%2C_NF_0.29dB%2C__GAIN_37dB.html


And HB9BBD can make you a cavity-input LNA:

http://www.hb9bbd.ch/article.php?key=34

But it really depends on how you define "significant".

The "signal" arrives with a certain SNR, and your goal as a system designer is to minimize the extent to which your
  system makes it *worse*.  The H1 "signal" arrives having already been contaminated with the non-H1 broad-spectrum
  continuum noise along the same path, as well as contributions from the atmosphere (perhaps 1-2K), and whatever
  ground-noise (300K diminished by whatever your spillover profile is) is making it into your feed.

To do a comparison between similar systems, you could compare with the antenna work that I did.  I got about 0.6dB of H1 peak
  during transit around Deneb using the para-grid and "stock" dipole with a SawBird H1 LNA.

You have a similar setup, except that you have a loop feed.  If your results are considerably worse than the above, then something
  else is wrong with your system.

I would say that spending lotsa $$$ on an LNA for an antenna setup that is only intended as a "get your feet wet" type of scheme might
  not be a good spend, unless the intention is to move it to a larger antenna later.

Alex

===================

On Saturday, October 2, 2021, 01:04:19 PM EDT, Paul Oxley <oxl...@att.net> wrote:

Jeff

True, but a lousy LNA will add significant noise. The noise added in dB = Noise in LNA / LNA gain.

Paul

--
--
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/1359039226.763211.1633195152958%40mail.yahoo.com.

--
--
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/5c5a7b8b-9b63-35dc-ddb3-f39d5a25a3c0%40gmail.com.

[Anthony]

Hi Alex,

I'm curious too, what software is that, SDR# with the IF plugin or RTLSDR Scanner?

Thanks! 

--
--
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/1092656074.844389.1633275463674%40mail.yahoo.com.

[b alex pettit jr]

AirSpy SDR# Studio w/ the IF_Avg plugin creating the data files.

[Paul Oxley]

Anthony via Marcus

I have done some work using Pelco Pan and Tilt units for tracking. The units are designed for use in Security cameras. They have motors for moving both azimuth and elevation. They come in various models for different load sizes. It is important to calculate the necessary torque and match it to the unit. They typically specify the torque by specifying a weight that can be handled at 2 inches above the mounting plate.

They can be obtained on e-bay but are no longer in manufacturing. The small web based cameras have made them obsolete.

Paul

https://groups.google.com/d/msgid/sara-list/ab530fd8-f035-1803-e04e-c9e6fab5d5ff%40gmail.com

.

[Larry Mayfield]

Ebay is your friend if looking for one. Use this search:   pelco pan and tilt unit and there are quite a few. Not really inexpensive but not a bank breaker either.

 

Link:   pelco pan and tilt unit | eBay

Control   Pelco CX9115RX Pan/Tilt Zoom Control Panel 115AC With Control Cable. | eBay

Titl/Pan unit  Pelco PT1250P/PP Heavy Duty Pan / Tilt outdoor camera mount. 120VAC. 140W | eBay

 

Paul, thanks for the thought. Tilt  and pan same as alt and azimuth…

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1867264720.1476456.1633276090956%40mail.yahoo.com.

[Anthony]

Thanks Paul, Larry, 

I'm not sure how would I mount this onto one of the dishes? Looks like I would have to make some major modification? 

Imaged attached, one of the 3-meter dish mounts. 

Ebay is your friend if looking for one. Use this search:   pelco pan and tilt unit and there are quite a few. Not really inexpensive but not a bank breaker either.

 

Link:   pelco pan and tilt unit | eBay

Control   Pelco CX9115RX Pan/Tilt Zoom Control Panel 115AC With Control Cable. | eBay

Titl/Pan unit  Pelco PT1250P/PP Heavy Duty Pan / Tilt outdoor camera mount. 120VAC. 140W | eBay

 

Paul, thanks for the thought. Tilt  and pan same as alt and azimuth…

 

Larry

Pahrump, NV

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1867264720.1476456.1633276090956%40mail.yahoo.com.

[Marcus D Leech]

I would stick with the Linear actuator you already have and drive your dishes in elevation only. 

It’s unlikely one of those camera platforms would be even remotely capable of Driving your 3M dish. 

Sent from my iPhone

On Oct 3, 2021, at 4:59 PM, Anthony <itpart...@gmail.com> wrote:



<IMG_1183.jpg>

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TxrREvR6zeiO9BO700mkPcV-bvQMCPpcrjU5vas3Qkg%3Dw%40mail.gmail.com.

<RAS_telescope_pole_v3.jpg>

[Paul Oxley]

Marcus

The software to drive the mount is very important. It needs to be able to make small changes during tracking. For the P/T units, I do this by issuing a short pulse (50 mS) on the motors when needed.

The same would be true with actuators. 

Paul

https://groups.google.com/d/msgid/sara-list/C6C6D7FE-58E5-4737-BF2B-F6FFE9CF1EB8%40gmail.com

.

[Marcus D Leech]

Yup. True of any astronomical motion control system.  

With a transit-type instrument, there is no tracking. Only slewing. 

How big an antenna has anyone actually driven with these Pelco camera units? The main issue is the wind loading. Even if you counter-balance the dish, the fairly small (possibly plastic) gears on those things can easily get turned into cheeseburger. 

Sent from my iPhone

On Oct 3, 2021, at 5:16 PM, Paul Oxley <oxl...@att.net> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/637158436.1540101.1633295797374%40mail.yahoo.com.

[Anthony]

I was thinking the same thing, the torque or weight is only 100 lbs on that camera mount and it's used. I don't care to purchase used electronics any longer. 

I have a 12" SuperJack for elevation and may just remove the horizontal 18" actuator and use it for elevation. As you said let the Earth provide RA. 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/C6C6D7FE-58E5-4737-BF2B-F6FFE9CF1EB8%40gmail.com.

[Anthony]

I live up on a steep hill and the dishes are even at a higher elevation and there is lots of wind coming through there, especially in winter, when the leaves on the trees are no longer there to block the winds. 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/3967B0EF-56D9-4D3D-B84C-8832AF7450E9%40gmail.com.

[Marcus D Leech]

Standard SaTV mounts need to be lightly modified to be driven in elevation. Normally the actuator drives them along an arc aimed at the Clarke Belt (where all the geosats live). 

The actuator we use on our mapper dish is fairly long in order to give us about 110deg of elevation range while at the same time loading the actuator quite lightly. 

Sent from my iPhone

On Oct 3, 2021, at 5:36 PM, Anthony <itpart...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TwWeb3aS3O%3D-dnaQKCoExEUpKgBcirM%2Bez3y-qPLMrQUw%40mail.gmail.com.

[Anthony]

Wouldn't the 18" actuator suffice, instead of the 12" SuperJack?

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/70FC3F8E-B3F2-4889-9EE7-C8CAD858399C%40gmail.com.

[Marcus D Leech]

You’ll have to play with geometries and see how far you can get. Closer to the center of the mount ring gives you more motion per inch of linear travel, but increases the load. 

Sent from my iPhone

On Oct 3, 2021, at 5:48 PM, Anthony <itpart...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65Tyid4Cz105AkPuDHDpY--JRBPa5kQ6Twa%2B3hSLt6GKU3Q%40mail.gmail.com.

[Anthony]

I've requested a quote for the PR1200 actuator movers, and two 16 gauge wires for carrying voltage, and two 22 gauge wires for the sensor to count the pulses from the actuator. The PR1200s come with remote controls and are capable of moving either both dish's horizontal or vertical actuators. If I don't wish to do that, I'll need to cover one of the PR1200's IR sensors. 

I'm still putting this all together. Hopefully within 4 months or I can have the 10x12 shed built and convert it to an observation station. 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/BD3AB98B-CFC7-4BF7-B1FD-79B559107076%40gmail.com.

[Paul Oxley]

Marcus

They come in different sizes. Tomorrow, when the light is better, I will go down stairs and read the tag on the largest one I have. 

The medium duty PT680-24P can handle 50 pounds (22.7 kg). It operates on 24 VAC. It is for both indoor and outdoor operation.

Paul

https://groups.google.com/d/msgid/sara-list/3967B0EF-56D9-4D3D-B84C-8832AF7450E9%40gmail.com

.

[Marcus D. Leech]

On 2021-10-03 7:00 p.m., Paul Oxley wrote:

Marcus

They come in different sizes. Tomorrow, when the light is better, I will go down stairs and read the tag on the largest one I have. 

The medium duty PT680-24P can handle 50 pounds (22.7 kg). It operates on 24 VAC. It is for both indoor and outdoor operation.

Paul

For some perspective.  The wind pressure at 70MPH  is 12.5lb/ft**2.  That assumes the dish is perfectly normal to the wind, and you're going to
  engineering for peak local wind speed.

A 10ft dish has an area of about 78 ft**2, which means a peak force of about 975lbs in peak expected wind.   Now, peak wind doesn't happen
  that often, and the dish will only rarely be pointed so that it's perfectly-normal to the wind.

[Paul Oxley]

Marcus

When very high winds are expected, the dish should be in the stow position (bird bath).  That is what is done at most if not all professional dishes (i.e Green Bank, Pari, etc.) 

I also didn't mention that the Pan/Tilt units apply a brake when the motors are not operating. I don't know if the actuators do this. Do you? It would have to be in both directions of motion on each axis.

Paul

--
--
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/feb3c3ee-4af7-8519-e629-7c1b11ad69f9%40gmail.com

.

[Larry Mayfield]

I like this!  Would a stepper motor pulsed DC voltage-out driver module work being driven for a microcontroller at the correct equation rate (i.e. the rate for either the RA axis and the rate for the elevation or declination rates).  Seems possible…

 

Thanks!

 

Larry

Pahrump, NV

 

From: sara...@googlegroups.com <sara...@googlegroups.com> On Behalf Of Paul Oxley
Sent: Sunday, October 3, 2021 2:17 PM
To: sara...@googlegroups.com
Subject: Re: [SARA] Low(er) Noise Amplifier ?

 

Marcus

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/637158436.1540101.1633295797374%40mail.yahoo.com.

[Paul Oxley]

Larry

On the control system that I built several years ago, it only issued one pulse when the position monitor vs target position indicated. It compared the stored value of degrees/pulse vs the target degrees on each axis. This allowed the speed to be minimal while still reacting for tracking.

I need to update the software/firmware, but have not been able to do this. The firmware was written in compiled C with the software in C++ to match ASCOM. The compiled C operated on a microprocessor (PIC). The C++ on Windows which was also operating either ASCOM or Radio-Eyes. The tracking values were updated based on universal time for the location involved.

Paul

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/003001d7b8b5%2454455420%24fccffc60%24%40mayfco.com.

[Marcus D. Leech]

On 2021-10-03 8:15 p.m., Paul Oxley wrote:

Marcus

When very high winds are expected, the dish should be in the stow position (bird bath).  That is what is done at most if not all professional dishes (i.e Green Bank, Pari, etc.) 

I also didn't mention that the Pan/Tilt units apply a brake when the motors are not operating. I don't know if the actuators do this. Do you? It would have to be in both directions of motion on each axis.

Paul

Sure, you try to "avoid the badness" but engineering a system that will only survive relatively-mild conditions is asking for trouble.

Even at only 30MPH, a 10ft dish wind loading will exceed the stated 100LB capacity.

Back in the 1980s, I was involved in an interferometer project involving two 12ft dishes.  We'd engineered a dual-axis mount based on what we
  thought were "adequately beefy" design margins.  Got the first dish assembled and on the mount.  Gust of wind came along.  Turned one of the
  two slewing bearings into scrap metal.  Wasn't even a particularly big gust, and I guarantee those slewing bearings were much beefier than anything
  in these camera pan-tilt mounts.

Linear actuators have an inherent brake in a way.  They're a little like a worm drive in that regard.  They turn rotary motion of the jack-screw into linear motion
  of the push rod.  Pushing on the push-rod cannot cause much in the way of rotational motion of the jack-screw.

[Paul Oxley]

Marcus

Since you are concerned about the gear chain in the P/T units, do you know the capability of the jack screw gears?

As previously stated the best high wind reaction is to stow the antenna with it pointed at zenith. This gives the minimum wind load since the wind is parallel to the earth. One could also tie down the dish to prevent movement.

In my case, all of the systems are portable. I would bring the antenna inside or in the back of my 4 door Jeep. If the wind is strong enough to tip over the Jeep, disaster happens.

Paul

--
--
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/b9784cc3-47cd-bfa9-958e-d7ebd65cd3b8%40gmail.com

.

[Marcus D. Leech]

On 2021-10-03 10:33 p.m., Paul Oxley wrote:

Marcus

Since you are concerned about the gear chain in the P/T units, do you know the capability of the jack screw gears?

The drive gears in a jack-screw unit are not in the significant load path--that's one of the beauties of a linear-actuator/jack-screw type arrangement.  If they
  fail, it's because the jack-screw has become seized.  The main load-bearing "thing" in a linear actuator is the follower nut.  The type of actuator we use
  is rated for 1200lbs--the follower nut on ours appears to be some type of hardened steel (had it apart for maint this past summer).  Some of the cheaper
  linear actuators use a nylon follower nut, and they tend to fail after a few years.  The very *best* linear actuators use a ball-screw follower nut, which
  can take some insane loads, but also provides extremely smooth motion.

The 18M dishes that D.S. Kennedy used to make use a very-large motorized jack-screw for the elevation axis, and a fairly-conventional pinion-and-ring gear
  drive for the azimuth.   At least one of the larger dishes at Green Bank also uses a jack-screw for the elevation axis, but I can't remember which one
  at the moment.

It's entirely possible that one of those P/T heads will work for a larger dish in benign conditions.  But if they're rated for a 100Lb load right at the mount plate,
  they're not going to do well when a big part of the load is on the end of a lever arm a considerable distance from that mount plate.  A 10ft dish will have a
  feed sitting about 4ft from the vertex, where, presumably, the dish would be mounted to the P/T mount plate.  You can obviously counter-weight it so the
  net load is much less, or even zero if you do it right.  But the dynamic loads can still be dicey.

I'd like to see a tear-down of these P/T heads.  They clearly were NOT designed for this task, but a different, more-benign one.  Whereas a 1200LB linear
  actuator, designed to move large consumer satellite dishes, clearly IS.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1946980006.1597166.1633314830314%40mail.yahoo.com.

[Paul Oxley]

Marcus

Your assumptions are incorrect!

Any wind load is not on the gear chain. Rather it is on the brake that automatically operates when not in motion (powered down and stopped).

Collecting data during a severe storm is not a good idea. Not only is the wind load large, the lightening could destroy your receiver and other items.

My assumption is that there is ALWAYS an operator when data collection is occurring. In all other times, the dish should be stowed and secured.

Paul

https://groups.google.com/d/msgid/sara-list/bcd62a53-7f07-a74a-d7b9-20dce19ddb3d%40gmail.com

.

[Marcus D. Leech]

On 2021-10-03 11:21 p.m., Paul Oxley wrote:

Marcus

Your assumptions are incorrect!

Any wind load is not on the gear chain. Rather it is on the brake that automatically operates when not in motion (powered down and stopped).

According to the drawings here:

https://cctvvideo.com/wp-content/uploads/2019/11/PT1280PPP-Manual.pdf

Load is delivered to the drive gear and worm regardless of the state of the brake module that's on the front of each motor, unless I'm
  wildly misinterpreting the drawings.  When we saw a wind failure all those years ago, it was precisely with one of these worm-gear
  drives, but *considerably* heavier duty than what is in the PT1280 shown above.

The only units I could actually find currently on eBay are the much-lower-rated (20LB) PT175 units.  Very very similar internal structure, but
  smaller, and rated for only 20Lbs.

Collecting data during a severe storm is not a good idea. Not only is the wind load large, the lightening could destroy your receiver and other items.

A storm isn't even necessary to deliver higher-than-rated load even to the "heavy duty" unit mentioned above, like I said, 30MPH, and you're
  at nearly twice the rated load.

My assumption is that there is ALWAYS an operator when data collection is occurring. In all other times, the dish should be stowed and secured.

We run automated observing systems for things like sky-mapping.  We could never have that as an operational requirement.  Our dish survived
  storms on the roof of our previous building no problem, including gusts to over 90km/h.  Observation quality was noticeably poorer, and we simply
  discarded that data.  But the thing survived.  Because we built the mount to survive, using one of the previously-mentioned linear actuators with a
  1200Lb rating.  Had we also decided to have azimuth drive, it would be similarly robust.   I considered one of those Pelco units years ago (when they
  were a bit more plentiful than today), and rejected them because it was obvious they wouldn't survive a use for which they clearly were not designed.
  I saw a listing for a brand-new Pelco PT1280 unit (the heavy duty one), and it was over $3K USD.  For that kind of money, you could easily put together
  a custom industrial-slewing-bearing+linear-actuator mount and still have money left over, and have much better *confidence* in it.

For small-portable-dish with a temporary setup, as you mentioned, these would be great--if you didn't have to pay too much for them.

Paul

[Anthony]

Can this site be used for calculating antenna noise or the power of the antenna noise?

https://www.satellite-calculations.com/Satellite/gtsys.htm

On Sat, Oct 2, 2021 at 9:27 PM Marcus D. Leech <patchv...@gmail.com> wrote:

On 2021-10-02 8:17 p.m., 'b alex pettit jr' via Society of Amateur Radio Astronomers wrote:

A few questions ..

1) From Marcus' calculations, this  " 0.6 dB "  data with my current setup is reasonably Ok ?

Yup, quite reasonable for your small antenna.

A larger antenna, where the H1 cloud fills the aperture can produce up to 2dB above the
  noise floor.

2) The noise floor is mostly Cosmic and Earth background rather than LNA Noise ?

Actually, from the equations I provided, a goodly portion of the total noise at any given time
  *IS* from the system itself.   If your Tsys is 100K (typical for an amateur effort), then the
  sky is only perhaps 15-50K, depending on aperture and where you look.

3) Then, What might this hardware provide over the current setup ?

https://www.radioastronomysupplies.com/store/p61/1420_MHz._HIGH_PEFORMANCE_LNA%2C_NF_0.29dB%2C__GAIN_37dB.html

The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB
  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the
  Tant from before -- 15K for "interesting" and 5K for "boring":

(15+70)/(5+70)

=

About 0.5dB



A little bit better than with the SawBird H1.  Whether that's "worth it" for a small antenna?
  That's up to you and what your budget will bear.

Thanks,

Alex

This is a set of  8 ea, 10 minute drift averages  taken at Dec 0 Deg ,. RA  6:30 - 7:30 Hrs. ( or thereabout )

( Vert Axis : dB ref to Background )

.

--
--
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/1802066057.799123.1633220246621%40mail.yahoo.com.

--
--
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/fc9d31ba-7710-2c58-15c3-704de4b2962f%40gmail.com.

[Marcus D Leech]

Here is a good description of G/T measurements and what they mean. 

Sent from my iPhone

On Oct 3, 2021, at 11:59 PM, Anthony <itpart...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TzsJh%2BZGGvyZ1_Nxruaem4zQny8k30gGDHZyYsKYJWyQg%40mail.gmail.com.

[Peter East]

The SNR dB conversion confusion arises for two reasons,

1. The output of a square-law detector and the equivalent of FFT I/Q squaring is actually a voltage and it is proportional to the input power (Tsys+Tant).

2. The IEEE definition of signal-to-noise ratio is as a power ratio at the measurement point.

So strictly it should be calculated as 20xlog at the receiver output.

But we are interested in the signal-to-noise ratio at the antenna input so we usually calculate it as 10xlog

To minimise confusion it is probably best for radio astronomers dealing in detected noise to agree our measurements as a linear input power ratio (Tsys+Tant)/Tsys and forget dB.

On Sunday, October 3, 2021 at 9:59:33 AM UTC+1 a_pet...@yahoo.com wrote:

POWER    or    VOLTAGE ??

For Rx, would the amplitude comparisons not be 20 Log ?

( If you note from the plot, "20" was my Y axis scaling factor )

A Power comparison drops my dB values in half ..

Alex

==========================================

From Marcus's Previous Posts

 If your Tsys is 100K (typical for an amateur effort), then the
  sky is only perhaps 15-50K, depending on aperture and where you look.
 

The NooElec LNA is about 0.7dB NF (51K) and the amplifier above claims about 0.3dB

  (21K).  So, your Tsys would go down from (let's say) 100K to 70K.  So if we take the
  Tant from before -- 15K for "interesting" and 5K for "boring":

(70 + 15 /(70 + 5 ) = 115 / 105
= about 0.5 dB     ( Power 10*Log(x) )

vs ( the Nooelec LNA )

( 100 + 15 ) / ( 100 + 5 )  = 115/105
= 0.4 dB      ( Power 10* Log(x) )

So, I've gone over this before, but let's review.

Your Tsys will be somewhere in the range of 100K (100-150K, let's say) with these setups.   The beam-averaged Tant (antenna temperature) may be about 20-30K    when a "good" part of the hydrogen sky occupies some part of the beam, and perhaps 10K when there's "nothing particular" in the beam.

So, the ratio between:

(Tsys + Tsky_interesting ) / (Tsys+Tsky_boring)

=  130 / 110

=  Is about 0.7dB  Power,   1.4 dB Voltage  ?

BUUUUTT

You're only going to "see" half of those Tsky temperatures, because of polarization loss (source is randomly polarized, instrument is fixed).

So, it then becomes:  115/105  approximately 0.4dB

You result is somewhere in between, at about 0.6dB.   You're doing fairly well.

If we drop Tsys by, let's say, 10K, we get (keeping the other very-rough approximations constant):

105/95  = 0.43dB

The main thing that happens with very-small apertures is that the "thing of interest" only occupies a smallish part of your beam area.  Even though H1 clouds are   "extended" they generally *will not* fill your entire beam/FOV for these very-small aperture antennas.  Which means your beam is "diluted" with a lot of sky   that is quite a bit colder (in the radio brightness sense) than the H1 clouds you're trying to measure.  Once your Tsys is "reasonable", at this scale, more   aperture is what is called for, IMHO.

-

[Marcus D. Leech]

On 2021-10-04 7:16 a.m., Peter East wrote:
> The SNR dB conversion confusion arises for two reasons,
> 1. The output of a square-law detector and the equivalent of FFT I/Q
> squaring is actually a voltage and it is proportional to the input
> power (Tsys+Tant).

The units in the output of a "raw" FFT are the same as the input units. 
This means that when you square those units, you are producing a number
that is
  proportional to the *power* represented in the input at the given
frequency.  That proceeds directly from Ohms Law.

What is not clear to me is what the units are that are in the IF_Average
output of the SDR# software.

My own software uses the scaling conventions that are typical in the
industry, and used throughout Gnu Radio software, which is to say that the
  MAG**2 is used, and then converted to dBFS (dB full-scale) on a
10*Log10 basis.

> 2. The IEEE definition of signal-to-noise ratio is as a power ratio at
> the measurement point.
> So strictly it should be calculated as 20xlog at the receiver output.
> But we are interested in the signal-to-noise ratio at the antenna
> input so we usually calculate it as 10xlog
> To minimise confusion it is probably best for radio astronomers
> dealing in detected noise to agree our measurements as a linear input
> power ratio (Tsys+Tant)/Tsys and forget dB.
>

I have gone "back and forth" on this issue many times in the last
17(!!!) years of being involved with SDR and back-yard radio astronomy. 
Since we mostly operate at
  the "edge" between strictly science and strictly engineering, and 99%
of the lab equipment we might encounter is overwhelmingly likely  to use
dB units, I use
  dB units in my outputs.   When producing things like maps, I just
convert them back into linear form, and I usually preserve enough digits
in the dB form that
  there's no significant loss of precision during that conversion.

When I give the next BAA seminar, one of the topics will be the format
of the data files, and we could have that discussion there as well....

[Paul Oxley]

Marcus

The mounting of the dish on the mount is also important. Wherever possible, the dish should be centered as near as possible on the mount. This would tend to create equal but opposite forces from a given wind direction.

In the designed operation of the outdoor units, there would be substantial wind load from the camera housing. These cameras were big and were enclosed for protection from the weather.

Paul

--
--
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/da415c21-ef44-8763-cb93-35ad4df236e7%40gmail.com

.

[Marcus D. Leech]

On 2021-10-04 10:03 p.m., Paul Oxley wrote:

Marcus

The mounting of the dish on the mount is also important. Wherever possible, the dish should be centered as near as possible on the mount. This would tend to create equal but opposite forces from a given wind direction.

In the designed operation of the outdoor units, there would be substantial wind load from the camera housing. These cameras were big and were enclosed for protection from the weather.

Paul

Sure.  Although, there's a *significant* difference between the couple of ft**2 wind loading of a camera housing, and the rather-larger 78ft**2 of a 10ft dish.
  Also different shapes have different Cds, and thus are harder to compare directly without knowing the precise Cd, even if you know the precise projected
  area.

Like I asked several messages ago in this thread, has anyone ever actually used the largest Pelco P/T mount for a 10ft (or larger?) dish?  The proof of
  the pudding, and all that.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/1529430471.2048986.1633399426079%40mail.yahoo.com.