OTA recommendations

[Tower Guy]

Originally Posted by rabbit73

This part was added by holl_ands, who does modeling and is an RF engineer, so I took it on faith because I'm not smart enough to verify it:

Holland’s is correct, but note that he started with two signals already, so add 6 dB to his numbers to get the Peak power ratio to a single signal.

[rabbit73]

Originally Posted by tripelo

This topic probably could be in another thread.

Would like your thoughts on dynamic range of Airspy vs. SDRplay RSP1.

Rabbit73, do you see differences in instantaneous dynamic between the two scanners, for:

- Wideband (several channels displayed)
- Single channel (display maybe 8-10 MHz or so)

No hurry on response. Do not wish to derail current thread.

Thanks

Good to hear from you.

The OP and I are both interested in SDRs as a diagnostic tool. I see no harm in a brief response.

The dynamic range of the SDR is determined by the bit depth of the ADC (Analog to Digital Converter). The Airspy R2 and SDRplay RSP1A are very similar in that respect; 12 bits. Both are far superior to the RTL-SDR.COM V3 dongle which has an 8 bit depth ADC, giving it a dynamic range of about 50 dB. Comparison chart here:
Comparisons with other common Wideband Commercial Software Defined Radios
https://www.rtl-sdr.com/about-rtl-sdr/

For a scan of 8-10 MHz for a single channel, both are quite rapid. For a wider scan, it takes slightly longer as would be expected, but it is real-time that can show changes. The Spectrum Spy scanner is faster than the SDRplay scanner. Both scanners are much more rapid than the RTL-SDR Scanner which takes forever to piece together individual sections into a complete scan; not real-time.

The separate scanner software available for the RSP1A looks like an old-time SA. The Spectrum Spy software that is included in the SDR# download for the R2 is very colorful. I prefer it to the scanner for the RSP1A, but the R2 costs twice as much as the RSP1A.

Some videos
Channels are 16, 19-21, 31-33, cellular

Scanner for RSP1A

still



https://www.dropbox.com/s/xdysaeqzc3...27-59.mp4?dl=0

https://www.dropbox.com/s/xdysaeqzc3...27-59.mp4?dl=1

hide controls

https://www.dropbox.com/s/nl1876xejj...32-41.mp4?dl=0

https://www.dropbox.com/s/nl1876xejj...32-41.mp4?dl=1

Spectrum Spy for R2

still



https://www.dropbox.com/s/gars272dis...12-56.mp4?dl=0

https://www.dropbox.com/s/gars272dis...12-56.mp4?dl=1

[rabbit73]

Originally Posted by Tower Guy

Holland’s is correct, but note that he started with two signals already, so add 6 dB to his numbers to get the Peak power ratio to a single signal.

Thank you.
73

post by majortom about 1/4 wave stubs
use 7th harmonic
https://www.digitalhome.ca/threads/o...5/post-3109007

[rabbit73]

Originally Posted by verder

I keep checking to see if I can post directly to the reply. I still cannot.

I actually picked up the MediaSonic tuner as a result of watching Tyler the Antenna Man's youtube video. The signal quality display view is identical to your CM7003 converter box.

When I had dropout on the MediaSonic, the quality was easily 70% or so. Then, it would go to 0% and lose signal, then come back and be back in the 70% range again. So, using that a measure of performance wasn't all that useful.

The Mediasonic is telling you the truth.

When it goes to 0% signal quality, there are more errors than the FEC (Forward Error Correction) is able to correct. The FEC has a limit to its ability to correct errors.

When it comes back to the 70% range, there are no uncorrected errors.

If you had an HDHR, it would show a green Symbol Quality bar at 100% when there was 70% signal quality on the Mediasonic.

The HDHR Symbol Quality bar would be red at the times when the Mediasonic was going to 0%.

The increase in errors is most likely from the signal passing through the trees. You will continue to have multipath dropouts as long as there are trees in the signal path.

The same thing is happening to me because there is a tree in front of my antenna; it's worse with my weakest channels.

https://www.dropbox.com/s/q79o3x433h...41-08.mp4?dl=0

https://www.dropbox.com/s/q79o3x433h...41-08.mp4?dl=1

[rabbit73]

Originally Posted by verder

Support indicates that KRDK and WDAY are still coming in too strong.

You tried an attenuator, and it didn't help with the ghost artifact or dropouts, did it? You could try more attenuation.

They also recommended using an LTE filter. I actually have one and used it previously but it did not work. I put it back in-line and also moved to the -7dB output of the BDS103H (previously I was on the -3.5dB output.)

There are two CM LTE filters with the same model number CM-3201. You would need the newer 600 MHz LTE filter if your cellular signals were strong enough to cause interference, but they don't look that strong in your scan.

I also mentioned that the back and forth with support since I have no access to the tuner's quality measures is a pain. I brought up that the HDHomeRun provides this information and it is disappointing that the Tablo doesn't expose this info. Further, my request for 'inside' access to the Tablo went completely ignored. No surprise, but equally disappointing.

In other news, I've gotten absolutely no where with the US distributor of the Airspy R2. They wanted to pawn me off to community support since I had already tried everything they could think of. I pushed back and they directed me to contact Airspy HQ directly. If I don't hear back from them, I plan to contact my credit card company and see if they can rattle their cage. Granted it's only been two days since I e-mailed Airspy HQ, but I haven't even gotten an acknowledgement yet that they received my request so I have my doubts.

Two examples of poor customer support; typical performance for these times. Good customer support would cost them money; the bottom line is more important than the customer.

Their first line of defense is that the problem must be your fault. The current form of capitalism is heartless.

Can you use your Airspy R2 now for scans?

Did you try removing the GitHub SNR meter plugin?

Did you try a new download of SDR# after removing the old one?

There is a strong FM signal coming from the NNW. It's only 7 kW, but it's 1 mi away.
http://www.fmfool.com/modeling/tmp/9...c/Radar-FM.png

Makes me wonder about FM interference.

[rabbit73]

Originally Posted by verder

I forgot to mention that I *know* the splitter is doing something. After I moved to the -7dB output and rescanned on the Tablo, K28MA is now showing up as 3 Orange dots rather than 5 Green dots for signal strength. Everything else is 5. Thankfully, K28MA is a duplicate of KXJB so I'm not really concerned here.

A quick < 1 minute view of all the XX-1 channels shows that they are at least coming in. No clue if they suffer from any dropout or pixelation issues.

That tells you the attenuator is making the signals weaker, but you don't yet know if reducing the signal strength reduces the dropouts and pixelation.

[tripelo]

Originally Posted by rabbit73

...

The dynamic range of the SDR is determined by the bit depth of the ADC (Analog to Digital Converter). The Airspy R2 and SDRplay RSP1A are very similar in that respect; 12 bits. Both are far superior to the RTL-SDR.COM V3 dongle which has an 8 bit depth ADC, giving it a dynamic range of about 50 dB...

Yes, thanks for your response & images/videos.

As you imply above, for a particular bandwidth,
the number of bits in A/D converter set a limit on dynamic range.

In addition,
linearity of preceding preamplifier & mixers can degrade dynamic range from that of A/D converter alone.

Possibly your received signals are not strong enough to stress the linearity of the RTL front end.

When you have the scanners them properly adjusted, have you seen any spurs in either scanner?

----------------------

Your images and videos are informative.

Good to see a comparison of both scanners displays.

Reading the scale looks like about 60+ dB dynamic range displaying those particular signals.

Have you checked the accuracy of the dB scale?

Is there a means in software to calibrate the dB scale against a reference?

Wonder, when calibrated at one or two points, if it maintains some accuracy over a larger range of dB?

Do you think one of the two scanners could be more accurate than the other?

Thanks

[rabbit73]

Originally Posted by tripelo

When you have the scanners them properly adjusted, have you seen any spurs in either scanner?

There appears to be spurs in all three of my scanners, but it's not unexpected in equipment that isn't lab grade. I would need a pure source to really tell the spurs from actual signals.

The RTL-SDR dongle has a lot of spurs when I put a 75 ohm termination resistor on the front end, but they are minimized when signals are measured.



I think the Spectrum Spy scanner for the R2 is a lot cleaner than the RSP Spectrum Analyzer for the RSP1A.

Your images and videos are informative.

Good to see a comparison of both scanners displays.

Thank you. I was curious to make the comparison myself.

Reading the scale looks like about 60+ dB dynamic range displaying those particular signals.

A 12 bit ADC should be able to do at least a 60 dB dynamic range.

Have you checked the accuracy of the dB scale?

The dB scales on my three SDRs are relative, not absolute. There is one exception, the calibration of the RSP1A with the SDRuno software is calibrated under certain conditions:

SDRplay RSP2 for accurate RF power Measurement
https://www.youtube.com/watch?v=MRR-x_TjOp4



This is my test of the SDRplay calibration:



I have used that feature to measure the strength of TV signals, but a correction factor of about +14 dB is needed:

Measuring TV Signal Strength with an SDR
https://www.avsforum.com/threads/the.../post-59870798

Is there a means in software to calibrate the dB scale against a reference?

I think there are offset adjustments for the relative scale, but I haven't hunted for them because the position of the signal changes with adjustments of gain and display.

I think the calibration feature of the SDRuno mentioned above is fixed in the hardware or firmware.

Wonder, when calibrated at one or two points, if it maintains some accuracy over a larger range of dB?

I have not tested the linearity of the relative dB scales. The accuracy of the "use under certain conditions" calibration of the SDRuno with RSP1A SDR appears to hold well over a wide range.

Do you think one of the two scanners could be more accurate than the other?

I don't think the relative scales of either one could be considered "accurate" with so many stages in series.

[rabbit73]

I am working on a method of measuring the strength of a TV signal using only the TV tuner and an attenuator. It should help verder to be able to measure the strength of his strongest signal which seems to be in question.

I will also do some SDR accuracy measurements at the same time which should be of interest to tripelo.

I need some sleep now.

[tripelo]

Originally Posted by rabbit73

There appears to be spurs in all three of my scanners, but it's not unexpected in equipment that isn't lab grade. ...

Yes.

The RTL-SDR dongle has a lot of spurs ...

Have noticed that in the low cost RTL devices. Most of the time not a problem if one is aware of their presence.

...I think the Spectrum Spy scanner for the R2 is a lot cleaner than the RSP Spectrum Analyzer for the RSP1A...

Interesting.

..The dB scales on my three SDRs are relative, not absolute.

Yes, the unit of dB itself is relative.

When uses as dBm or dB(other), then it takes on more absolute meaning.

-------------------

Curious about scale accuracy, sorta related to scale linearity:

Is a dB indicated, an actual dB.

For example:

Is a -20dB indication really 10dB less than a -10dB indication? Or,

Is the delta, 9 or 11 dB or some other number.

-------------------

...I think there are offset adjustments for the relative scale,...

....I don't think the relative scales of either one could be considered "accurate" with so many stages in series.

Thanks for the information and the images, helps to better understand these interesting devices.

.

[rabbit73]

Here is an example of using an attenuator to measure the strength of a digital TV signal. I picked RF channel 20 to measure.



This is my signal report in dBm:
https://www.rabbitears.info/searchma...145947&opkey=C



This was my setup:

                                  TV for SNR
                 0 to 10 dB          /
Ant > 25ft RG6 > Attenuator > A/B Switch 
                  at 0 dB            \
                                  Meter or SDR

The Sadelco DisplayMax 800 signal level meter reading was -7.3 dBmV = -56.1 dBm

I added attenuation, for a total of 30 dB, to bring the signal down to 15 dB SNR at the "Digital Cliff."

                                                TV for SNR
                 0 to 10 dB      Fixed             /
Ant > 25ft RG6 > Attenuator > Attenuators > A/B Switch 
                  at 7 dB        20+3 dB           \
                                                Meter or SDR

This is what the scan looked like at dropout. Note that I had to increase the SDR gain for the proper display:



So, to calculate the signal strength, the basic rule says to add the amount of attenuation needed for dropout to -85 dBm, which is the average dropout point of a tuner:
-85 dBm + 30 dB = -55 dBm

The signal level meter reading for the signal was -56.1 dBm; close enough for a good estimate to determine the signal strength.

The 30 dB of attenuation needed to bring the signal down to 15 dB SNR is sometimes called Margin to Dropout.

Channel 20 is listed at -26 dBm on my report. The trees are causing a lot of signal loss.

Note that I really only needed the TV and the attenuator to determine the signal strength. I added the SDR and the signal level meter to verify the method.

[rabbit73]

Originally Posted by tripelo

Curious about scale accuracy, sorta related to scale linearity:

Is a dB indicated, an actual dB.

The only dB scale on an SDR that I have confidence in is the dBm calibrated scale for the RSP1a with SDRuno software. I used it during the above test.









I couldn't figure out a way to get a direct reading of the signal power as shown in the YouTube video or with my NTSC modulator, so I had to settle for using a correction factor. It's probably measurement-bandwidth related.

Correction factor
99.5 - 85 = 14.5
-70.1 dBm + 14.5 = -55.6 dBm

Regarding linearity:
The signal does appear to be about 30 dB weaker at dropout on the calibrated dBm scale AND the relative dB scale.

With a flat signal, the placement of the measurement window isn't very critical, but with a signal that doesn't have a flat top, it's up to you to place it at an average level point.

This is the way my signal level meter does it, and takes the average:

[tripelo]

Originally Posted by rabbit73

The only dB scale on an SDR that I have confidence in is the dBm calibrated scale for the RSP1a with SDRuno software....

Thanks.

That is quite a difference between your calibrated RSP1a and the Airspy.

.

[rabbit73]

Originally Posted by verder

It still functions. I get ghost spikes now is the problem.

SNR meter plugin is a non-issue. As mentioned previously, the ghost spikes occur on completely separate machines with their own installs of the latest SDR# (and those did not use the plugin.)

If it's FM interference, then it's a mystery why I didn't see this for the first two weeks or so of owning the device. I've done a bit of messing around with trunked voice comms and never saw this issue.

Thank you for your reply.

[rabbit73]

Originally Posted by verder

As of this morning I still get weak signal reports (recording failed @ 1am). Current setup:

antenna input > cm3201 > splitter/7dB drop > tablo

I have the newer CM3201.

We watched a football game yesterday on the same channel (KVRR) as this morning's failed recording and didn't notice any issues except for maybe one short ~1 second occurrence that had digital smearing. I believe the weather was clear overnight.

I'm currently waiting for Tablo support to get back to me in this continuing saga of e-mail tag. I learned that they send your case to the back of the line every time you update them with new info via email. I guess it pays to be quiet. :P

I hope that they will take your request seriously enough to give you some useful help.

I have half a mind at this point to get a ClearStream 2Max and replace the 4Max. Wondering if the wider beamwidth will help with KVRR reception or what.

The CS 2MAX does have a wider beamwidth but a little less gain. Difficult to predict the results of a test.

We are now at post #55 on your thread and I'm running out of ideas. I've learned a lot trying to help you and have had a chance to show off some of my ideas about SDRs and how to measure the strength of a TV signal without a signal level meter.

I keep picturing your antenna above the trees, but I know that isn't going to happen.

If you went to AVS, Calaveras would probably say get your antenna above the trees.

Example of Vegetation Affecting Signal Strengths-Part 1
https://www.avsforum.com/threads/exa...art-1.2005866/

[rabbit73]

I was wondering why Tablo support thinks your signals are too strong. This seems to answer the question:
Getting Technical with Over-the-Air TV Reception (AKA: Why Can’t I Watch This Channel?)
https://www.tablotv.com/blog/getting...-tv-reception/



A GOOD tuner should be able to receive signals much stronger than -50 dBm. Apparently, the Tablo tuners don't meet the ATSC specs:

ATSC Recommended Practice:
Receiver Performance Guidelines

Document A/74:2010, 7 April 2010

RECEIVER PERFORMANCE GUIDELINES

5.1 Sensitivity

A DTV receiver should achieve a bit error rate in the transport stream of no worse than 3x10E-6 (i.e., the FCC Advisory Committee on Advanced Television Service, ACATS, Threshold of Visibility, TOV) for input RF signal levels directly to the tuner from –83 dBm (-34 dBmV) to –5 dBm (+44 dBmV)for both the VHF and UHF bands.

-50 dBm + 49 = -1 dBmV

Cable guys want you to have at least that much signal at the wall plate for your TV.

Maybe they got the idea from SiliconDust. 100% on the HDHR GUI is only one dB more at 0 dBmV.

Most of my Sony TVs say MAX or 100 at 0 dBmV except for one model that goes much higher:

[rabbit73]

Originally Posted by verder

I appreciate the time you've put into this so far! I'm learning quite a bit in the process.

I find this comment from Calaveras very telling:

Without elevating things above the tree-line, I'm thinking this is as good as things are going to get for me. A CS2M may help out with KVRR reception... I'd most likely remove the splitter or go back to the -3.5dB output in that case. If it does, I could probably donate my CS4M to my father-in-law who lives out in the country and relies on an el-cheapo mudflap for his OTA reception. So, it wouldn't be a total loss.

That sounds like a good plan.

[verder]

My account has finally been activated. I PM'd the "admin" user and was approved last night I guess. Cue this clip!

[rabbit73]

Originally Posted by verder

My account has finally been activated. I PM'd the "admin" user and was approved last night I guess.

Congratulations!

A little late, but they finally got to you.

Cue this clip!
For the moment you get excited about whatever makes you "somebody." (Steve Martin; from the 1979 movie The Jerk).

Ha Ha, that's a good one.

Welcome, verder; you are now "somebody" on the TVFool website.

[eclipsme]

Welcome. I have been following along, picking up a few things along the way.
Congrats!