Donate to Remove ads

Got a credit card? use our Credit Card & Finance Calculators

Thanks to johnstevens77,Bhoddhisatva,scotia,Anonymous,Cornytiv34, for Donating to support the site

Spectrum Analyzer?

Seek assistance with all types of tech. - computer, phone, TV, heating controls etc.
XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Spectrum Analyzer?

#403419

Postby XFool » April 11th, 2021, 3:27 pm

Has anyone experience of using and/or can recommend an affordable, easily available, analytical RF spectrum analyzer, covering at least 80 MHz - 800 MHz.

As an example, this looks to be the kind of thing, though strangely it doesn't seem to me to quote a frequency range and it's also an American sourced product: http://rfexplorer.com

TIA


PS. Oh I now see there are different models covering different RF ranges.

AleisterCrowley
Lemon Half
Posts: 6381
Joined: November 4th, 2016, 11:35 am
Has thanked: 1880 times
Been thanked: 2026 times

Re: Spectrum Analyzer?

#403424

Postby AleisterCrowley » April 11th, 2021, 4:00 pm

'affordable' and 'RF spectrum analyzer' rarely appear in the same sentence...
There is a lot of 'test equipment' available on eBay - mostly Chinese origin- but I have my doubts about accuracy and reliability
A lot of stuff is based around counterfeit chips - I saw one power meter based on the Analog Devices AD8307 which was selling for less than the bulk cost of the chip itself... so almost certainly a counterfeit with unknown performance
Do you need portability ? What is the maximum frequency span you want to look at?
There are some solutions available based on SDR receivers (which plug in to a laptop)

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#403429

Postby XFool » April 11th, 2021, 4:26 pm

AleisterCrowley wrote:'affordable' and 'RF spectrum analyzer' rarely appear in the same sentence...

That's what I thought. Though I see some cheap ones listed at the usual places online.

AleisterCrowley wrote:Do you need portability ? What is the maximum frequency span you want to look at?

Yes, portable/hand-held would be ideal. Frequency range as I said, 80 - 800 MHz. In other words mainly for VHF/FM and DTT RF broadcast frequencies.

AleisterCrowley
Lemon Half
Posts: 6381
Joined: November 4th, 2016, 11:35 am
Has thanked: 1880 times
Been thanked: 2026 times

Re: Spectrum Analyzer?

#403437

Postby AleisterCrowley » April 11th, 2021, 4:48 pm

Well, if you want 'affordable' it probably means Chinese - something from here?
https://uk.banggood.com/Wholesale-Spect ... 11847.html
I've no experience of this company

Do you want to look at the whole 80-800MHz span at the same time? It's not clear what the max span is for some of these. I know the SDR ones are sometimes limited to any 10MHz chunk within their range

servodude
Lemon Half
Posts: 8271
Joined: November 8th, 2016, 5:56 am
Has thanked: 4435 times
Been thanked: 3564 times

Re: Spectrum Analyzer?

#403506

Postby servodude » April 12th, 2021, 1:00 am

AleisterCrowley wrote:Well, if you want 'affordable' it probably means Chinese - something from here?
https://uk.banggood.com/Wholesale-Spect ... 11847.html
I've no experience of this company

Do you want to look at the whole 80-800MHz span at the same time? It's not clear what the max span is for some of these. I know the SDR ones are sometimes limited to any 10MHz chunk within their range


I've ordered stuff from Banggood (...this is a reminder to run business names past a focus group ;) )
- it arrived and did what it was meant to

And I have a friend who was refunded the money for a build your own portable scope that was just terrible from them

If you can find reviews (real reviews!) or people using the hardware in a blog post write up or similar you'll be fine

saying that if I was looking for a cheap wide range spec an(1) for home use I'd probably roll my own using a:
https://www.adafruit.com/product/1497
and
https://www.rtl-sdr.com/tag/spectrum-analyzer-2/

- sd

(1) for lower bandwidth stuff there are now plenty of decent cheap portable or small scopes with spectrum capabilities (Owon/Rigol/Siglent)

AleisterCrowley
Lemon Half
Posts: 6381
Joined: November 4th, 2016, 11:35 am
Has thanked: 1880 times
Been thanked: 2026 times

Re: Spectrum Analyzer?

#403532

Postby AleisterCrowley » April 12th, 2021, 9:36 am

I'd probably go down the SDR route, but some of those portable SAs look amazing for the price IF they;
(a)meet the specs
(b)carry on working for years, not weeks
I suspect they will have lots of internally generated spurious responses etc, and will also fall over if connected to an expternal antenna but happy to be proved wrong... I may even buy one for fun..
I haven't used any decent test equipment for years (as I now seem to be an acccountant) All the good stuff was from the likes of HP/Agilent, Rohde&Schwarz, Marconi etc

scotia
Lemon Quarter
Posts: 3561
Joined: November 4th, 2016, 8:43 pm
Has thanked: 2371 times
Been thanked: 1943 times

Re: Spectrum Analyzer?

#403861

Postby scotia » April 13th, 2021, 12:42 pm

AleisterCrowley wrote:I'd probably go down the SDR route, but some of those portable SAs look amazing for the price IF they;
(a)meet the specs
(b)carry on working for years, not weeks
I suspect they will have lots of internally generated spurious responses etc, and will also fall over if connected to an expternal antenna but happy to be proved wrong... I may even buy one for fun..
I haven't used any decent test equipment for years (as I now seem to be an acccountant) All the good stuff was from the likes of HP/Agilent, Rohde&Schwarz, Marconi etc

I have been playing around with one of the more common SDR chip sets (RTL2832U and R820T2) - I bought the NooElec NESDR SMArt v4 bundle (with three aerials) for £38 from Amazon. I chose this one since it has an aluminium enclosure - which dissipates the considerable heat generated by the chips. I had heard problem reports of cheaper products that were plastic encapsulated.
I'm using it as a radio - with numerous free software bundles. I listen into 1090MHz for the aircraft ADS-B transmissions, to VHF/FM broadcasts and to Digital Radio Broadcasts. So scanning through 80MHz to 800Mhz (in steps) should be well within its capabilities. But yes - you will get spurious responses from strong signals, particularly if you operate it at high gain. Many of the SDR (free) software packages display the section of the spectrum you are currently looking at, and there are also software packages which specifically describe themselves as Spectrum Analysers - but I haven't tried any.
https://www.rtl-sdr.com/about-rtl-sdr/ will provide lots of links

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#403899

Postby XFool » April 13th, 2021, 2:46 pm

...Meanwhile. Aerial guy in attendance. He has been in attendance for quite a long time now - couple of hours.
(I am currently getting a better quality, if weaker, signal on R4 FM with the aerial lead unplugged from the wall then I've had for ages with it plugged in!)

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#403912

Postby XFool » April 13th, 2021, 3:07 pm

AleisterCrowley wrote:I haven't used any decent test equipment for years (as I now seem to be an acccountant) All the good stuff was from the likes of HP/Agilent, Rohde&Schwarz, Marconi etc

Ah! All the 'old' names. :)

Who was it made those old UK make of impedance bridge testers.

Anyway... Aerial guy finished: water in cable; had to replace. Plus strong 4G signal (water in cable could have made worse); 4G filter fitted.
He had his own spectrum analyser. :D

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#403925

Postby XFool » April 13th, 2021, 3:58 pm

Next mystery! (We are never going to run out of mysteries in today's digital world...)

Programme channel 'London Live' is transmitted from Crystal Palace. It is on the local service multiplex (LCN 8) for London, the LTVMux L-LON or MuxLW - London (take yer pick!) transmitted at 30kW on RF channel 35 at 586 MHz.

I remember a while ago now watching the old 60s/70s series 'Callan' on it. Not recently though, it pixelated all over the place. Before the aerial guy did his work I had noticed this (relatively lower power) channel/Mux was indicated by my box as having a somewhat weak signal (~13 dB ?) but with a zero error rate. Thinks: So why is it pixelating?

Now, after the aerial guy has done his work, signal strength is indicated as 19 dB (95%) with zero errors. Hurray! So why is it pixelating and shimmering all over the place - exactly the same way as before the aerial guy?

This MUST be an interfering signal, no? But then, on DTT this would show as a high error rate, no? But it does not have, nor has it ever had a high error rate... And if it is now shown as full strength...

The only thing I can think of, if no interference on this channel, is SNR. It is transmitted at rather lower power than the main Muxs and I am some way from Crystal Palace. But zero error rate? I don't understand. Something to do with type of modulation? (Sometimes I want to go back to 405 lines...)

Anyone any suggestions?

pochisoldi
Lemon Slice
Posts: 941
Joined: November 4th, 2016, 11:33 am
Has thanked: 31 times
Been thanked: 462 times

Re: Spectrum Analyzer?

#404072

Postby pochisoldi » April 14th, 2021, 12:48 am

XFool wrote:But zero error rate? I don't understand. Something to do with type of modulation? (Sometimes I want to go back to 405 lines...)

Anyone any suggestions?


Data can be sent with redundant data added.

https://en.m.wikipedia.org/wiki/Error_correction_code explains this better than I can (scroll down to Forward Error Correction)

In short - add error correction bits and you can detect and correct errors at the receiver, at the cost of reduced usable bandwidth.

What could be happening in your case is that the data stream is being demodulated, error correction applied and you are actually getting an error free signal.

However in many telly's the incoming signal will be a 576/720/1080 line signal, but your screen only does (say) 1080 or 1920 native. The TV will have to convert. Throw in an incorrect wide-screen setting and you start getting "artifacts" - blocky areas that kind of look kind of like pixelation, but the techie in you says "nah", or my favourite in my local pub with a mahoosive but cheap LED screen, footballs that seem to move with a jerky motion.

In my experience, if the receiver says the signal quality is OK, but you are seeing artifacts, then it's incorrect (usually obscure) screen settings or p***poor TV software.

Number one thing to look in respect of wrong setting for is evidence of overscanning. In the case of my local, the Sky Sports logo is often off the edge of the screen indicating that the TV is doing unnecessary image processing, which leads to an "underpowered" TV giving poor picture quality.

Also look for what I would call "underscanning" - adding unwanted black stripes top/bottom or left/right. Not an issue with an old 4:3 format picture, but not required on a 16:9 telly being fed a 16:9 image.

HTH

servodude
Lemon Half
Posts: 8271
Joined: November 8th, 2016, 5:56 am
Has thanked: 4435 times
Been thanked: 3564 times

Re: Spectrum Analyzer?

#404076

Postby servodude » April 14th, 2021, 1:31 am

pochisoldi wrote:
XFool wrote:But zero error rate? I don't understand. Something to do with type of modulation? (Sometimes I want to go back to 405 lines...)

Anyone any suggestions?


Data can be sent with redundant data added.

https://en.m.wikipedia.org/wiki/Error_correction_code explains this better than I can (scroll down to Forward Error Correction)

In short - add error correction bits and you can detect and correct errors at the receiver, at the cost of reduced usable bandwidth.

What could be happening in your case is that the data stream is being demodulated, error correction applied and you are actually getting an error free signal.

However in many telly's the incoming signal will be a 576/720/1080 line signal, but your screen only does (say) 1080 or 1920 native. The TV will have to convert. Throw in an incorrect wide-screen setting and you start getting "artifacts" - blocky areas that kind of look kind of like pixelation, but the techie in you says "nah", or my favourite in my local pub with a mahoosive but cheap LED screen, footballs that seem to move with a jerky motion.

In my experience, if the receiver says the signal quality is OK, but you are seeing artifacts, then it's incorrect (usually obscure) screen settings or p***poor TV software.

Number one thing to look in respect of wrong setting for is evidence of overscanning. In the case of my local, the Sky Sports logo is often off the edge of the screen indicating that the TV is doing unnecessary image processing, which leads to an "underpowered" TV giving poor picture quality.

Also look for what I would call "underscanning" - adding unwanted black stripes top/bottom or left/right. Not an issue with an old 4:3 format picture, but not required on a 16:9 telly being fed a 16:9 image.

HTH


I'd agree
- it sounds like it's either encoded poorly or in a way the TV doesn't like

Normally the Bit Error Rate reported is based on the amount of work the error correction had to do (if you didn't have the error checking//recovery bits you couldn't know if you had a problem)
It's not clear whether it's the "recoverable" or "unrecoverable" count that is being shown
- but either way 0 implies you're getting out what the broadcaster expected ;)

Anyways getting a bit geeky:

from https://www.ofcom.org.uk/spectrum/information/transmitter-frequency
TVMux Multiplex L Local TV Multiplex Comux UK Ltd. DVB-T: QPSK, code rate 3/4, 8K FFT
- QPSK is about as robust/simple a modulation strategy you'd use these days (constellation only has two bits per quadrant)

so again it points to the noise likely being in the data - or injected after it's been unpacked

- sd

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#404252

Postby XFool » April 14th, 2021, 5:17 pm

pochisoldi wrote:Data can be sent with redundant data added.

https://en.m.wikipedia.org/wiki/Error_correction_code explains this better than I can (scroll down to Forward Error Correction)

In short - add error correction bits and you can detect and correct errors at the receiver, at the cost of reduced usable bandwidth.

What could be happening in your case is that the data stream is being demodulated, error correction applied and you are actually getting an error free signal.

However in many telly's the incoming signal will be a 576/720/1080 line signal, but your screen only does (say) 1080 or 1920 native. The TV will have to convert. Throw in an incorrect wide-screen setting and you start getting "artifacts" - blocky areas that kind of look kind of like pixelation, but the techie in you says "nah", or my favourite in my local pub with a mahoosive but cheap LED screen, footballs that seem to move with a jerky motion.

In my experience, if the receiver says the signal quality is OK, but you are seeing artifacts, then it's incorrect (usually obscure) screen settings or p***poor TV software.

Number one thing to look in respect of wrong setting for is evidence of overscanning. In the case of my local, the Sky Sports logo is often off the edge of the screen indicating that the TV is doing unnecessary image processing, which leads to an "underpowered" TV giving poor picture quality.

Also look for what I would call "underscanning" - adding unwanted black stripes top/bottom or left/right. Not an issue with an old 4:3 format picture, but not required on a 16:9 telly being fed a 16:9 image.

Thanks, pochisoldi, but I am not convinced any of that really applies in this case.

London Live is the only programme channel I use that has this problem: jittering picture, flicking at top of picture, blocking, rapid large scale pixelation, frame breaking up/slipping, frame freezes. All while still at constant, unvarying, 95% Signal Strength, 100% Signal Quality.

Whenever I have problems with other channels it was either a wipeout: Signal Strength 0%, Signal Quality 0% - 'No signal. Please check antenna/cable' Or intermediate, with low and varying Signal Strength and Signal Quality, accompanied by picture freezes or no picture. Nothing like on London Live.

I cannot see how picture formatting/processing can have anything to do with this, if it were I would never be able to see a steady picture on this channel, which I do, briefly, now and then. (Seemingly more in advert breaks, but not sure)

I tried stepping through all 106 channels... I just don't believe 'The Professionals' was ever filmed in widescreen; 'Accurist timepieces' - Gee!; 'Blippi' on Ketchup TV was just a blank screen and no sound, so who knows. I got a bit fed up around Kiss Chat & Date, so I gave up.

I did find one transmitting channel that was behaving the same as London Live. That was Sony Movie Classic on LCN 51. Which is also transmitted on LTVmux! So there's the culprit, but how and why?
Last edited by XFool on April 14th, 2021, 5:26 pm, edited 1 time in total.

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#404258

Postby XFool » April 14th, 2021, 5:22 pm

"You're gonna need a bigger filter!"

Since the aerial guy yesterday, I found this morning that VHF FM will still have problems.

Tuner signal strength indicator has 5 bars. SHOULD be max 5 bars for R4 from Wrotham (and most other stations). This morning R4 dropped to 4 bars with some hiss on stereo (ought to be no hiss at that level, surely?). Other stations also down/hissy. Going by yesterday, when aerial guy was working, I decided to pull aerial cable out of wall socket. Sure enough signal dropped to 2/3 bars. No surprise there. BUT, quality of signal improved - no hiss on stero on R4.

This confirms what I had suspected, quality of signal is being affected by more than signal strength (had earlier wondered if signal indicator on my 1980s era tuner was out of adjustment). That intermittant: stero hiss, loss of signal strength, not to mention annoying 'gas flow in pipe' sound on R4 are being caused by external interference. Even more amusing, just 'showing' the plug at the end of the tuner aerial lead the FM wall socket (~2/3 inches), without plugging in, cuts R4 off entirely!

Looks like I could use a pretty powerful Band II bandpass filter inline with the aerial lead to the tuner.

As an aside, have you tried AM recently? I did, in truth I was pretty shocked. OK, "Who listens to AM these days?" I hear you say. Well not me but: ADSL 2+, which I still use, uses 'tones' that are in the Medium Wave band.

Is terrestrial RF entertainment broadcasting coming to the end of its shelf life? Its got too much 'competition' in today's environment.

AleisterCrowley
Lemon Half
Posts: 6381
Joined: November 4th, 2016, 11:35 am
Has thanked: 1880 times
Been thanked: 2026 times

Re: Spectrum Analyzer?

#404286

Postby AleisterCrowley » April 14th, 2021, 6:51 pm

I think terrestrial analog radio is on the way out - sadly, as DAB is power hungryand needs a good signal
With ADSL and LED bulbs (and other SMPS devices, and power line transmission etc) the HF spectrum is mush in urban areas - I know of radio hams pulling (what's left of) their hair out...

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#404292

Postby XFool » April 14th, 2021, 7:10 pm

AleisterCrowley wrote:I think terrestrial analog radio is on the way out - sadly, as DAB is power hungryand needs a good signal
With ADSL and LED bulbs (and other SMPS devices, and power line transmission etc) the HF spectrum is mush in urban areas - I know of radio hams pulling (what's left of) their hair out...

Yes. Not just the HF spectrum. I still remember one night in Birmingham as a teenager in the 1960s when I went out in the back garden, to fetch in something from the washing line I think, I happened to look upwards: WTF!

It was the Milky Way. I knew what it was, I'd just never seen it before. Nowadays in London at night, if I see a small light in the sky it's either an aircraft, I must be suffering migraine attacks again, or it's the little green men arriving. :shock:

servodude
Lemon Half
Posts: 8271
Joined: November 8th, 2016, 5:56 am
Has thanked: 4435 times
Been thanked: 3564 times

Re: Spectrum Analyzer?

#404377

Postby servodude » April 15th, 2021, 1:08 am

XFool wrote:I did find one transmitting channel that was behaving the same as London Live. That was Sony Movie Classic on LCN 51. Which is also transmitted on LTVmux! So there's the culprit, but how and why?


Perhaps the Error Rate reports for LTVMux are bogus? (given it's a different transmission mode from the channels)
- having a different decoder in the same spot (like a set top box) would allow you to A/B the performance - and isolate whether it might be down to the TV

XFool wrote:As an aside, have you tried AM recently?


yeah - it's a great cheap way to check for EMI ;)

- sd

pochisoldi
Lemon Slice
Posts: 941
Joined: November 4th, 2016, 11:33 am
Has thanked: 31 times
Been thanked: 462 times

Re: Spectrum Analyzer?

#404397

Postby pochisoldi » April 15th, 2021, 8:15 am

From digging around, it turns out that the LTVmux uses FEC 3/4 and the PSO muxes use FEC 2/3.

So, taking a common denominator of 12 transmitted bits, LTV sends out 9 bits of payload with 3 error correction bits.

PSO muxes send out 8 bits of payload with 4 error correction bits.

The PSO raw bitstream is inherently more robust than the LTV one.

Combine this with the difference in transmitted power, and the reach of the LTVmux is going to be lower - and simply using a higher gain aerial to pick up LTV won't help if the TV tuner is then overloaded by the PSO signals - tuners can only accept a certain range of signal strength. (Think of a car driver's eyes, blinded by low sun that can't make out oncoming traffic)

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: Spectrum Analyzer?

#404542

Postby XFool » April 15th, 2021, 4:12 pm

pochisoldi wrote:From digging around, it turns out that the LTVmux uses FEC 3/4 and the PSO muxes use FEC 2/3.

So, taking a common denominator of 12 transmitted bits, LTV sends out 9 bits of payload with 3 error correction bits.

PSO muxes send out 8 bits of payload with 4 error correction bits.

The PSO raw bitstream is inherently more robust than the LTV one.

Combine this with the difference in transmitted power, and the reach of the LTVmux is going to be lower - and simply using a higher gain aerial to pick up LTV won't help if the TV tuner is then overloaded by the PSO signals - tuners can only accept a certain range of signal strength. (Think of a car driver's eyes, blinded by low sun that can't make out oncoming traffic)

This sounds more promising. However: https://www.ofcom.org.uk/spectrum/information/transmitter-frequency

LTVMux	Multiplex L  Local TV Multiplex	Comux UK    DVB-T:   QPSK, code rate 3/4, 8K FFT
COM4 Multiplex A SDN SDN DVB-T: 64 QAM, code rate 3/4, 8K FFT
COM5 Multiplex C Arqiva A Arqiva DVB-T: 64-QAM, code rate 3/4, 8K FFT

I don't know if "code rate" here is the same as "FEC" but, if it is, then commercial Muxs use the same rate as the LTVmux with no problem. Though they are transmitted from Crystal Palace at 200 kW like the PS Muxs.

https://en.wikipedia.org/wiki/Crystal_Palace_transmitting_station#After_switchover

I notice LTVmux uses QPSK modulation as opposed to QAM on all the main Muxs. I don't know if this is relevant.

servodude
Lemon Half
Posts: 8271
Joined: November 8th, 2016, 5:56 am
Has thanked: 4435 times
Been thanked: 3564 times

Re: Spectrum Analyzer?

#404634

Postby servodude » April 16th, 2021, 12:50 am

XFool wrote:I don't know if "code rate" here is the same as "FEC" but, if it is, then commercial Muxs use the same rate as the LTVmux with no problem. Though they are transmitted from Crystal Palace at 200 kW like the PS Muxs.


yes it is
code rate is the ratio of information/data to bytes: 3/4 means 75% of a frame is data and 25% error correction
- that should be a pretty good ratio

XFool wrote:I notice LTVmux uses QPSK modulation as opposed to QAM on all the main Muxs. I don't know if this is relevant.


QPSK is simpler - each quadrant of the phase difference between signal and carrier represents two bits
QAM breaks the space down further in to a constellation of points (located by phase and amplitude)
e.g 16-QAM would represent 4 bits as 16 points, 64-QAM 6bits as 64points

So on paper QPSK should be easier to decode and more tolerant of noise/interference but with a lower data capacity (for the same clocking rates)
- but that doesn't mean your TV front end handles it as well as the other modulation methods ;)

But unless the reported 0 error rate is bogus - it points to the information in the stream being sub par

- sd


Return to “Technology - Computers, TV, Phones etc.”

Who is online

Users browsing this forum: No registered users and 13 guests