开云体育

So, this may not be an appropriate forum for this question, but I had some questions on the TinySA.? As Spectrum Analyzers have been almost completely out of my price range, my familiarity with them is exceptionally limited.? However, doesn't the minimum resolution bandwidth of the TinySA (2.6kHz) act as a significant impediment to its functionality?? Based on my minimal understanding, such a wide setting would render it not useful for audio-level work, and severely limit the ability to use it for SSB work?? I think it would be quite useful for evaluating transmission spurs, which would be the other primary use for me on my workbench.

Don't get me wrong - I think it's an amazing bit of kit for the price!? I just would like some advice on if this would be a good bit of kit for someone who's not overly familiar with SA's and working around a particular design's shortcomings?

Thanks!

~Josh
W0ODJ

Josh,
You are right. As stated in the wiki the lower RBW is a limitation. For analysis of SSB signals you can use an RTL-SDR with appropriate? attenuation.

Josh,

I would agree with Shirley on the use of the TinySA.? I have been searching for a good SA since I got back into the hobby and have tried may "low cost" alternatives.? Everything from the RFExplorer to the Discovery 2 series.? Each of them has an area that does not meet the full requirements.? It would be a long email to contrast all of the options that I have gone through.

Currently, I have three devices that I go to for diagnostic use:

1 - NanoVNA for filter or SWR testing
2 - TinySA for harmonic and spur hunting
3 - RSP1A for close in audio frequency testing at RF frequencies. (this is just a more expensive RTL-SDR type of device with an upconverter built-in)

I have not found a tool that does it all.? At best I could replace the TinySA and RSP1A with a Rigol or Sigelent SA, and had I started there would be money ahead taking into account all of the failed purchases.

I am still learning more about how to use the TinySA to overcome the RBW issues.? That would include using AM to test amplifier linearity to get enough separation to measure IIMP3.? I still do not understand that process at this point.? There are also other ways that only require an oscilloscope (an oscilloscope is a given in the test equipment list and should be purchased before the SA choices).

So as others on the thread have stated you first need to decide what it is you want to measure (the use case) for the instrument.? That will then guide you to the best device for the job at a price that you can afford.

Above are my opinions, and maybe in error.? Please check for yourself as much as possible.
73
Evan
AC9TU

Thank you all!

My use case is... tinkering.? I am interested in things like hunting down transmitter spurs, and other "general use" - but wanted to make sure that I wasn't going to buy something that I'd quickly outgrow (or that I wouldn't be able to follow "learning exercises" from youtube, etc. on, to learn about SAs).? Granted, for the cost, it's hard to go wrong in any sense (and I'm going to still buy one), but I'll have to have another look at SDRs - it looks like that is going to be more useful for me right the moment.? I didn't get much use out of my (non-hf sampling) RTLSDR, and my RSP1A got sold because it didn't play well with linux (perhaps it's better now?).

Thank you so much!
~Josh
W0ODJ

The cool factor of just being able to wear a spectrum analyzer as a fashion item is worth the money!

TinySA earrings, anyone?

73

-Jim
NU0C

On Sat, 03 Oct 2020 16:09:35 -0700
"Joshua Wood" <WoodJRx@...> wrote:

Thank you all!

My use case is... tinkering.? I am interested in things like hunting down transmitter spurs, and other "general use" - but wanted to make sure that I wasn't going to buy something that I'd quickly outgrow (or that I wouldn't be able to follow "learning exercises" from youtube, etc. on, to learn about SAs).? Granted, for the cost, it's hard to go wrong in any sense (and I'm going to still buy one), but I'll have to have another look at SDRs - it looks like that is going to be more useful for me right the moment.? I didn't get much use out of my (non-hf sampling) RTLSDR, and my RSP1A got sold because it didn't play well with linux (perhaps it's better now?).

Thank you so much!
~Josh
W0ODJ

On Sat, Oct 3, 2020 at 09:43 AM, Evan Hand wrote:

Currently, I have three devices that I go to for diagnostic use:

1 - NanoVNA for filter or SWR testing
2 - TinySA for harmonic and spur hunting
3 - RSP1A for close in audio frequency testing at RF frequencies. (this is just a more expensive RTL-SDR type of device with an upconverter built-in)

Have you tried the Steve Andrews Spectrum Analyzer program for the RSP1A? Has markers, peak, averaging, different RBW etc. ? Available on the SDRplay Web site.

Roger

Roger,

Yes I did try the Spectrum Analyzer software and use it for it for the low RBW situations to like SSB purity testing ?It has issues when spanning large frequency spans. ?I found this when searching for 3rd harmonic of 20 meters (14MHz)

73
Evan
AC9TU

Roger,

The last message needs clarification.??

I did try the RSP1A with the Spectrum Analyzer software and found it to be very good for spans up to 10MHz.? When I tried to measure the 3rd harmonic of 14MHz (42MHz) displaying both the fundamental and that harmonic, there were steps in the noise floor and inconsistent level readings that did not maintain step attenuation of the input.? This situation did not happen for the lower span settings.? This is what led me to look for other solutions.

There are ways around the RSP1A issue, namely to measure the values at the lower spans by changing the center or start settings.? With the step in the noise floor, I was not confident of the relationship to the fundamental.? I also found the markers and math to be problematic.? They were hard to get to follow the signal levels when adjusting the step attenuator to validate measurements.? It was better with the last update that I used, but still not consistent and hard to set.

There is also a cost difference.? The TinySA is $50, the RSP1A is $100.? Unfortunately, it still takes both to do all of the purity tests to my way of thinking.? There maybe calibration steps of the RSP1A software that would solve the large span problem.? There may also be ways to measure SSB signals without needing less than 3kHz RBW.

For sure the TinySA will not get an upgrade in the RBW, as that is a hardware filter issue, at least at this point in time.? There may be ways to use digital comparison techniques to narrow down the bandwidth that I am not aware of.? As stated before, there may also be ways to adjust the calibration process of the RSP1A software.? Until then I will use both tools for their appropriate measurements.? The combined cost is still less than the lower cost Rigol SA model.

The above are my opinions based on my experiences.? They may be in error, so it is recommended that you validate if possible.
73
Evan
AC9TU

The step in the noise floor of the RSP1A (also the RSP1) is caused by the internal upconverter when measuring below 60MHz. The conversion loss is not compensated by extra amplification inside the MSI001 chip used in the RSP1A
My main problem with the RSP1 (I do not own a RSP1A) is you sometimes see spurs that do not exist and wide span scanning is slow.

What would be the maximum minimum RBW a possible tinySA upgrade must support to enable SSB analysis etc??

The rbw needed to assess intermodulation first depends on how you do it. ?The best way is two high quality signal (high power and totally isolated) generators 100khz apart, then the Tinysa will be fine.

However most amateur radio people do not have such kit, so two audio tones are put into the transmitter, usually as low and as high (eg 600hz and 2.3khz) to pass through the audio banwidth. So the output spectrum consists of signals 1.7khz apart. ?A resolution better than 25db is needed to assess intermod distortion of 30db or thereabouts (acceptable intermod distortion for most amateur kit). Tones can be a bit further apart but start to get near the edge of the audio passband in the radio.
Depending on how steep the rbw filter skirts are, a rbw of 500hz or less is desireable, otherwise the first intermodulation spur cannot be separated from the wanted signals to assess all but terrible intermodulation distortion ( usually due to overload of the amplifier). The skirts of the rbw cannot be too steep (on an old analogue sweep analyser) or the sweep speed has to be greatly reduced to avoid “ringing”.

Sorry if I have got the exact figures slightly incorrect, but it is a long time since I was ?familiar with these details. A rbw of 500hz or less would be ok, but I do not think it is realiseable. Most middle-aged spectrum analysers used digital fft in the final narrow resolution stages, I believe. Older ones, like my HP141 used crystal filters for the narrower rbw ranges. Alignment of the rbw filters for an old boatanchor was quite a task, as the skirts of each one had to be carefully adjusted. It was almost like aligning a Racal RA17 communications reciever (first synthesised valve set circa 1966).

I hope this helps. I think sdr is the way to go for ssb intermodulation measurements, but one can always wish......

steve L. G7PSZ

As an experiment.
HDSDR and SoftRock Ensemble RXTX at 7.1MHz LO, USB transmission at 7.1MHz with two tones, 300Hz and 2.5kHz
Using an RPS1 to verify the amount of intermodulation products

First with less then -50dBc first intermodulation products



With -40dBc first intermodulation product



with -30dBc intermodulation



and with -20dBc intermodulation



Its not possible to do a quantitative analysis but at least you see when it gets really bad.

If you offset the transmission frequency 4kHz from the LO (easy to do in HDSDR you can see the intermodulation product much better as they appear at the other side of the LO

Dear Erik,

Yes, your screenshots I think demonstrate what I was trying to say. I cannot set up my old boatanchor to show a screenshot of intermodulation products generated by a power amplifier within the ssb audio bandwidth (and I do not have the two-tone audio generator to produce 300hz and 2.7khz - yes I have, the Ipad has an app to do it). ?However, the tinysa will never be able to show it unless the two RF power generator method is used with frequencies at least 10khz apart.

I do not think anyone would expect the tinysa to do it. Actually very few radio enthusiasts would need it unless building or setting up a power amplifier, and then might never need to do it again in their lifetime. This excellent device should be used for what it can do very well, with a minute footprint, minute cost, maintenance and power consumption ( I had to deliberately avoid using the word tiny just there).

Steve L. G7PSZ

Dear Erik,

Looking again at your screenshots, I think the last one where you increased the separation to 4khz actually shows the two fundamentals, but not of equal amplitude. An intermod scan would show the two original frequencies of equal amplitude with a symetrical ?diminishing “comb” of intermod products either side of the fundamentals, spaced at the difference of the two fundamental frequencies (every 4khz).

steve L

Steve,

I did not increase the separation to 4kHz as the audio filters in HDSDR makes that impossible
As the HDSDR uses an i/q audio (192kHz bandwidth) to steer the output of the SoftRock when you offset TX away from the LO you can get intermodulation products at the other side of the LO

The LO of the softRock was 7.1MHz, TX was at 7.104MHz (USB) with the fundamentals at 7.104300MHz (300Hz) and 7.106500MHz (2500Mz) , The LO of the I/Q mixer was at 7.1MHz so what you see to the left of 7.1MHz is an intermodulation product originating in the I/Q mixer causing the mirror suppression to fail so it appears at the other side of the LO
The same (equally strong) products appear to the right of the LO around the 300Hz and 2500Hz signals but as the original signals are not visible at the other side of the LO these do not mask the intermodulation products at the other side of the LO.

I am one of those "few radio enthusiasts" who would do this. And you are correct in that I did not expect the TinySA to be able to do it. But I bought one anyway. :)

I have equipment that can do this but it takes up considerable more bench space than the tiny. My current project is evaluating substitute PA transistors to replace devices in a transceiver that have gone obsolete or dramatically increased in price. This is more what someone doing transmitter repair would need. The attached photo is the screen of my HP 8594E doing just such a measurement.

73

-Jim
NU0C

On Sun, 04 Oct 2020 09:18:19 -0700
"Stephen Laurence" <Gaslaurence@...> wrote:

Dear Erik,

Yes, your screenshots I think demonstrate what I was trying to say. I cannot set up my old boatanchor to show a screenshot of intermodulation products generated by a power amplifier within the ssb audio bandwidth (and I do not have the two-tone audio generator to produce 300hz and 2.7khz - yes I have, the Ipad has an app to do it). ?However, the tinysa will never be able to show it unless the two RF power generator method is used with frequencies at least 10khz apart.

I do not think anyone would expect the tinysa to do it. Actually very few radio enthusiasts would need it unless building or setting up a power amplifier, and then might never need to do it again in their lifetime. This excellent device should be used for what it can do very well, with a minute footprint, minute cost, maintenance and power consumption ( I had to deliberately avoid using the word tiny just there).

Steve L. G7PSZ

On Sun, Oct 4, 2020 at 11:20 AM, Jim Shorney wrote:

I am one of those "few radio enthusiasts" who would do this. And you are correct in that I did not expect the TinySA to be able to do it. But I bought one anyway. :)

I am also one of the "few radio enthusiasts" who would do this, and in fact, the first thing I did when I got my first tinySA was look at close-in performance.? I was hopeful, but it won't do SSB transmitter IMD tests.? But I have another SA that will, and the tinySA will do so much amazing stuff, including LOTS of things that my bigger, much more expensive SA won't do, or won't easily do, so the tinySA is well worth the price..? I currently own 3 tinySAs, and I'm thinking about getting another one to keep in the car.? (You never know when you might need a spectrum analyzer, you know.)

I believe that the close-in performance of the tinySA can be improved somewhat, but haven't had the time to do experiment with it, and even with the improvements I have in mind, it still won't do SSB transmitter IMD tests.? Nonetheless, a fine piece of equipment.? Everybody should have one (or several).

Hy other heavy iron is a gently used HP 3585B that I got for free. It is a sweet instrument but so large I had to put it in a different room from where the workbench is.

I've been playing with the TinySA too much lately when I should be doing other work.

73

-Jim
NU0C

On Sun, 04 Oct 2020 15:20:02 -0700
"Jim Allyn - N7JA" <jim@...> wrote:

On Sun, Oct 4, 2020 at 11:20 AM, Jim Shorney wrote:

I am one of those "few radio enthusiasts" who would do this. And you are
correct in that I did not expect the TinySA to be able to do it. But I
bought one anyway. :)

I am also one of the "few radio enthusiasts" who would do this, and in fact, the first thing I did when I got my first tinySA was look at close-in performance.? I was hopeful, but it won't do SSB transmitter IMD tests.? But I have another SA that will, and the tinySA will do so much amazing stuff, including LOTS of things that my bigger, much more expensive SA won't do, or won't easily do, so the tinySA is well worth the price..? I currently own 3 tinySAs, and I'm thinking about getting another one to keep in the car.? (You never know when you might need a spectrum analyzer, you know.)

The American Radio Relay League (United States ham organization) has a lab test procedure that uses 700 and 1900 tones for a two tone intermodulation test. Their procedure sets the SA RBW to 100 Hz, but that may be a bit overkill. As long as the skirt of the RBW filter is down 50 dB at 1.2 kHz, you could reliably measure -40 dB intermodulation. Don't know how much relief that gives on the filter performance requirement, but I am sure it is some.