Why are old spectrum analyzers so expensive?

[hendorog]

New Group: I'll ignore all the modern little handheld units with a limited display as these are only useful for very casual/portable use. You either need this size/portability or you don't.

But I like my little handheld headless SA.

And worth noting that your BB60C can do many things the old stuff can't - stuff which is relevant nowadays.

I'm very happy with my SA124B too. The bandwidth limitations of the Rigol and Siglent would have ruled them out for me if they were around then. The USB devices often get overlooked because they don't have a screen - they are a bargain for those who can see past that

[joeqsmith]

And worth noting that your BB60C can do many things the old stuff can't - stuff which is relevant nowadays.

I'm very happy with my SA124B too. The bandwidth limitations of the Rigol and Siglent would have ruled them out for me if they were around then. The USB devices often get overlooked because they don't have a screen - they are a bargain for those who can see past that

I really don't know what Siglent and Rigol offers.  The person posting all the Siglent data stopped once I started posting the QAM demodulation.  I would guess that most of the systems could decode that Extech meter I showed.  I played with the HDSDR software which is really slick.  I am not sure why (pure luck?) but that software has yet to hang with the BB60C.  I don't think I have ever seen it hang running swept mode.     

The whole headless thing has not really bothered me.  Often I run my equipment from a PC anyway and having the hooks for custom software is a must.  With the BB60C, using the USB3, I can move a fair amount of data fairly quickly into the PC.  I tried some tests without using the decimation at the highest sample rates and was able to keep up with LabVIEW. 

Show us the Siglents and Rigols doing some basic demodulation. 

Here's the BB60C decoding a very basic 2FSK, 2400 BAUD with a 915MHz carrier.

https://youtu.be/f1UtLpSZ2W0?t=1756

   

[G0HZU]

New Group: I'll ignore all the modern little handheld units with a limited display as these are only useful for very casual/portable use. You either need this size/portability or you don't.

But I like my little handheld headless SA.

By handheld I mean the new and cheap spectrum analysers with a built in (limited) display that can be picked up and viewed as you walk around a test site. Think more in terms of the RF Explorer than the Signalhound USB + PC products.

The Signalhound USB based stuff probably deserves its own category along with the Tek RSA306. They can't compete with the best of the old school analysers (or even the mid range analysers) for downconverter performance but they do have that huge real time bandwidth in the BB60C and all of the toys associated with the post processing in the PC. Powerful stuff indeed. The original SH44 with its narrow real time BW would be a good choice for radio hams and maybe also for EMC work but the unconventional design in terms of how it achieves its image/spurious rejection means that it will have a few holes in its performance/integrity as a spectrum analyser. Probably fine for casual use but risky for some types of use.

[joeqsmith]

I rate my old SAs at the bottom of the heap, or maybe floor sweepings if there is a category for that.  I don't think there is anything that my old Tektronix SA does better than my Signal Hound except using more power and shelf space.  The handhelds you mention  may even have better specs.  The last time I used the HP8569A was when we were playing with those el cheapo oscillators. 

https://www.eevblog.com/forum/projects/el-cheapo-oscillator/

[Fraser]

Before getting overly excited about USB spectrum analyzers, remember that a full size spectrum analyzer is built as a 'system'. That is to say, it is a combination of RF, analogue and digital technology in a box that is tested for performance and EM radiation. Any serious internal spurii are searched to source and addressed. This can be simple RF screening tape, additional filter chokes or more sophisticated digital electronics taming techniques such as lowering clock speeds and rounding the corners off of the digital square wave signals to reduce harmonic content. Full RF enclosure (biscuit tin) techniques are also employed where appropriate. Modern PC's run at amazing bus speeds and messing with the quality of the digital data stream to lower harmonic content is not normally a concern for the manufacture.

I have lost count of the number of times I have helped someone who is struggling with unexpected spurii whilst looking at the sensitive areas of an RF circuit with a spectrum analyzer or whilst listening to weird noises coming out of a receiver under test. I walk over to the adjacent PC, TV, SMPSU or even fluorescent light, and switch them off. Like magic, the unexpected signals all disappear. Basic stuff, yes, but all techs have to learn about such sometime and the first time can be perplexing for them to say the least !

If anyone decides to carry out diagnostics on RF converters or receivers, they should think carefully about whether having a PC running nearby is really such a good idea. Before anyone shouts European EMC directives, I know, I know, but once you connect cables and other equipment to a PC you can get all manner of weird emissions. USB3 was infamous for its horrendous RFI production when a cable was connected to the port.

USB based SA's do have their place, but, like all equipment, they have limitations, some serious when used in EMC work. Most can, however, be addressed by a knowledgeable tech when they are understood. Placing laptops and desktop PC's anywhere near a sensitive receiving system is asking for trouble.

One of the worst RFI offenders I found in my radio kit was an AOR SDU5500 panoramic display adapter. The AR5000A receiver was picking up all manner of spurii from the SDU. The SDU5500 radiated from the LCD and its backlight inverter. It was impossible to use when the AR5000 was connected to any form of indoor antenna. AOR expected all users to use an antenna outside the house noise field so obviously did not address the RFI emissions from their SDU.

Finally, it is a wise tech the who tests his spectrum analyzer for internal spurri PLUS any spurii that the input circuit presents to the DUT. It is not a great anlayzer that actually injects spurii or local oscillator/ 1st Mixer products into the device that is connected to its input. Yes, such crappy SA's do exist !

Fraser

[G0HZU]

I rate my old SAs at the bottom of the heap, or maybe floor sweepings if there is a category for that.  I don't think there is anything that my old Tektronix SA does better than my Signal Hound except using more power and shelf space.  The handhelds you mention  may even have better specs.  The last time I used the HP8569A was when we were playing with those el cheapo oscillators. 

https://www.eevblog.com/forum/projects/el-cheapo-oscillator/

I have fond memories of using the HP8569 back in the 1980s in my youth but it was never a high performance analyser I'm not sure which old Tek analyser you have but I don't rate any spectrum analyser from Tektronix. I've used a few over the years and hated them all. Their RTSA products are quite good though.

I guess it's unrealistic to compare the BB60C against the best of the 1980s analysers but you did link to the cheapo oscillator thread so I'll give an example where an old HP8566B can still shine today. The best of today's lab grade analysers will be much better but you can buy an HP8566B for about £1000 today if you are patient. I paid £1000 for this one over 5 years ago as an ex rental sell off.

A while back I made a suite of crude/cheapo oscillators to test my HP8566B for phase noise here at home because I didn't have a low phase noise source up at many GHz here at home back then. So I designed some crude free running oscillators using a printed resonator on Rogers 4003C PCB material with a target phase noise of -120dBc/Hz at 100kHz offset as this is close to the phase noise of the HP8566B here. In the youtube video below you can see me test one of them for stability. This was a 4.4GHz oscillator and it only drifted a few kHz in the test.



From memory the loaded Q was just over 20 and I designed it on an EM simulator (Sonnet EM) to make sure the loop response was correct for optimal phase noise. A very crude analysis using Leeson's equation predicted about -120dBc/Hz at 100kHz offset at 4.4GHz for this oscillator design. The (typical) phase noise of the old HP8566B isn't quite this good but it did at least allow me to measure the phase noise to within a few dB of this at 100kHz offset at 4.4GHz. The oscillator noise is slightly lower than the HP8566B here.

On a decent E5052A SSA at work I think this little oscillator managed -119dBc/Hz at 100kHz offset.

What can the BB60C manage here? I'm going to guess it will be in the ballpark of -95dBc/Hz at 100kHz offset at 4.4GHz. Maybe >20dB worse than this 35 year old HP8566? So it would be fairly useless here as a tool to evaluate this oscillator. I could list other stuff the BB60C will be poor at but I think it is really aimed at people who want that big real time bandwidth to capture/process modern wideband signals. The HP8566B obviously can't compete here. It really is a case of horses for courses

[G0HZU]

I managed to find this old test oscillator and I just measured it again on the HP8566B and you can see below that the combined phase noise of the oscillator and the HP8566B is about -117dBc/Hz at 100kHz offset. I think I've probably messed with this oscillator a few times since I made that youtube video but the performance is about the same.

The true PN of the HP8566B is probably close to -120dBc/Hz here which is a pretty good result for such an old analyser. My Advantest TR4172 analyser manages -128dBc/Hz phase noise at 100kHz offset at 1GHz but it only covers up to 1800MHz. But this is one area where the big old analysers perform well because of the YIG oscillator based LO system.

[joeqsmith]

The Tektronix I have is a 496P. I offered it to a local radio club but there was little interest. 

Most of the oscillators I play with can be measured with a scope.  This was video where I compared a 1GHz oscillator using an old Wavecrest, my old LeCroy and then post processing the data using Labview.     

https://youtu.be/Bg45FuoeHZk?t=754

I could list other stuff the BB60C will be poor at but I think it is really aimed at people who want that big real time bandwidth to capture/process modern wideband signals. The HP8566B obviously can't compete here. It really is a case of horses for courses

As you suggest, at the price it allows a hobbyist like myself to have some fun.  The data sheet for the BB60C shows -93dBc/Hz @ 1GHz.  I have no way to evaluate it at the frequencies you are interested in.  As you suggest, this is not going to be the tool of choice for looking at phase noise.  I think the Siglent had a wider dynamic range and lower phase noise.  I really don't know what they offer to the digital crowd. 

[G0HZU]

As you suggest, at the price it allows a hobbyist like myself to have some fun.

I'm sure you will agree that it could also be used to do some serious signal analysis and it also offers the ability to scan very rapidly. So I suspect that this product will have very wide appeal professionally too.

I guess the point I'm trying to make is that a lot of the old mid range HP/Tek/Advantest/Anritsu analysers from yesteryear really are obsolete now and offer limited functionality to a modern user. The exception to this will possibly be for ham/CB radio servicing/checking and for old school RF design of fairly simple devices. There's so much choice out there today I can't see the appeal of them now and I don't understand why people pay big bucks for them. For casual use I would much rather have a Siglent analyser or the BB60C than a lot of the old mid range analysers. But for serious RF design I would still reach for a decent old school lab analyser for most classic measurements.

The very best $$$ analysers from this era are still very desirable for RF design and offer remarkable value for money today. My HP8568B cost me £125 about 10 yrs ago, the TR4172 was just £400 prior to this. I bought a faulty TR4172 a couple of years ago for £250 and it was easy to repair at no cost. But these fabulous old analysers can't offer the wideband SDR tools/toys of the modern digital IF type analysers of today like your BB60C.

[Bicurico]

"I can't see the appeal of them now and I don't understand why people pay big bucks for them."

I tried to explain that in my post: the reason, amongst other reasons, is simply the appeal of owning a premium lab equipment which in its days would cost a fortune.

Call me crazy, but for sure I am not the only one, who would love to own such a HP, Agilent or R&S premium spectrum analyser in working condition. It's a collectors item. Some like vintage cars, some like vintage lab equipment.

To give another example of a field I am more comfortable in: TV field meters. There is a device called the Kathrein MSK33 (http://www.radiomuseum.org/r/kathrein_sat_tv_fm_messempfaenger_msk_33g_bn_208302.html), which was in it's days (1999) the premium TV field meter, covering TV, CATV and SAT bands with options ranging from MPEG2 decoder, DVB-T board or return path measurement. Those devices can still be bought at eBay and their prices, after more than 15 years(!) sometimes still exceed 500 Euro. If you are lucky and patient, you can get one for much less.

Is it worth it? I am not sure. For 1000 Euro you can purchase a brand new Rover Instruments entry level field meter that does MPEG2/MPEG4, DVB-S/S2/C/T and all new measurements with realtime spectrum. Much better than the Kathrein MSK33, except in one thing: the Kathrein MSK33 has the best constellation diagram implementation I have seen so far.

AND: The Kathrein MSK33 is like a brick, solid and robust. It works fine, does its job and can be fully remote controlled through RS232. It even offers some kind of terminal remote interface! This is a classic device, which was marketed as an OEM by no other than R&S! Now look at the attached picture! This is the SAME device, but with R&S logos printed on it. It did not sell, but it makes my point!

ANYTHING from R&S is sold for an expensive price. It is rare, it is highly collectable and in this case, reason does not matter. Even old R&S catalogs are sold on eBay for ridiculous prices.

Regards,
Vitor

[G0HZU]

"I can't see the appeal of them now and I don't understand why people pay big bucks for them."

I tried to explain that in my post: the reason, amongst other reasons, is simply the appeal of owning a premium lab equipment which in its days would cost a fortune.

Call me crazy, but for sure I am not the only one, who would love to own such a HP, Agilent or R&S premium spectrum analyser in working condition. It's a collectors item. Some like vintage cars, some like vintage lab equipment.

I think we agree on the premium analysers having a special appeal. I have several of the very best analysers here that money could buy in the 1980s. Eg an HP8568B, HP8566B and a couple of TR4172 analysers from Advantest. I was lucky enough to be a daily user of the HP8568B and the TR4172 back from 1990 onwards in our design labs. I now own the very same analysers I was using back then.

Although it isn't my first choice analyser today, I still get a tingle when I turn on the HP8568B because I can remember how special this analyser was back then. The company couldn't afford to buy it so it was donated to us on a project by the customer and was effectively paid for by the UK taxpayer. This analyser was held in very high regard by all the top engineers at the company so I felt very privileged to be allowed to use it for design work back then. A year or so later the customer donated the TR4172 to the company and this analyser was outrageously expensive and it could outperform the HP8568B in several key areas. To now have both of them here at home is very special.

The analysers that I think have limited appeal are the ones in the performance classes way below the HP8568B. I'm referring to the HP8590/1/x range, anything from Tek and pretty much anything from Anritsu or Advantest (apart from the TR4172).

[joeqsmith]

As you suggest, at the price it allows a hobbyist like myself to have some fun.

I'm sure you will agree that it could also be used to do some serious signal analysis and it also offers the ability to scan very rapidly. So I suspect that this product will have very wide appeal professionally too.

I guess the point I'm trying to make is that a lot of the old mid range HP/Tek/Advantest/Anritsu analysers from yesteryear really are obsolete now and offer limited functionality to a modern user. The exception to this will possibly be for ham/CB radio servicing/checking and for old school RF design of fairly simple devices. There's so much choice out there today I can't see the appeal of them now and I don't understand why people pay big bucks for them. For casual use I would much rather have a Siglent analyser or the BB60C than a lot of the old mid range analysers. But for serious RF design I would still reach for a decent old school lab analyser for most classic measurements.

The very best $$$ analysers from this era are still very desirable for RF design and offer remarkable value for money today. My HP8568B cost me £125 about 10 yrs ago, the TR4172 was just £400 prior to this. I bought a faulty TR4172 a couple of years ago for £250 and it was easy to repair at no cost. But these fabulous old analysers can't offer the wideband SDR tools/toys of the modern digital IF type analysers of today like your BB60C.

Actually, I think my 496 has better phase noise than the BB60C, that's not saying much! I was surprised the local club had no interest.   

http://www.ebay.com/itm/Tektronix-496-Spectrum-Analyzer-/271899015273?hash=item3f4e71b869:g:Hn0AAOSwpdpVefx1
 

I think what the BB60C can do for the price is pretty good and with time I suspect their software will continue to improve. 

[mmagin]

Maybe among sub-3GHz spectrum analyzers.   

I spent about $2k a year and a half ago on a used HP 8562A.  It's certainly doesn't have amazing phase noise specs or minimum resolution bandwidth, but it does go to 22 GHz, and is generally a reliable piece of equipment.  As far as I can tell, I'd be looking at $13k+ (MSRP anyway) on the new market to get something which can do that frequency range with decent specifications.

[Electro Fan]

How come the price gap between an old crt analyzer on eBay and a new Rigol our Siglent lower end signal analyzer is so narrow?

There are a lot (almost all of the) users here with more experience, knowledge, and insight than me on EE stuff in general and Spectrum Analyzers in particular, but here are a few thoughts.

As background, a while back I found a bit of a deal on a HP8561E.  At the time Rigol looked like a good alternative and now Siglent looks good too.

My sense is that a significant tradeoff between a vintage HP and the much newer Rigol/Siglent models is bandwidth.  If you don't need the bandwidth of the HP vintage units, I'd think seriously about going with a Rigol or Siglent.  Secondarily, I think there is also an advantage (in addition to bandwidth) in favor of the vintage HP's with respect to some performance specs, but I doubt the performance difference will be noticeable for most applications.  So, what I think drives the market price proximity is mostly banwidth, followed by some combination of performance and nostalgia.  However, in the end (or as the first decision point) either you need a certain amount of bandwidth or you don't.  If you can live with the bandwidth and price of the Rigols and Siglents you will get not only decent performance but many more features and much greater (modern software driven) functionality.  You will also get a warranty.  As for which will last longer, a vintage HP purchased in 2016 or a new Rigol or new Siglent purchased in 2016, I don't know.  It might depend on how well the vintage HP was maintained and how lucky we are with any of the models:  HP, Rigol, or Siglent. 

In my case, I don't use the SA a real lot and I don't do a lot of serious stuff with it, but when I need it or have an idea about something I'd like to checkout (usually wifi related or ham related), it is a very enjoyable tool to use.

Net, net: unless you have a specific stringent performance specification requirement, start by determining your required bandwidth; if the Siglent or Rigol meet the bandwidth requirement and fit your budget, go with the new technology.  If you need more bandwidth than you can afford with Siglent or Rigol and you can find a good vintage HP that gives you the needed bandwidth and fits your budget, go with the HP.  Either way, a Spectrum Analyzer is a pretty cool piece of test equipment that can help you measure and learn. 

PS, in my case I was attracted by 6 GHz bandwidth, pretty low noise specs, and very narrow RBW (1 Hz), plus some nostalgia and general confidence in HP.  FWIW, after experiencing the fairly long sweep times that come with very narrow RBW and the additional but only modestly practical information derived from very narrow RBW, I probably put too much emphasis on the RBW.  I don't think I will ever use 1 Hz RBW, and to be square most of the time I'm above 1 kHz and often much higher on the RBW.  (But if I hadn't seen it for myself, I'd always be thinking "wow, I should have held out for 1 Hz RBW.")  For anyone thinking about a SA, if you need narrow RBW and the sweep times are acceptable, go for it, but be careful about over paying for RBW you don't need.  Bandwidth on the other hand, is a nice capability and I'm very happy that I held out for 6 GHz (still more bandwidth would have been nicer but for my purposes I don't know what I would have done with it).  More FWIW, I get a much bigger kick out of looking at a 40 picosecond rise time on a Tektronix 7904 than I get out of looking at 1 Hz RBW on the HP8561E; go figure... YMMV

[hendorog]

Net, net: unless you have a specific stringent performance specification requirement, start by determining your required bandwidth; if the Siglent or Rigol meet the bandwidth requirement and fit your budget, go with the new technology.  If you need more bandwidth than you can afford with Siglent or Rigol and you can find a good vintage HP that gives you the needed bandwidth and fits your budget, go with the HP.  Either way, a Spectrum Analyzer is a pretty cool piece of test equipment that can help you measure and learn. 

Consider the Signalhound SA series along with Rigol and Siglent. You get more bang for your buck with SH, more flexibility, similar phase noise to Siglent (well ahead of the Rigol), portability and more RF bandwidth. Downsides are you need a PC, and the unorthodox LO.

Consider the Signalhound BB60C and Tek RSA306 if you want big realtime bandwidth.

[David Hess]

What's a used 24GHz oscilloscope going for now days?  I gave away my old HP 24GHz SA over the summer.  Checked eBay to get some idea and prices first. Many of the SAs I looked at seemed to have a fair price.

$400 for my 14 GHz digital sampling oscilloscope.

[Electro Fan]

Net, net: unless you have a specific stringent performance specification requirement, start by determining your required bandwidth; if the Siglent or Rigol meet the bandwidth requirement and fit your budget, go with the new technology.  If you need more bandwidth than you can afford with Siglent or Rigol and you can find a good vintage HP that gives you the needed bandwidth and fits your budget, go with the HP.  Either way, a Spectrum Analyzer is a pretty cool piece of test equipment that can help you measure and learn. 

Consider the Signalhound SA series along with Rigol and Siglent. You get more bang for your buck with SH, more flexibility, similar phase noise to Siglent (well ahead of the Rigol), portability and more RF bandwidth. Downsides are you need a PC, and the unorthodox LO.

Consider the Signalhound BB60C and Tek RSA306 if you want big realtime bandwidth.

hendorog, sorry I previously missed your comments (I scanned the thread too fast) when I made my post.

To be clear and square, I have very little knowledge about Signalhound; I just used Rigol and Siglent as a couple ways to compare newer technology to vintage HP SA's.  I have a hunch that folks like me who like nostalgic gear are probably into knobs and buttons on standalone gear, but I get the attraction of PC-based products and if they address the need for relatively higher bandwidth, it's good to have choices.  So if Signalhound is better than a Rigol or Siglent for someone's needs, that's cool.  We might be preaching to the same choir or a similar choir - I think bandwidth is a key decision-making criteria.

PS, it might be a classic "pick 2 of the 3"; "you can have bandwidth, price, buttons - but only 2 of the 3" 

[hendorog]

Net, net: unless you have a specific stringent performance specification requirement, start by determining your required bandwidth; if the Siglent or Rigol meet the bandwidth requirement and fit your budget, go with the new technology.  If you need more bandwidth than you can afford with Siglent or Rigol and you can find a good vintage HP that gives you the needed bandwidth and fits your budget, go with the HP.  Either way, a Spectrum Analyzer is a pretty cool piece of test equipment that can help you measure and learn. 

Consider the Signalhound SA series along with Rigol and Siglent. You get more bang for your buck with SH, more flexibility, similar phase noise to Siglent (well ahead of the Rigol), portability and more RF bandwidth. Downsides are you need a PC, and the unorthodox LO.

Consider the Signalhound BB60C and Tek RSA306 if you want big realtime bandwidth.

To be clear and square, I have very little knowledge about Signalhound; I just used Rigol and Siglent as a couple ways to compare newer technology to vintage HP SA's.  I have a hunch that folks like me who like nostalgic gear are probably into knobs and buttons on standalone gear, but I get the attraction of PC-based products and if they address the need for relative higher bandwidth, it's good to have choices.  So if Signalhound is better than a Rigol or Siglent for someone's needs, that's cool.  We might be preaching to the same or a similar choir - I think bandwidth is a key decision-making criteria.

Yep - it was just that you were talking new devices, price, and bandwidth. In that space SH just can't be left out as bandwidth at a low cost is one of their big advantages.

[Fraser]

This has been an interesting thread.

I was looking at micro metal lathes on eBay today and was surprised at how much people pay for very elderly Unimat lathes and even used more modern types. A modern SIEG C0 sells for around £300 delivered in the UK. People are paying more than that for a 1960's Unimat 1000 and a current auction has almost reached £300 on a very used modern SIEG C0.

What has this to do with old SA prices ? Well sometimes the secondary market for equipment appears illogical. That is just life. If two or more people get into a bidding war over something they want, things can get silly.

With respect to the lathes.... the venerable old Austrian made Unimat 1000 has a large following. Some are collectors, others had one, or wanted one when young. Each person has their reasons for buying stuff and a price that they are willing to pay. Some will pay way over the odds for what they desire.

The question that really needs to be asked is why is it that anyone should be troubled by someone else overpaying on an old piece of equipment. It is their prerogative to do so is it not ? I personally look for bargains such as equipment poorly described, faulty or just too exotic to be recognised for its usefulness 😄 I am not rich enough to overpay on kit.

A good question these days is how can a manufacturer make a half decent Spectrum analyser so cheaply. They have historically been a niche product that cost significant sums of money to manufacture. Modern electronics miniaturisation has enabled hobbyists to have access to acceptable performance within a reasonable budget. Be happy with that rather than wasting time wondering why other people still buy old SA's and pay, what seems to some, high prices. They have their reasons. Be happy for them getting the object of their desires 😄

On a personal note, I own Advantest SA's. No HP kit in my lab and yet I am still very satisfied with what I can achieve with my SA's. You do not always need the best to get the job done. Way back when, people had to design and repair  without access to an SA, they were just too exotic and expensive. These people got the job done by being top of the line techs who could innovate and make do with what they had available.

Enjoy your life and worry not what others do with their money. It will do you no good. It is sick that people spend £10K on Christmas decorations for a posh London party...... but that is their life, not mine.

Just my 2p worth

Fraser

[tggzzz]

This has been an interesting thread.

You do not always need the best to get the job done. Way back when, people had to design and repair  without access to an SA, they were just too exotic and expensive. These people got the job done by being top of the line techs who could innovate and make do with what they had available.

Very true. That is the inspiration for the gliding aphorism in my .sig.

[nctnico]

Way back when, people had to design and repair  without access to an SA, they were just too exotic and expensive. These people got the job done by being top of the line techs who could innovate and make do with what they had available.

And waste oodles of time in the process but that didn't matter because salaries where low. Also back then FCC/CE certification wasn't an issue like it is today. Nowadays you quickly need a spectrum analyser if you are serious about developing products which contain electronics.

[G0HZU]

This has been an interesting thread.

You do not always need the best to get the job done. Way back when, people had to design and repair  without access to an SA, they were just too exotic and expensive. These people got the job done by being top of the line techs who could innovate and make do with what they had available.

Very true. That is the inspiration for the gliding aphorism in my .sig.

In my case I absolutely did need the very best analysers available. My task back in the 1990s was to design the RF converter section of a suite of high performance receivers whose task it was to scan and digitise the RF spectrum from HF through several GHz and feed the output to a digital IF as it scanned. Back in those days the digital IF was a huge floorstanding rack system costing a lot of money. Today's equivalent digital IF DSP can be found in analysers like the Siglent at a tiny fraction of the cost, size and power.

The HP8568B was sometimes used in place of the DSP system as an IF and it needed to maintain a very good dynamic range whilst displaying a span of several MHz with a fluid display rate. Back then the benchmark for a decent analyser was 80dB (IP3) SFDR with a 30kHz RBW (span of several MHz typically) and it needed to have very low phase noise as it was also a design tool for verifying the low noise synthesisers in the receiver. The 2HI performance had to be very good too. The HP8568B could just about match the 80dB requirement but the TR4172 was better by several dB and its 2HI performance was a class above the HP8568B here and its phase noise at 100kHz offset was much better than the 8568. The synthesisers I designed had to have ultra low spurious levels down to very low signal levels and the HP8568B and the TR4172 were just about good enough to do this work. To use a basic analyser and try and innovate around this wasn't a realistic option back then.

The analysers we really wanted to use simply didn't exist in those days. We explored the limits of the HP8568B and eventually went over to the TR4172 for some of the design work. I think this particular TR4172 cost more than £60k when it was new in the mid/late 1980s but there were rumours that it cost even more because of the strong Yen back then. Luckily it was paid for with public money by the customer and you could have bought a very nice 4 bed house in the south east of the UK for that money back in those days

[joeqsmith]

If measuring phase noise in the GHz range is your hobby, the old HPs may be a better buy.  If you want to play with decoding your Extech wireless meter, I don't know what's on the new or used market that a hobbyist could afford.  Of course the Extech meter is not a real good example, so here are QPSK and QAM16 signals at 5GHz.



 

[joeqsmith]

The BB60C in swept, realtime and demodulation modes with a 5GHz QAM64 signal.

[joeqsmith]

BB60C in swept mode with a 5.5GHz QAM256 signal.  Well within the 27MHz bandwidth but Spike does not support demodulating it.  National Instruments offers a communications toolkit for free.  Maybe we can use this to decode the data......