Spectrum Analyzers, USB Dongle Software Defined Radios, FlexRadio (Ham Radio)

[Electro Fan]

Background:  I’ve been using test equipment as a way to learn about electronics at the hands-on level.  So far I have a few items:  some DMMs, a power supply, oscilloscope, function generator, breadboards and parts, etc.  I understand some of the basics and intermediate theory; I find that actually wiring things together helps the learning process.

I’m starting to think about exploring RF at the hands-on level and I'm considering purchasing a spectrum analyzer but they seem somewhat pricey (Rigol DS815, etc.)  In looking at spectrum analyzers I’ve come across Software Defined Radios based on PC software and USB dongles – these seem to provide some level of spectrum analyzer functionality.  Exploring these possibilities has caused me to trip across FlexRadio (Flex-1500).  This would be somewhat expensive ($700) but it opens the path to not only some spectrum analyzer functionality but also the ability to receive radio signals and potentially the ability to transmit signals (if I get a FCC Technician Class License in the U.S.).  So now we have “mission creep” but that seems to be par for the course with everything related to EE – it’s just a matter of speed and direction.
 ?
In looking at USB dongles and FlexRadio I came across Baofeng handheld radios but they seem to be less exciting with their FRS and GMRS modes; they are interesting as communications devices but they don’t seem to do much for technology learning.  But maybe at the price they provide some learning utility. 

I’ve found this to be pretty compelling:

http://www.flex-radio.com/Products.aspx?topic=F1.5k_features



FlexRadio looks like the intersection of RF, some spectrum analysis, operation of both receivers and transmitters, opportunity to learn about antennae, and the ability to communicate with people around the country/world.

So, I’m at a cross roads between USB dongles and something like FlexRadio, or maybe there is another RF and spectrum analyzer type path to pursue?

I’m up for any advice on the best path for exploring RF at the hands-on engineering/operations level.
 
Thanks, EF

[Rigby]

That's just a software defined radio that can transmit.  I wonder if the HackRF will outperform it in any way.  The HackRF will certainly cost less.  I don't know of a software GUI that's as good as the one in that video.  The open source community has apparently lost all interest in creating an open SDR front-end; all the decent ones are closed source.  GNU Radio is very powerful, but it is not enough.

[granz]

Keep in mind that those SDRs etc. are just sampling the input fast and then performing a FFT to produce a spectrum for viewing.  It's a different process than a true swept spectrum analyzer.  Just a heads up.

[Rigby]

Keep in mind that those SDRs etc. are just sampling the input fast and then performing a FFT to produce a spectrum for viewing.  It's a different process than a true swept spectrum analyzer.  Just a heads up.

Is that a Left Twix vs. Right Twix differentiation or something that matters?

"Sir, do you have a sec?"
"Sure."
"We’re the Right Twix factory making cookie layers with caramel covered chocolate for the Right side of the pack."
"Yup."
"And next door is the Left Twix factory and they make cookie layered with caramel covered in chocolate."
"No, cloaked in chocolate. Totally different process."

[KD0RC]

Electro Fan, another VERY cheap alternative to the Flex Radio is the SoftRock (see links, below).  I built one of the receive-only units and did exactly as you are suggesting - I learned a great deal about RF.  Plus, I got a pretty cool receiver out of it.  Note that this class of equipment cannot touch the specs of the Flex, but you can get in, build it yourself and get an idea of whether or not to go all-in on the Flex or other higher shelf technology. 

This will also give you a poor-man's spectrum analyzer for roughly 1 - 30 MHz, but will not have the features available on a 'real' SA like the Rigol (9 KHz - 1.5 GHz).  For poking around at off the air signals, however, it is a really good way to start.  I recently graduated to the Rigol DSA 815 TG.  I have always wanted a spectrum analyzer, but they were waaay out of my price range.  The 815 is around $1,500 US, making it irrestable... 

Go for the ham radio license - you won't regret it.

http://www.wb5rvz.com/sdr/

http://fivedash.com/

[Electro Fan]

Electro Fan, another VERY cheap alternative to the Flex Radio is the SoftRock (see links, below).  I built one of the receive-only units and did exactly as you are suggesting - I learned a great deal about RF.  Plus, I got a pretty cool receiver out of it.  Note that this class of equipment cannot touch the specs of the Flex, but you can get in, build it yourself and get an idea of whether or not to go all-in on the Flex or other higher shelf technology. 

This will also give you a poor-man's spectrum analyzer for roughly 1 - 30 MHz, but will not have the features available on a 'real' SA like the Rigol (9 KHz - 1.5 GHz).  For poking around at off the air signals, however, it is a really good way to start.  I recently graduated to the Rigol DSA 815 TG.  I have always wanted a spectrum analyzer, but they were waaay out of my price range.  The 815 is around $1,500 US, making it irrestable... 

Go for the ham radio license - you won't regret it.

http://www.wb5rvz.com/sdr/

http://fivedash.com/

Hi KDORC,

Thanks for the info and encouragement.

Based on your post I did some searching.  Looks like SoftRock is indeed a very good possible starting point.

I'm now looking for more info to compare FlexRadio and SoftRock.  So far I've found this:
http://www.sdrtransceiver.com/

You have me thinking harder more about the ham radio license.  Are you a ham with a license?  (Which license?) What have you found most interesting with the 815TG?

If I'm going to get a license, maybe it makes sense to go straight for a SoftRock transceiver?

I need to get up to speed on the various meters/bands, etc.  Any recommendations welcome...

Thanks again, EF

[GreyWoolfe]

+1 on the ham license.  Just don't stop at Technician, you only really get some 10 meter privileges.  At the very least, go for your General.  As a General, you have portions of all the ham HF bands.  Much more funner.  By the way, if you live in central Florida, I do testing every month except December as the VE Liaison for the Daytona Beach Amateur Radio Association.

Tom, NW0LF

[granz]

Keep in mind that those SDRs etc. are just sampling the input fast and then performing a FFT to produce a spectrum for viewing.  It's a different process than a true swept spectrum analyzer.  Just a heads up.

Is that a Left Twix vs. Right Twix differentiation or something that matters?

"Sir, do you have a sec?"
"Sure."
"We’re the Right Twix factory making cookie layers with caramel covered chocolate for the Right side of the pack."
"Yup."
"And next door is the Left Twix factory and they make cookie layered with caramel covered in chocolate."
"No, cloaked in chocolate. Totally different process."

LOL, that's a good commercial.  In all seriousness though, the two techniques are different, and each has its advantages and disadvantages.  The DSA815 is a swept analyzer, where-as the SDRs will give a FFT spectrum.  Some of the high-end Agilent SAs will let the user choose the technique depending on the application.

Have a look at this for a comparison (there is a table further down):

http://www.mpdigest.com/issue/Articles/2009/Jan/agilent/Default.asp

Personally, I think nothing really beats having a dedicated SA (such as the DSA815 or anything else that's affordable).  In general, for the same price, you'll get much higher frequency swept-SA vs. an FFT analyzer.

[Electro Fan]

+1 on the ham license.  Just don't stop at Technician, you only really get some 10 meter privileges.  At the very least, go for your General.  As a General, you have portions of all the ham HF bands.  Much more funner.  By the way, if you live in central Florida, I do testing every month except December as the VE Liaison for the Daytona Beach Amateur Radio Association.

Tom, NW0LF

Thanks - any chance you could summarize what the limitations (of being confined to 10 meter privileges) would mean from a practical experience?  What would you/could you experience at the General level (in what meter ranges) that you would miss with the Technician license at 10 meters?  Thanks again

[w2aew]

+1 on the ham license.  Just don't stop at Technician, you only really get some 10 meter privileges.  At the very least, go for your General.  As a General, you have portions of all the ham HF bands.  Much more funner.  By the way, if you live in central Florida, I do testing every month except December as the VE Liaison for the Daytona Beach Amateur Radio Association.

Tom, NW0LF

Thanks - any chance you could summarize what the limitations (of being confined to 10 meter privileges) would mean from a practical experience?  What would you/could you experience at the General level (in what meter ranges) that you would miss with the Technician license at 10 meters?  Thanks again

Propagation on 10 meters is much more dependent on ideal atmospheric conditions for long distance communications.  As the Solar indices and sunspot numbers drop, so too does the maximum usable frequency as far as global signal propagation is concerned. Most times, 10m completely closes up at night  Thus, bands such as 20, 17, and 15 meters provide more reliable long distance communications more often than on 10m. Lower bands such as 40 and 75/80m provide much more reliable short to medium distance communications around the clock.

[KD0RC]

Hi KDORC,
Thanks for the info and encouragement.
Based on your post I did some searching.  Looks like SoftRock is indeed a very good possible starting point.
I'm now looking for more info to compare FlexRadio and SoftRock.  So far I've found this:
http://www.sdrtransceiver.com/

You have me thinking harder more about the ham radio license.  Are you a ham with a license?  (Which license?) What have you found most interesting with the 815TG?
If I'm going to get a license, maybe it makes sense to go straight for a SoftRock transceiver?
I need to get up to speed on the various meters/bands, etc.  Any recommendations welcome...

Thanks again, EF

Yep, I am a ham.  My call is the same as my name here - KD0RC.  I have an Extra Class license, so I have all the privileges available.  I suggest getting a book on the subject so that you can see what is available to the various license classes, and also to see what it takes to attain those licenses.  Building a receiver or transciever while you are working on getting your license is a good way to stay motivated to keep going.  A good source of info is  http://www.arrl.org/

Buying a used receiver to use while you are learning is not a bad way to go either.  You can get a sense of operating procedure and start to figure out what to do for an antenna (not all receiving antennas make good transmitting antennas, but many do).

Comparing the SoftRock with a FlexRadio is like comparing a go-kart with a Lamborghini...  I bought my SoftRock receiver to learn from; it was fun to build and I play around with it all the time, but I don't use it in my regular ham communications.

The Technician license essentially gives you 10 Meters (28 - 28.5 MHz) and all higher bands.
General, Advanced and Extra give you all bands starting at 160 Meters (1.8 - 2 MHz).

It is not bad to start with the Technician and work your way up.  If you are more interested in HF (10 M and lower), then it might be good to go straight for the General.

The best part of the DSA 815 is how much I am learning from it.  I have wanted a spectrum analyzer for years, and this is the first one that I can afford and has a tracking generator that I can use to characterize filter response and also use to check my antennas.

Best of luck with this, ham radio has been a real fun thing for me for many years.

[ConKbot]

\
LOL, that's a good commercial.  In all seriousness though, the two techniques are different, and each has its advantages and disadvantages.  The DSA815 is a swept analyzer, where-as the SDRs will give a FFT spectrum. Some of the high-end Agilent SAs will let the user choose the technique depending on the application.

Have a look at this for a comparison (there is a table further down):

http://www.mpdigest.com/issue/Articles/2009/Jan/agilent/Default.asp

Personally, I think nothing really beats having a dedicated SA (such as the DSA815 or anything else that's affordable).  In general, for the same price, you'll get much higher frequency swept-SA vs. an FFT analyzer.

The FFT on signal analyzer is faster than a pure sweep in most applications. There are some that arent, but I forget what it was.  (I think I read it in an Agilent appnote)

I know that a having a signal analyzer with realtime/FFT made my life less miserable when I was doing some oscillator characterizations.  1 Hz RBW/VBW with a ~200Hz sweep made for ~5 minutes/sweep (dont even think about averaging...) on a swept SA, and 1.8s/sweep for the FFT analyzer, even with the sweep expanded upto 1kHz.

[Electro Fan]

+1 on the ham license.  Just don't stop at Technician, you only really get some 10 meter privileges.  At the very least, go for your General.  As a General, you have portions of all the ham HF bands.  Much more funner.  By the way, if you live in central Florida, I do testing every month except December as the VE Liaison for the Daytona Beach Amateur Radio Association.

Tom, NW0LF

Thanks - any chance you could summarize what the limitations (of being confined to 10 meter privileges) would mean from a practical experience?  What would you/could you experience at the General level (in what meter ranges) that you would miss with the Technician license at 10 meters?  Thanks again

Propagation on 10 meters is much more dependent on ideal atmospheric conditions for long distance communications.  As the Solar indices and sunspot numbers drop, so too does the maximum usable frequency as far as global signal propagation is concerned. Most times, 10m completely closes up at night  Thus, bands such as 20, 17, and 15 meters provide more reliable long distance communications more often than on 10m. Lower bands such as 40 and 75/80m provide much more reliable short to medium distance communications around the clock.

Thanks.  Not sure if I fully understand but overall it would seem that a key difference between the 10 meter band and the other bands that the 10 meter band is likely to cover shorter distances and provide less around the clock windows for communications - so the ability to reach people far away and at any time of day will be relatively limited.  It would be great to be able to communicate via Ham with people far away around the clock but in terms of experiencing the general experience (communicating via radio and learning the underlying technology), the 10 meter band with the Technician license would seem to provide at least a glimpse into the overall experience without having to pass more than the basic test.  Yes/No/Sort of?   Thanks again

[Electro Fan]

Hi KDORC,
Thanks for the info and encouragement.
Based on your post I did some searching.  Looks like SoftRock is indeed a very good possible starting point.
I'm now looking for more info to compare FlexRadio and SoftRock.  So far I've found this:
http://www.sdrtransceiver.com/

You have me thinking harder more about the ham radio license.  Are you a ham with a license?  (Which license?) What have you found most interesting with the 815TG?
If I'm going to get a license, maybe it makes sense to go straight for a SoftRock transceiver?
I need to get up to speed on the various meters/bands, etc.  Any recommendations welcome...

Thanks again, EF

Yep, I am a ham.  My call is the same as my name here - KD0RC.  I have an Extra Class license, so I have all the privileges available.  I suggest getting a book on the subject so that you can see what is available to the various license classes, and also to see what it takes to attain those licenses.  Building a receiver or transciever while you are working on getting your license is a good way to stay motivated to keep going.  A good source of info is  http://www.arrl.org/

Buying a used receiver to use while you are learning is not a bad way to go either.  You can get a sense of operating procedure and start to figure out what to do for an antenna (not all receiving antennas make good transmitting antennas, but many do).

Comparing the SoftRock with a FlexRadio is like comparing a go-kart with a Lamborghini...  I bought my SoftRock receiver to learn from; it was fun to build and I play around with it all the time, but I don't use it in my regular ham communications.

The Technician license essentially gives you 10 Meters (28 - 28.5 MHz) and all higher bands.
General, Advanced and Extra give you all bands starting at 160 Meters (1.8 - 2 MHz).

It is not bad to start with the Technician and work your way up.  If you are more interested in HF (10 M and lower), then it might be good to go straight for the General.

The best part of the DSA 815 is how much I am learning from it.  I have wanted a spectrum analyzer for years, and this is the first one that I can afford and has a tracking generator that I can use to characterize filter response and also use to check my antennas.

Best of luck with this, ham radio has been a real fun thing for me for many years.

Thanks, I've been using arrl.org - it's a great resource.  You have me extra motivated to pursue both SDR and a FCC license.

I'd say a Lambo has more horsepower, sounds better, looks better, is more fun to drive (although go karts can be pretty fun) and will attract more attention than a go kart - among other differentiators.  I imagine that some advantages of FlexRadio over SoftRock might be 1) better sound quality receiving and transmitting, 2) better user interface, 3) more features for control of the radio's configuration, and probably others - but I'd be up for hearing more about the key differences.  Thanks again 

[GreyWoolfe]

+1 on the ham license.  Just don't stop at Technician, you only really get some 10 meter privileges.  At the very least, go for your General.  As a General, you have portions of all the ham HF bands.  Much more funner.  By the way, if you live in central Florida, I do testing every month except December as the VE Liaison for the Daytona Beach Amateur Radio Association.

Tom, NW0LF

Thanks - any chance you could summarize what the limitations (of being confined to 10 meter privileges) would mean from a practical experience?  What would you/could you experience at the General level (in what meter ranges) that you would miss with the Technician license at 10 meters?  Thanks again

the 10 meter band with the Technician license would seem to provide at least a glimpse into the overall experience without having to pass more than the basic test.  Yes/No/Sort of?   Thanks again

Yes, the 10 meter band will give you a glimpse and a pretty good starting descriptor of life on HF.  It's like Lay's potato chip though, you can't eat just one .  Additionally, only having that small slice of HF bandwidth to play on will be severely compromised during the many HF contest weekends.  The more bands and bandwidth you have, the better chance of just being able to "ragchew" without contesters clogging up all the frequencies.  12 and 15 meters are great places to play during contests as contesters don't play there.  I play mostly on 20 and 40 meters myself.  I agree with W2AW that the more reliable, longer range communications comes from the lower bands.  However, 10 meters will allow you to put on your swimmies, jump into the shallow end and have fun.  Just don't be surprised if it is but a short time when you get the itch to upgrade.  The technician license mostly gives you VHF/UHF and the microwave bands.  Unless you are interested in emergency communications or microwave experimentation, I'm not sure how much fun you will have there.  In many parts of the country 2 meter and 70 cm band repeater activity is almost non existent except for NTSC traffic nets.  I am trying to drive more activity on our club primary repeater with a weekly "elmers and newbies" net where new hams can practice and develop their radio chops and can ask questions on everything amateur radio related.  So far the net has been well received and people have told me they appreciate the open forum concept which makes the net more fun and a good place to learn. KD0C makes some good points also.  I agree with all of his pointers.  The cool thing about ham radio is that there are so many facets to the hobby, it is almost impossible to find an aspect that won't light a fire for you.  Good luck on your journey.

[granz]

The FFT on signal analyzer is faster than a pure sweep in most applications. There are some that arent, but I forget what it was.  (I think I read it in an Agilent appnote)

I know that a having a signal analyzer with realtime/FFT made my life less miserable when I was doing some oscillator characterizations.  1 Hz RBW/VBW with a ~200Hz sweep made for ~5 minutes/sweep (dont even think about averaging...) on a swept SA, and 1.8s/sweep for the FFT analyzer, even with the sweep expanded upto 1kHz.

In general, FFT will be faster for narrow bandwidths, but swept will be faster for wider bandwidths.  FFT is just computed (in the digital sense), so the faster you can crunch the numbers on your DSP / FPGA / PC the faster you can produce the spectrum.  This is why the FFT analyzers have gotten progressively better with time.  It might seem like I favor the swept SAs, but I really just wanted to mention (in case the OP wasn't aware of) the differences 

[w2aew]

The FFT on signal analyzer is faster than a pure sweep in most applications. There are some that arent, but I forget what it was.  (I think I read it in an Agilent appnote)

I know that a having a signal analyzer with realtime/FFT made my life less miserable when I was doing some oscillator characterizations.  1 Hz RBW/VBW with a ~200Hz sweep made for ~5 minutes/sweep (dont even think about averaging...) on a swept SA, and 1.8s/sweep for the FFT analyzer, even with the sweep expanded upto 1kHz.

In general, FFT will be faster for narrow bandwidths, but swept will be faster for wider bandwidths.  FFT is just computed (in the digital sense), so the faster you can crunch the numbers on your DSP / FPGA / PC the faster you can produce the spectrum.  This is why the FFT analyzers have gotten progressively better with time.  It might seem like I favor the swept SAs, but I really just wanted to mention (in case the OP wasn't aware of) the differences 

This isn't really true anymore, given that there are relatively wideband signal analyzers that can perform FFTs over a wide chuck of spectrum, and stitch the results together.  So, at narrow RBWs and wide spans, the FFT based solutions can be much faster. 

For example, a typical swept analyzer may take 10-30 seconds to sweep 1GHz with a 10kHz RBW, and over 20 minutes to sweep 1GHz with a 1kHz RBW.  I have a signal analyzer that can do those sweeps in 1 second and 3 seconds respectively.

[granz]

The FFT on signal analyzer is faster than a pure sweep in most applications. There are some that arent, but I forget what it was.  (I think I read it in an Agilent appnote)

I know that a having a signal analyzer with realtime/FFT made my life less miserable when I was doing some oscillator characterizations.  1 Hz RBW/VBW with a ~200Hz sweep made for ~5 minutes/sweep (dont even think about averaging...) on a swept SA, and 1.8s/sweep for the FFT analyzer, even with the sweep expanded upto 1kHz.

In general, FFT will be faster for narrow bandwidths, but swept will be faster for wider bandwidths.  FFT is just computed (in the digital sense), so the faster you can crunch the numbers on your DSP / FPGA / PC the faster you can produce the spectrum.  This is why the FFT analyzers have gotten progressively better with time.  It might seem like I favor the swept SAs, but I really just wanted to mention (in case the OP wasn't aware of) the differences 

This isn't really true anymore, given that there are relatively wideband signal analyzers that can perform FFTs over a wide chuck of spectrum, and stitch the results together.  So, at narrow RBWs and wide spans, the FFT based solutions can be much faster. 

For example, a typical swept analyzer may take 10-30 seconds to sweep 1GHz with a 10kHz RBW, and over 20 minutes to sweep 1GHz with a 1kHz RBW.  I have a signal analyzer that can do those sweeps in 1 second and 3 seconds respectively.

Where can I get such an signal analyzer? (seriously )  DSPing has gotten way way faster for sure, but I was more thinking in the price range of the DSA815 which was mentioned here.  Just for a little background, I personally only have an HP 8562a (22GHz swept) and a 35665A (low freq. signal analyzer), but I'd love to get something newer to replace that beast.

You have an excellent youtube channel btw.

[ConKbot]

Where can I get such an signal analyzer? (seriously )  DSPing has gotten way way faster for sure, but I was more thinking in the price range of the DSA815 which was mentioned here.  Just for a little background, I personally only have an HP 8562a (22GHz swept) and a 35665A (low freq. signal analyzer), but I'd love to get something newer to replace that beast.

You have an excellent youtube channel btw.

Agilent N9020A is the one I was using in my above post. But I dont think they are available on the used market cheap enough to consider yet, even for a base model.  Its just a high-spec SA on the base model. By the time you up the input bandwidth, and up realtime bandwidth to 100MHz, I/Q, AM,FM demod, high stability oscillator, software to capture/analyze the input, you can make it go from the cost of a car to the cost of a house.

But even with the basic model (what we have) its still great for the aforementioned reason (fast sweeps) and the dynamic range is great. With the 1Hz RBW, it was ~135 dB of dynamic range IIRC.  One of the RF greybeards at work says he actually got a spectrum of GPS signal with a 30dB active antenna, a 30 dB preamp, a bunch of filtering, and a slow sweep/peak hold run, and letting it run for a few hours.

[w2aew]

The FFT on signal analyzer is faster than a pure sweep in most applications. There are some that arent, but I forget what it was.  (I think I read it in an Agilent appnote)

I know that a having a signal analyzer with realtime/FFT made my life less miserable when I was doing some oscillator characterizations.  1 Hz RBW/VBW with a ~200Hz sweep made for ~5 minutes/sweep (dont even think about averaging...) on a swept SA, and 1.8s/sweep for the FFT analyzer, even with the sweep expanded upto 1kHz.

In general, FFT will be faster for narrow bandwidths, but swept will be faster for wider bandwidths.  FFT is just computed (in the digital sense), so the faster you can crunch the numbers on your DSP / FPGA / PC the faster you can produce the spectrum.  This is why the FFT analyzers have gotten progressively better with time.  It might seem like I favor the swept SAs, but I really just wanted to mention (in case the OP wasn't aware of) the differences 

This isn't really true anymore, given that there are relatively wideband signal analyzers that can perform FFTs over a wide chuck of spectrum, and stitch the results together.  So, at narrow RBWs and wide spans, the FFT based solutions can be much faster. 

For example, a typical swept analyzer may take 10-30 seconds to sweep 1GHz with a 10kHz RBW, and over 20 minutes to sweep 1GHz with a 1kHz RBW.  I have a signal analyzer that can do those sweeps in 1 second and 3 seconds respectively.

Where can I get such an signal analyzer? (seriously )  DSPing has gotten way way faster for sure, but I was more thinking in the price range of the DSA815 which was mentioned here.  Just for a little background, I personally only have an HP 8562a (22GHz swept) and a 35665A (low freq. signal analyzer), but I'd love to get something newer to replace that beast.

You have an excellent youtube channel btw.

I should clarify...  When I said "I have an analyzer...", more accurately I should have stated that "I have an analyzer with my job..."   So, it's not "mine personally", it belongs to my employer.   These analyzers are fantastic, but outside the reach of hobbyists, etc.  The specific unit I'm referring to in this case is a Tektronix RSA5126A realtime spectrum/signal analyzer.

[Richard Crowley]

I thought this was a quite elegant DIR spectrum analyzer with SDR front-end...

http://youtu.be/6YhrKMBrJ2g

[Fank1]

Google YU1LM for a diy SDR receiver.
I built a SA designed by K7ZOI several years ago and it worked so good I bought a used HP, not cheap but GOOD.
WA8HTO