Signal Hound BB60C

[joeqsmith]

Signal Hound offers a 30 day trial for their products.  I have been looking at the BB60C for a while and finally picked one up.   The first problem I ran into is it will not run on my laptop which is an i7 2630QM with a Fresco Logic FL1000 USB 3 controller.   Basically the software crashes when changing settings.  Stable just running swept.

The manual is not very comprehensive.   The USB "Y" cable is very short (24").   

Runs a bit on the warm side.  I measured an increase of 13C for the outside case temperature (27 amb, 40C case).   

There are a fair number of spurs in the 200 - 300KHz range (power supply perhaps).   Spurs seen at every 10MHz to 100MHz.   Above 100MHz, spurs at 160, 240, 320, 360, 370, 380.....  1200, 2400, 2560    Interesting behavior from 708 to 713 when span is set to 5MHz.  Same form 728-733, 748-753, 768-773 ...     




Tried a few different frequencies and the amplitude seems fair compared with HP 8660D.  At 99MHz (away from spur), I measured the following:
0dB, -0.02
-10, -9.96
-20, -19.9
-30, -29.9
-40, -39.8
-50, -49.3
-60, -59.1
-70, -68.9
-80, -79
-90, -89
-100, -98.6
-110, -108.8
-120, -119
-130, -129

Phase noise at 1GHz, 100Hz from carrier is spec'ed at -70.  Looks about -77 with the HP8660D(spec'ed to -95dB at 100Hz).  1MHz looks within spec. 

All in all, looks like a nice setup.  I just need to get a better PC to run it.   

[joeqsmith]

I have not connected it to the external ref yet.   Freq drift (Tempco) not spec'ed that I could find but looks decent.  Shown looking at 8660D set to 1GHz (using GPS) and external 10Hz AM modulation.   

[joeqsmith]

Using the GPS,  1.234567890 GHz signal. 

[joeqsmith]

Using the 8660 to sweep a 9MHz BPF compared with the 3589A.

[rx8pilot]

Do you suppose the spurs are internal to the unit? Are they enough to be a problem?

Sent from my horrible mobile....

[joeqsmith]

Do you suppose the spurs are internal to the unit? Are they enough to be a problem?

Sent from my horrible mobile....

Those are with a terminator attached to the input connector.  I wonder if some could be the PC's noise bleeding into it.   I think for home use, I can work around it.   One thing is certain, it's a whole lot better than my old Tektronix.  It's much faster, better filters, lots of options... 

Will start to look for a better PC to replace my old P4. 

Plotting the drift of my 8754A VNA warming up and one of my early single transistor oscillators.

 

[joeqsmith]

First plot is with the RF generator and BB60C both locked to GPS.   4.0GHz with 400Hz AM modulation.

Unit has been on most of the day and the room temperature is fairly stable.  Next, I use the BB60C's internal reference.   It's fairly stable until I touched the case to cool it.   I then took a small fan and blew it across the SA.   

[rs20]

Haha, awesome. Just for the sake of demonstrating my amazing ability to divide two numbers, I figure that the drift there (with the fan) is nearly 0.05ppm?

I am a few days away from buying one of these myself (because I'm visiting the US; the Aus distributor has a fairly hefty markup). I appreciate the mini-review.

[joeqsmith]

Haha, awesome. Just for the sake of demonstrating my amazing ability to divide two numbers, I figure that the drift there (with the fan) is nearly 0.05ppm?

I am a few days away from buying one of these myself (because I'm visiting the US; the Aus distributor has a fairly hefty markup). I appreciate the mini-review.

You read my mind!      Just a ballpark test but it looks like the total drift was 531.9Hz.   The case temperature went from 41C to 32C with the fan before it became stable or 59.1Hz / degC.   With a 4GHz carrier, that's what, 14.78ppb or 0.0147ppm?  Seems a little too good.  Maybe I screwed up.

Wish I could play with some of the advanced features.  Maybe next week.  I do like what I see so far.   

[rx8pilot]

I take it you saw the Signal Path video on this? Excellent video.

Sent from my horrible mobile....

[jjoonathan]

I've been eyeing the BB60C too, largely because I've got a SA44B that I love but the slow sweep rate chafes. I've taken a look at the RSA306, but I've also seen some intriguing competition from Aaronia:

http://www.aaronia-shop.com/products/spectrum-analyzer/real-time/real-time-usb-spectrum-analyzer

There are options for 88 and 175MHz real time bandwidth, the series goes up to 20GHz, and they offer nifty handheld versions for 500€ more. Of course, all of this is broken out into price tiers that demand hard choices be made about how much one needs each feature. There's also the slight problem that the tracking gen software is vaporware at the moment. But even at the lowest tier, the existence of options and higher tiers suggests a greater potential for, err, "price efficient unofficial upgrades" than the BB60C and RSA306, which piques my curiosity.

Does anyone here have experience with Aaronia products? I'm awfully tempted to spend a bit more vs the BB60C and have a poke around inside, but the relative lack of reviews makes me nervous.

[joeqsmith]

I take it you saw the Signal Path video on this? Excellent video.

Sent from my horrible mobile....

I watched Signal Hound's as well as parts of the Signal Path review.  Skipped the first 20 minutes or so and some of his tests (where he shows some of the tracking generator).   It's complex enough, he really could have made a several parts and dove into it with more detail.   There's not much for reviews of it.  The persistence bug he shows is something I keep running into.  The delta markers is a little strange like he mentions but really I am pretty happy with how it drives. 

I did not get the tracking generator for it.  If the phase noise was better and they allowed you to control it independent from the SA and it had modulation.....  if it could replace my old 8660D, I would get one in a heart beat! 

Attempted to look at the input impedance using the four attenuator settings and I get some strange results.  Appears to be independent of the pre-amp selection.  I suspect it's time for some sleep. 

[Ivan7enych]

Center 4.000000GHz
Stop    4.000000GHz

Push them to add more digits when frequency is > 1GHz. I've told them but this bug is still here.

[joeqsmith]

Center 4.000000GHz
Stop    4.000000GHz

Push them to add more digits when frequency is > 1GHz. I've told them but this bug is still here.

I am surprised they don't increase it.  Maybe there is a limit in their design that prevents it.   

I did ask about using it with a hub and sounds like this is not a problem.  I'll go ahead and order up some new cables and hubs.   Also pulled the trigger on a new PC.  The desktop is about 10 years old now!  Still very reliable but will not run most of the new software.   

Reran the tests looking at the input to the BB60C and I would be interested in seeing someone else measure return loss and impedance.  I have a hard time believing it is as poor as what I am seeing and suspect I am missing something.    Would really like to measure it over the entire 6GHz.   

I played around a little with the audio output.  Very impressive compared to the old Tektronix.   If I change the frequency at all, the software crashes so I have to just tune it, then enable the sound.   Hoping the new PC will solve all of these problems like they are suggesting.   

[hendorog]

I did not get the tracking generator for it.  If the phase noise was better and they allowed you to control it independent from the SA and it had modulation.....  if it could replace my old 8660D, I would get one in a heart beat! 

Attempted to look at the input impedance using the four attenuator settings and I get some strange results.  Appears to be independent of the pre-amp selection.  I suspect it's time for some sleep. 

FWIW the tracking generator does come with a little program which allows it to be used on its own. It's might already installed on your PC in the same directory as Spike.

Also regarding the crashing, I've only got an SA124B but I had strange problems with it at one stage which which were caused by a particular USB cable - and that cable worked fine with other devices. I never worked out if it was dropping a bit too much voltage or if it was something else. So perhaps be suspicious of cables - and USB ports supplying enough power.

[joeqsmith]

I am plugging the cable that comes with the unit directly into the PC.  The system is solid in swept mode.   I ran it nonstop for more than 10 hours yesterday with no problems.  They wrote me that it may not be compatible with the USB chipset.  May laptop has an i7-2630QM and sounds like they are also concerned it may not keep up with the data.  We will see in week.

[joeqsmith]

I wrote SH about my input impedance measurements on the  BB60C.  Sounds like what I am seeing is normal. 

Spectrum analyzers never have good 50 ohm impedance at 0 dB attenuation (essentially straight into a filter then mixer).
Our SWR spec is <3.0:1 at 0 dB attenuation, <1.5:1 for 10-30 dB. A VSWR of <1.5 corresponds to a resistance between 33.33 and 75 ohms.

If you need to improve this, adding a high quality 6 or 10 dB pad to the input will get you significantly closer to the ideal 50 ohm load (at the cost of sensitivity).

I looked in the manual and sure enough,

RF Input VSWR at tuned frequency
? 3.0:1 (<10 dB attenuation)
? 1.5:1 (?10 dB attenuation)

Push them to add more digits when frequency is > 1GHz. I've told them but this bug is still here.

I asked about this as well and received the following responses.   

As far as the resolution on the start, center, and stop frequencies, yes, they are rounded for display purposes. For a high resolution frequency counter, you can look at the modulation analysis mode in zero span.

The frequency resolution internally is much higher, we just truncate to around 6 or 7 digits since that is the approximate accuracy of the internal time base, is easier to display, and most users prefer this amount of resolution.  I believe if you place a marker it will give you more digits of resolution.

[rx8pilot]

The more I look at these units, they seem like a good choice for 'My First SA' choice. For me, it would be a learning tool first but will have business value after I get my feet wet evaluating various radios to be integrated into my existing products as well as pre-compliance testing in house. Being able to add the tracking generator as a separate piece is a nice option to get an SNA when I want.

Thanks for sharing the details of your experience. Is the USB-ness a problem at all? Do you have any wish that it was a more traditional stand-alone instrument?

[joeqsmith]

The more I look at these units, they seem like a good choice for 'My First SA' choice. For me, it would be a learning tool first but will have business value after I get my feet wet evaluating various radios to be integrated into my existing products as well as pre-compliance testing in house. Being able to add the tracking generator as a separate piece is a nice option to get an SNA when I want.

Thanks for sharing the details of your experience. Is the USB-ness a problem at all? Do you have any wish that it was a more traditional stand-alone instrument?

I have never found USB to be a very robust bus but I don't have a lot of experience with the 3.x flavors.  I am not at all apposed to headless equipment like the Signal Hound and Copper Mountain.  Most of my equipment I run headless anyway using Ethernet to GPIB or just direct Ethernet.   My concern with the BB60C was if it would be a problem running it from a hub away from the PC.   As long as the hub and extension cables work out and it's stable, I think it's going to be a good little setup. 

[rx8pilot]

There is no question your USB3 bus has to be totally glitch free. It would be rather frustrating to setup some tests or experiments only to troubleshoot the USB. I suspect that a hub would be no problem - as long as the BB60C is the only thing on that bus. Everything else using a separate bus - tracking gen, mouse, card readers, etc.

That aside, this unit seems like a lot of bang for the buck overall.

[joeqsmith]

The BB60C does not use much of the BW.  It seems like you could run more than just it off a hub.  Most of my USB devices are 1 and 2.  So basically no BW.    I'll try some different combos and post what I find out.   First thing is if it will actually be reliable with a direct connection.   I can understand it not having enough BW to run or not being compatible in some way.  I don't understand how it could just flat out crash.   It just seems like the low level code may not be very robust.   We will see.   

If there is anything you or anyone else wants to see done with it, feel free to ask.  I'm in the same boat, just trying to see if it is a good fit for my hobby use.   

[joeqsmith]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

[joeqsmith]

Glad to help.  Anything else, feel free to ask. 

They have offered to extend my 30 day evaluation.  This is not something I asked for.   They are aware of the problems I am seeing and are aware that I have a new PC on the way.  Communications and service have been excellent considering I am just some hobbyist.   

There is an article on their website about a person using a drone with a Signal Hound attached that he uses to map antenna radiation patterns.   What a great idea!

[rx8pilot]

Glad to help.  Anything else, feel free to ask. 

They have offered to extend my 30 day evaluation.  This is not something I asked for.   They are aware of the problems I am seeing and are aware that I have a new PC on the way.  Communications and service have been excellent considering I am just some hobbyist.   

There is an article on their website about a person using a drone with a Signal Hound attached that he uses to map antenna radiation patterns.   What a great idea!

At the price point, high-end hobbyists are probably a sales opportunity.

On the drone, curious if they struggled with broadband noise from the motor systems.

[hendorog]

Glad to help.  Anything else, feel free to ask. 

They have offered to extend my 30 day evaluation.  This is not something I asked for.   They are aware of the problems I am seeing and are aware that I have a new PC on the way.  Communications and service have been excellent considering I am just some hobbyist.   

There is an article on their website about a person using a drone with a Signal Hound attached that he uses to map antenna radiation patterns.   What a great idea!

I find them great to deal with. There is something to be said for dealing with smaller companies where you are often communicating directly with the designer.

They have always been very receptive to feedback and bug reports as well, they have limited resources but do turn around bugs quickly - so it is worth reporting your thoughts to them.

[Ivan7enych]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

Show as well a region lower 10MHz, it is not shown when you click FullSpan.

The BB60C switches to direct sampling in 0..10MHz, that makes a different noise level (higher) and much better phase noise.

Here are pictures from my unit.
0-20MHz noise
spurs on 9-500KHz

leaked 10MHz
when you make span < 200KHz, it fully switches to one mode and show no noise step between <10MHz and >10MHz

a funny picture with asimmetric left\right phase noise on 10MHz input signal (left side of spectrum goes directly to ADC, right side has double frequency convertion path)

[joeqsmith]

Show as well a region lower 10MHz, it is not shown when you click FullSpan.

The BB60C switches to direct sampling in 0..10MHz, that makes a different noise level (higher) and much better phase noise.

Here are pictures from my unit.
0-20MHz noise
spurs on 9-500KHz

leaked 10MHz
when you make span < 200KHz, it fully switches to one mode and show no noise step between <10MHz and >10MHz

a funny picture with asimmetric left\right phase noise on 10MHz input signal (left side of spectrum goes directly to ADC, right side has double frequency convertion path)

Thanks for posting.  I was hoping someone else had one so we could compare.   Looking at your last plot, the first thing I notice is your "Native RBW".   The manual shows this button but I can not find any reference to it.  More strange is my software does not show it.    I am running version 3.1.2.   

As I mentioned in the first post, I too will see these spurs.   Attempting to repeat your last setup, I show the data with spur rejection both on and off. 

[rf-loop]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

What are detector settings used in these images. Is it quite important information?
Example Image 1, 0dB atten, RBW300kHz, no amplifier. Noise level looks high, even if detector is P-Peak what is perhaps mostly used as default. (..and what do not tell anything about DANL) But if it is Video Average or samples average it is really high.

[joeqsmith]

What are detector settings used in these images. Is it quite important information?

Everything default (detector set to average)

[Ivan7enych]

Thanks for posting.  I was hoping someone else had one so we could compare.   Looking at your last plot, the first thing I notice is your "Native RBW".   The manual shows this button but I can not find any reference to it.  More strange is my software does not show it.    I am running version 3.1.2.   

"Native RBW" has been renamed in new Spike to RBW Shape = Nutall as I remember.

[hendorog]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

What are detector settings used in these images. Is it quite important information?
Example Image 1, 0dB atten, RBW300kHz, no amplifier. Noise level looks high, even if detector is P-Peak what is perhaps mostly used as default. (..and what do not tell anything about DANL) But if it is Video Average or samples average it is really high.

Gain means IF gain. Preamp on/off is a different setting.

[rf-loop]

What are detector settings used in these images. Is it quite important information?

Everything default (detector set to average)

Thanks.

[joeqsmith]

Thanks for posting.  I was hoping someone else had one so we could compare.   Looking at your last plot, the first thing I notice is your "Native RBW".   The manual shows this button but I can not find any reference to it.  More strange is my software does not show it.    I am running version 3.1.2.   

"Native RBW" has been renamed in new Spike to RBW Shape = Nutall as I remember.

Mine were taken with the default filter shape (flat top). 

[Ivan7enych]

I don't understand gain0..3 settings, as I see the lowest noise is seen when I set all gain and preamp settings to auto and reference level to -60dBm.

[rf-loop]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

What are detector settings used in these images. Is it quite important information?
Example Image 1, 0dB atten, RBW300kHz, no amplifier. Noise level looks high, even if detector is P-Peak what is perhaps mostly used as default. (..and what do not tell anything about DANL) But if it is Video Average or samples average it is really high.

Gain means IF gain. Preamp on/off is a different setting.

So is this wrong information?   

.....preamp gain settings....

[hendorog]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

What are detector settings used in these images. Is it quite important information?
Example Image 1, 0dB atten, RBW300kHz, no amplifier. Noise level looks high, even if detector is P-Peak what is perhaps mostly used as default. (..and what do not tell anything about DANL) But if it is Video Average or samples average it is really high.

Gain means IF gain. Preamp on/off is a different setting.

So is this wrong information?   

.....preamp gain settings....

Yes i believe thats wrong.

[hendorog]

Would you mind to upload a no-input sweep over entire spectrum on various gain and preamp settings? I want to see its background noise.

Here are all of the attenuator and preamp gain settings w/ input terminated at connector.  Enjoy.

What are detector settings used in these images. Is it quite important information?
Example Image 1, 0dB atten, RBW300kHz, no amplifier. Noise level looks high, even if detector is P-Peak what is perhaps mostly used as default. (..and what do not tell anything about DANL) But if it is Video Average or samples average it is really high.

Gain means IF gain. Preamp on/off is a different setting.

So is this wrong information?   

.....preamp gain settings....

Yes i believe thats wrong.

Sorry I am wrong on this. On my device the gain setting controls IF gain, but the BB60C is different.

In the BB60C the same gain setting controls the preamp. This is confirmed in the manual:
"Gain control is achieved in the BB60C using the front-end attenuator and preamplifier."

Based on the pictures the effect is the same.

[Ivan7enych]

By the way,
Be careful with input SMA connector, I had to repair it once. Center pin is soldered to very tiny trace on a board, and soldering didn't have enough strength in my case. I spent much time with microscope to accurately add thin wire on top of this trace.

some components identified on a board -

Hittie HMC424 30dB attenuator, absoulte max input power +25dBm, probably the first candidate to replace if someone puts too much power.
HMC270 and HMC849 RF switches
39 - RF amplifiers (?)
...

[rx8pilot]

By the way,
Be careful with input SMA connector, I had to repair it once. Center pin is soldered to very tiny trace on a board, and soldering didn't have enough strength in my case. I spent much time with microscope to accurately add thin wire on top of this trace.

Good argument for the more traditional N connector.

[joeqsmith]

I don't understand gain0..3 settings, as I see the lowest noise is seen when I set all gain and preamp settings to auto and reference level to -60dBm.

I don't understand you not understanding gain settings.      I was using the Settings, Enable Manual Gain/Atten.    This allowed me to run all the combinations.   If I use your example, then select Gain 3 for Gain and 0 dB for Atten, I will see the same waveform .   

I have not figured out what the Preamp  selection does.  I would have thought it was controlled by the gain.  It seems to have no effect. 

[rf-loop]

What are detector settings used in these images. Is it quite important information?

Everything default (detector set to average)

This image 1 (atten 0, ampl 0) result is still so strange that I want ask agen more detailed. 

Question 1  (most important)

Trace and Detector settings.  I do not know what are S&H defaults. Im interest what are in use, exactly. (and because this result is so "amazing")

In S&H (BB60C) user interface there is perhaps these settings:

Left top corner
Traces
 - Type = Clear&Write     (If not this, then what?)

Right bottom corner

Acquisition
 - Video Units = Power  (If not this, then what?)
 - Detector = Average   (you tell it was this)


Question 2

If there is (of course detector - Positive Peak)
0 attenuation and 0 amplifier. What is maximum reference level.


Question 3   (but this need also enough accurate instruments)

Then next what is maximum signal level what do not clip or is not highly compressed when atten 0 and ampl 0.
(you can test it roughly IF you have good or better RF signal generator what have trusted levels what you can adjust and follow what level S&H cursor measure when you adjust level example from -50dB to up as far as S&H can follow this level without starting clip/compress.)
Best to use quite low frequency, example something between example 10 - <100MHz.

This is because I still wonder why base noise power with this setting looks like nearly 20dB higher than expected and around 20db  here is really lot of.

Reason for this question is here in these two images.
SSA3000X. Same settings, 0dB atten, Amplifier OFF, RBW300kHz, Detector WideoAve mode LogPower, ref_-30dBm

First image in « Reply #22 on: August 05, 2016, 11:45:31 PM »

[joeqsmith]

What are detector settings used in these images. Is it quite important information?

Everything default (detector set to average)

This image 1 (atten 0, ampl 0) result is still so strange that I want ask agen more detailed. 

No problem.  Hope I can help.

In S&H (BB60C) user interface there is perhaps these settings:

- Type = Clear&Write     (If not this, then what?)

  Set to Clear&Write

- Video Units = Power  (If not this, then what?)

Set to Power

- Detector = Average   (you tell it was this)

Set to Average

Then next what is maximum signal level what do not clip or is not highly compressed when atten 0 and ampl 0.
(you can test it roughly IF you have good or better RF signal generator what have trusted levels what you can adjust and follow what level S&H cursor measure when you adjust level example from -50dB to up as far as S&H can follow this level without starting clip/compress.)
Best to use quite low frequency, example something between example 10 - <100MHz.

Setting the generator to 99.0MHz, I set the signal and read the peak height using the marker.   I set the center to 99.0M with a 10M span but left RBW and VBW the same.  No signal (terminated) is roughly -76dB.

-60, -59.9
-50, -50.1
-40, -40.1
-30, -30.2
-20, -20.3
-10, -10.3
0, -2.8 (Uncal IF Overload, It appears the Uncal will happen at -4.4)

This is because I still wonder why base noise power with this setting looks like nearly 20dB higher than expected and around 20db  here is really lot of.

Reason for this question is here in these two images.
SSA3000X. Same settings, 0dB atten, Amplifier OFF, RBW300kHz, Detector WideoAve mode LogPower, ref_-30dBm

[rf-loop]

What are detector settings used in these images. Is it quite important information?

Everything default (detector set to average)

This image 1 (atten 0, ampl 0) result is still so strange that I want ask agen more detailed. 

No problem.  Hope I can help.

In S&H (BB60C) user interface there is perhaps these settings:

- Type = Clear&Write     (If not this, then what?)

  Set to Clear&Write

- Video Units = Power  (If not this, then what?)

Set to Power

- Detector = Average   (you tell it was this)

Set to Average

Thanks. This mean that SSA  and SH have exactly same settings in these two images.

I will look rest things with signal later when it is possible.

[joeqsmith]

No problem.  Just to add to this, using all of the same settings above except changing the Gain to Gain 2, with no input signal the marker displays -106dB.  I then reapplied the 99MHz signal and measured the following:

To make it a little easier to view, I ended up repeating the test using my old HP8660D for each gain setting with 0dB attenuation.

The 8660D has both a step and variable attenuator.  When I ran this test, I kept the variable attenuator at the same setting until the end of the test.  I then fine adjusted it for each gain setting and recorded the marker readout where it would start to display Uncal.     

[joeqsmith]

Show as well a region lower 10MHz, it is not shown when you click FullSpan.

Some difference.

Last plot showing HP8660D set to -130 with a MWM 20dB attenuator (-150).   Flat top, Power, Average, Gain 3, 0 Attn.   Picture of the same signal at -110dB on the old Tektronix.  30Hz RBW is the limit of it.     

[rf-loop]

Here around same with SSA3000X with Attenuator 0 and Aplifier 0  (later I will add levels with 20dB Preamp  ON)
If set Siglent Atten some dB  then they are more like same scale position (just different design, this itself mean not better or worse) but then Siglent useable  level range is clearly more, difference is tenfold.
(Propably reason is Siglent  much lower noise level.)

 (also note that Siglent noise level in this table is Trace averaged Positive Peak Detector  output noise level what is very different  if look  Video Average Detector output  noise level what is typically used for low DANL numbers).

[joeqsmith]

I had ordered an ANKER USB hub along with a few different length of extension cables from Cable Matters.  It's a little too early to say if it will work or not.  I tried just the hub and SH,  that seemed to work.  Then with just the SH with extension cables and saw no problems.  PC to hub to extension cables ti SH, no problems.  PC to extension cable to hub to SH, no problems.  Last, PC with two 6' extension cables (12' total) to hub to SH.  Ran it like this for a few hours and I saw no problems.   My biggest concern it that the hub seems to have the same brand chip set used in my laptop. 

I think until the new PC arrives and I can actually get the SH to run something other than swept mode, I can't really say if this combination will work or not. 

Ran one more test, keeping all the settings as before but used my old 8640B generator and set the RBW to 1Hz.  I only ran it with a gain setting of 0.   

[joeqsmith]

The new PC made it in yesterday and I have been playing around with the Signal Hound non stop and getting the PC setup, except for a little sleep.   I never even thought about it until after I got the PC that I had forgotten about needing a printer port.  I guess that's a thing of the past now but will need it to run my PROM programmer.   I'm also thinking there may be some problems with getting Labview working and I am not sure yet about how I am going to go about porting over some of my code I wrote for my motorcycle HIL simulator.   Lots to do.   

A little toe dip into the digital world.....

[Muxr]

Congrats on the new rig! Can't believe you held out that long on an old Pentium 4.

[G0HZU]

I can see you are sending
 HELLO EEVBLOG !!! FROM JOE Q SMITH !!!!
in ASCII in your data

[hendorog]

I can see you are sending
 HELLO EEVBLOG !!! FROM JOE Q SMITH !!!!
in ASCII in your data

Haha, love it!

[joeqsmith]

  I wondered if anyone was going to catch that.   

The old PC is almost 10 years old! Trying to see what compatibility problems I will run into next.  I wonder if my DOS games will still run.

Some friends of mine play a lot of video games and told me to get an M.2 drive.  I was planning on two SSDs on RAID 0.  Windows 10 boots from it.  Wow!    Now where is the printer port???!!!

[bson]

I never even thought about it until after I got the PC that I had forgotten about needing a printer port.  I guess that's a thing of the past now but will need it to run my PROM programmer.

Just pick up a TL866 programmer from eBay or Amazon... download the latest software from http://www.autoelectric.cn/en/TL866_main.html

Cheap.  Nothing fancy, but it works.

[joeqsmith]

I never even thought about it until after I got the PC that I had forgotten about needing a printer port.  I guess that's a thing of the past now but will need it to run my PROM programmer.

Just pick up a TL866 programmer from eBay or Amazon... download the latest software from http://www.autoelectric.cn/en/TL866_main.html

Cheap.  Nothing fancy, but it works.

Thanks for the link.  Nice little programmer.  Looks like they don't support the Xilinx and Altera serial PROMs. Other than that, it looks much better than what I have now.     

[KE5FX]

As it turns out, my Ethernet GPIB controllers from National Instruments are no longer supported.  Of course the drivers will not work under Windows 10.  To replace the two would be in the $2600 range.       And this is why I don't buy hardware from NI.  Why NI feels I need 1G to talk on GPIB      Of course there is other software that will not run on it as well.   Set up VirtualBox with Windows XP 32.  Installed all the old software along with Labview and the GPIB drivers.  And...... IT WORKS!   What a way to cripple a new PC. 

The GPIB-ENET boxes have actually been unsupported officially since Vista, possibly earlier.  I don't have one myself, but a user of my GPIB freeware sent a workaround that I added to the FAQ.  If you do want to use them under Windows 10, go here and search for "Q. How can I use a National Instruments GPIB-ENET adapter under Windows XP or Vista?"  If the utility in that zipfile still runs, it will probably be possible to get them working.

To be fair to NI, those boxes are, what, probably 20 years old now?  At some point every vendor is going to stop writing drivers for any given combination of hardware and OS.  The Ethernet box shouldn't need any drivers, of course, but at the time they came out, it probably seemed like a perfectly reasonable strategy for NI to create their own address configuration scheme.  When 16-bit Windows and DOS apps stopped being supported by Microsoft, so did their config tool.

[joeqsmith]

As it turns out, my Ethernet GPIB controllers from National Instruments are no longer supported.  Of course the drivers will not work under Windows 10.  To replace the two would be in the $2600 range.       And this is why I don't buy hardware from NI.  Why NI feels I need 1G to talk on GPIB      Of course there is other software that will not run on it as well.   Set up VirtualBox with Windows XP 32.  Installed all the old software along with Labview and the GPIB drivers.  And...... IT WORKS!   What a way to cripple a new PC. 

The GPIB-ENET boxes have actually been unsupported officially since Vista, possibly earlier.  I don't have one myself, but a user of my GPIB freeware sent a workaround that I added to the FAQ.  If you do want to use them under Windows 10, go here and search for "Q. How can I use a National Instruments GPIB-ENET adapter under Windows XP or Vista?"  If the utility in that zipfile still runs, it will probably be possible to get them working.

To be fair to NI, those boxes are, what, probably 20 years old now?  At some point every vendor is going to stop writing drivers for any given combination of hardware and OS.  The Ethernet box shouldn't need any drivers, of course, but at the time they came out, it probably seemed like a perfectly reasonable strategy for NI to create their own address configuration scheme.  When 16-bit Windows and DOS apps stopped being supported by Microsoft, so did their config tool.

The problem is not setting the the GPIB-ENET.   Setting the IP is easy enough and I use static anyway.     Yes, I would guess in the order of 20 or so years old now, still work fine.   If it were a card plugged into an ISA bus, I could understand but it's Ethernet.   

[KE5FX]

As it turns out, my Ethernet GPIB controllers from National Instruments are no longer supported.  Of course the drivers will not work under Windows 10.  To replace the two would be in the $2600 range.       And this is why I don't buy hardware from NI.  Why NI feels I need 1G to talk on GPIB      Of course there is other software that will not run on it as well.   Set up VirtualBox with Windows XP 32.  Installed all the old software along with Labview and the GPIB drivers.  And...... IT WORKS!   What a way to cripple a new PC. 

The GPIB-ENET boxes have actually been unsupported officially since Vista, possibly earlier.  I don't have one myself, but a user of my GPIB freeware sent a workaround that I added to the FAQ.  If you do want to use them under Windows 10, go here and search for "Q. How can I use a National Instruments GPIB-ENET adapter under Windows XP or Vista?"  If the utility in that zipfile still runs, it will probably be possible to get them working.

To be fair to NI, those boxes are, what, probably 20 years old now?  At some point every vendor is going to stop writing drivers for any given combination of hardware and OS.  The Ethernet box shouldn't need any drivers, of course, but at the time they came out, it probably seemed like a perfectly reasonable strategy for NI to create their own address configuration scheme.  When 16-bit Windows and DOS apps stopped being supported by Microsoft, so did their config tool.

The problem is not setting the the GPIB-ENET.   Setting the IP is easy enough and I use static anyway.     Yes, I would guess in the order of 20 or so years old now, still work fine.   If it were a card plugged into an ISA bus, I could understand but it's Ethernet.   

Hmm.  The only problems I've heard of (at least pre-Windows 10) have been related to the address configuration method that NI used.  It would be interesting to drill down and figure out exactly why they've stopped working in Windows 10, because as you say, it's just Ethernet.   

[joeqsmith]

A while back I had "upgraded" Labview and later discovered that they had jacked up the serial communications.   As I understand it they had rewrote the VISA interface in .NET.   Some software manager must have felt that was the future and would make things more maintainable.    Once again we are talking about a serial port, technology older than the PC itself!  Dare I say that once again, Windows handles the low level heavy lifting (like it's that complex).  And now we have a company who's whole business is based around T&M who can't figure out how to make a serial port work. 

I spent a fair amount of time working with them to try and sort it out.   They were getting close and actually had a workaround.  In the end, I switched to FTDI for all of my USB serial ports.   

They had jacked up the GPIB at one time.  I get on the phone with their support.  They were unable to replicate a problem that I could replicate on multiple PCs.   It took a few days until I had them walk me through their entire setup.  They had a GPIB controller in the PC, plugged into the the GPIB ENET.   They thought this is how it was used!         

[joeqsmith]

I'm still evaluating the unit.   For the most part I'm pretty happy with it but there are a few things I don't care for. 

You can't turn it off or put it into a deep power down mode.  Using hex to represent a variable size symbol for the demodulation trigger seems like a poor choice.  Why not let me select it.  There is no ASCII decode.

I have yet to try and run it from Labview.  Working on porting some of my other code which is taking a fair amount of time.  Hope to be able to pull the plug on the old P4 in a few weeks. 

[Ivan7enych]

I've spent some time looking at different radio protocols with BB60c. Let me show some pictures.

DECT home phone, a waterfall view in a realtime mode

and a single packet (2FSK with preamble) on FM chart in zero span mode

[Ivan7enych]

FRSky RC transmitter
http://www.frsky-rc.com/

frequency hopping packets over wide area in 2.4GHz range, some weak wifi is also visible here

shape of a single channel

shape of another channel, with some imperfections

decoded packet of data

For me it's rather hard to calculate bits per second rate of unknown signal, to find it I look at FM chart in zero span mode, measure time interval between first 10 pulses in apreamble and then test some round numbers near the calcuated value.

[Ivan7enych]

very cheap 27MHz RC car toy, AM pulses in zero span mode

dialog based car security system, in waterfall view there are visible - 1 short send packet, then short reply from car, then second short send packet, and a long reply from car, as usual 2FSK

a cordless mouse, 2FSK, one packet in FM chart in zero span mode

[Ivan7enych]

another RC controll transmitter, 1w LRS in 440MHz range.
connected directly to input with 40dB attenuator
8 frequency hopping channels, in a waterfall

FM chart in zero span mode, single packet, 2FSK

AM chart in zero span mode, some deviations of output power are visible within a single packet

a single decoded packet in protocol analysis mode

[Ivan7enych]

A small task, made with BB60c and tracking generator (plus directional coupler).

My RC hexacopter have 440MHz control receiver and 1.2G 1W video transmitter. I noticed that receiver looses cencitivity very much near the video transmitter even when transmitter emits nothing in 440MHz range. I think it's input circuit is oversaturated by a very strong signal.

Here is a reflection graph of a 440MHz dipole antenna from receiver, one can see the dipole has good impedance in 440MHz, but also can receive power in 1.34GHz, close to my video Tx range.

I've made a simple low pass filter (3 smd capacitors and 2 pieces of wire as inductors), here is a filter response, -0.4dB in 440MHz and -40dB in 1.2G, which is very good for me.

After combining the filter with the antenna, the reflection looks different, good impedance in 440MHz, some additional resonance at 510MHz (which I don't care of) and no other resonances upper.

That's all, I hope it will be interesting for you.

[joeqsmith]

I've spent some time looking at different radio protocols with BB60c. Let me show some pictures.

DECT home phone, a flow in a realtime mode

and a single packet (2FSK with preamble) on FM chart in zero span mode

Good stuff!   Thanks for the post!   

I'm curious how long you have owned yours if you have had any problems with it.  Are you cooling yours.  Do you unplug it when not in use?

[Ivan7enych]

I've spent some time looking at different radio protocols with BB60c. Let me show some pictures.

DECT home phone, a flow in a realtime mode

and a single packet (2FSK with preamble) on FM chart in zero span mode

Good stuff!   Thanks for the post!   

I'm curious how long you have owned yours if you have had any problems with it.  Are you cooling yours.  Do you unplug it when not in use?

Thank you.

I have it for ~8 months.

There was a problem with sma connector center pin -
https://www.eevblog.com/forum/testgear/signal-hound-bb60c/msg998092/#msg998092
As I remember, I've got on ebay 1-4GHz directional coupler with very tight SMA connectors, and attached it to analyzer with short SMA-female-SMA-female adapter. SMA connector on the coupler was so tight that it pushed out center pin of the adapter and that makes some damage to BB60 connector...

Cooling it with a fan and a heatsink should be a good solution, my unit heats up slowly from 27c to ~45c. And this temperature change shifts "store thru" calibration in scalar analysis mode.

[joeqsmith]

I've spent some time looking at different radio protocols with BB60c. Let me show some pictures.

DECT home phone, a flow in a realtime mode

and a single packet (2FSK with preamble) on FM chart in zero span mode

Good stuff!   Thanks for the post!   

I'm curious how long you have owned yours if you have had any problems with it.  Are you cooling yours.  Do you unplug it when not in use?

Thank you.

I have it for ~8 months.

There was a problem with sma connector center pin -
https://www.eevblog.com/forum/testgear/signal-hound-bb60c/msg998092/#msg998092
As I remember, I've got on ebay 1-4GHz directional coupler with very tight SMA connectors, and attached it to analyzer with short SMA-female-SMA-female adapter. SMA connector on the coupler was so tight that it pushed out center pin of the adapter and that makes some damage to BB60 connector...

Cooling it with a fan and a heatsink should be a good solution, my unit heats up slowly from 27c to ~45c. And this temperature change shifts "store thru" calibration in scalar analysis mode.

I have not looked into how they connect a tracking generator.  Are the two not locked together when running scalar?  In other words, is this a frequency drift or amplitude?

Have you had any problems with your needing to reset?  Disconnects?    Curious what USB chipset and CPU you are using and if you have any extension cables or a hub?    Reason I ask is I have had mine act up a few times now and have been unable to determine the cause.   

Auto data rate would be a nice feature.   Have you looked at controlling it with their API yet?   

[rs20]

As I remember, I've got on ebay 1-4GHz directional coupler with very tight SMA connectors, and attached it to analyzer with short SMA-female-SMA-female adapter. SMA connector on the coupler was so tight that it pushed out center pin of the adapter and that makes some damage to BB60 connector...

A couple of clarifications -- when you say very tight, do you mean a) that the eBay directional coupler is badly dimensioned, or b) that you did up the connection with more than the recommended 0.45 Nm of torque for a brass SMA connector?

Also, pedantically, when you say SMA-female-SMA-female adapter, do you mean this?:



Because that's male-to-male (it's based on the little centre pin, not on the threaded parts).

[Ivan7enych]

As I remember, I've got on ebay 1-4GHz directional coupler with very tight SMA connectors, and attached it to analyzer with short SMA-female-SMA-female adapter. SMA connector on the coupler was so tight that it pushed out center pin of the adapter and that makes some damage to BB60 connector...

A couple of clarifications -- when you say very tight, do you mean a) that the eBay directional coupler is badly dimensioned, or b) that you did up the connection with more than the recommended 0.45 Nm of torque for a brass SMA connector?

Also, pedantically, when you say SMA-female-SMA-female adapter, do you mean this?:
Because that's male-to-male (it's based on the little centre pin, not on the threaded parts).

This MAC directional coupler and SMA adapter (see photo).

You're right, I meant "female" by its thread. If SMA male-female is relied on center pin, than what about RP-SMA (with hole instead of pin on the same connector) ?

I tightened connectors by hand, I didn't realized I can possibly exceed recommended torque by fingers.

[Ivan7enych]

I have not looked into how they connect a tracking generator.  Are the two not locked together when running scalar?  In other words, is this a frequency drift or amplitude?

Have you had any problems with your needing to reset?  Disconnects?    Curious what USB chipset and CPU you are using and if you have any extension cables or a hub?    Reason I ask is I have had mine act up a few times now and have been unable to determine the cause.   

Auto data rate would be a nice feature.   Have you looked at controlling it with their API yet?

1. Yes, tracking gen is connected with USB to computer and with sync cable to analyzer, the BB60c generates sync pulses on every frequency change.

2. No issues. My laptop - Intel i7, Nvidia Gt640m, and some Intel USB3 chipset

3. No, but I wanted to make a sweep with tracking generator, to sweep in one region and look at another region (to check frequency doubler).
I wrote a small app based on theyr API and published my sweep software here -
https://signalhound.com/support/forums/topic/api-for-standalone-tg44/

[Ivan7enych]

About scalar analysis -

As I understand, BB60C has many frequency bins (27MHz wide), each one is calibrated by amplitude and internal temparature drift (of the amplifier gain). So in usual sweep mode or realtime mode all dBm measurements are correct over all temperature range of the unit.

But in scalar analysis BB60c turns off all corrections and rely on "Store Thru" calibration only, which you must do before any scalar measurement. So first you need to connect output to input, make "Store Thru" calibration and then connect any other DUT.

That works OK until temperature of the unit changes by ~5-10degree, than dB measurements goes off by 1-2dB and software warns you that you must repeat calibration again. So if you start measurements with cold unit, you must reconfigure your setup quickly (between calibration and actual measurement). Or you just need to turn everything on and wait until temperature stabilizes (half an hour at least).

I think much better would be to add external temperature stabilization, but it will require more power than can give one USB port..

Here are my 2 threads about this on HS forum
https://signalhound.com/support/forums/topic/bb60c-tg44-in-sna-mode-resets-store-tru-calibration-every-2-minutes/
https://signalhound.com/support/forums/topic/bb60c-tg44-strange-jumps-on-plot-in-sna-mode/

[joeqsmith]

What I was asking is it because of the frequency drift between the spectrum analyzer and tracking generator or are these two locked together?   

Looking at the manual for the TG44A, they talk about the sync but not the reference.   It looks like the BB60C can supply the reference for the tracking generator.   Do you run it this way?

[Ivan7enych]

What I was asking is it because of the frequency drift between the spectrum analyzer and tracking generator or are these two locked together?   

Looking at the manual for the TG44A, they talk about the sync but not the reference.   It looks like the BB60C can supply the reference for the tracking generator.   Do you run it this way?

I had 2 discussions about this with SH -
https://signalhound.com/support/forums/topic/bb60c-with-10mhz-external-reference-from-tg44-shows-me-large-frequency-deviation/
https://signalhound.com/support/forums/topic/scalar-analysis-why-resolution-is-limited-to-1khz/

The TG44 has 10MHz output (but no 10MHz ref input) which I can connect to bb60c ref input, theoretically.

But 10MHz output is simply an output of internal clock generator, while final frequency calibration of TG44 output is done in software, as a result, TG44 signal output has much better frequency accuracy than it's 10MHz "reference" output.

Current software unfortunately is limited to 1KHz resolution in scalar analysis, the internal frequency calibration of both units is much better than this value, so I see no use to connect 10MHz Ref to each other.

What I want from both units, is better frequency resolution in lower frequency range. I can do external sweep with any other generator and use MaxHold to get a track, but is makes dynamic range much lower. The native combination of TG44 + BB60c actually does 2 sweeps in scalar analysis (one with -10dBm and second with -30dBm) and have > 100dBm dynamic range in this mode.

[joeqsmith]

I had 2 discussions about this with SH -
https://signalhound.com/support/forums/topic/bb60c-with-10mhz-external-reference-from-tg44-shows-me-large-frequency-deviation/
https://signalhound.com/support/forums/topic/scalar-analysis-why-resolution-is-limited-to-1khz/

The TG44 has 10MHz output (but no 10MHz ref input) which I can connect to bb60c ref input, theoretically.

But 10MHz output is simply an output of internal clock generator, while final frequency calibration of TG44 output is done in software, as a result, TG44 signal output has much better frequency accuracy than it's 10MHz "reference" output.

Current software unfortunately is limited to 1KHz resolution in scalar analysis, the internal frequency calibration of both units is much better than this value, so I see no use to connect 10MHz Ref to each other.

What I want from both units, is better frequency resolution in lower frequency range. I can do external sweep with any other generator and use MaxHold to get a track, but is makes dynamic range much lower. The native combination of TG44 + BB60c actually does 2 sweeps in scalar analysis (one with -10dBm and second with -30dBm) and have > 100dBm dynamic range in this mode.

And this temperature change shifts "store thru" calibration in scalar analysis mode.

Locking the 10MHz reference may correct a fair amount of the temperature drift.  I don't have the tracking generator so I can't test it.   You are right, this will not help with resolution or accuracy.   I wonder if they remove the display resolution because customer complained about an error between the generator and spectrum analyzer.   

[Ivan7enych]

Here is frequency temperature drift between TG44 and BB60c

Both units are cold (temperature of BB60c is +28c)
I set 4GHz -30dBm on generator, connect it with -3dB attenuator and 20cm rg316 cable (which have some loss)
BB60c shows me -33.4dBm with 200Hz frequency error

After warmup (+38c) I see -33.7dBm and 85Hz frequency error

after more warmup (+43c) I see the same -33.7dBm and frequency error -15Hz

[Ivan7enych]

Here is amplitude drift and dynamic range in Scalar analisys mode.

1. Screen with -3dB atten, without calibration, 10MHz - 4GHz span

2. Calibrated at +30c, in cold state, with Store Thru

3. Disconnected input, ~100dB (+3dB for attenuator) dynamic range.

4. Warmed up to +36c, amplitude drifted by 1dB, warning about it is visible.

5. Dynamic range with "High Range" option off. Much less range, but 2times faster update.

[joeqsmith]

I read your posts on their site.   Sorry not getting back sooner.  I have been in manic mode for last few days getting my motor cycle simulator under Windows 10.  Done for the most part and am surprised how well the new PC can handle it.  All of the low level interface to the PCI bus was coded in C and now I am doing it all in Labview.  I'm impressed.  Next I need to look at the GPIB controller.   

I connect TG44 10MHz output to BB60c reference input, and I expect to see exactly the same frequency, but it’s not.

After further discussion here, we are going to add the ability to specify whether you are using the TG 10MHz reference output, so it will disable this frequency correction. We think allowing the user to manually specify this is the best way forward. It will likely be a new checkbox in the TG control panel. Look for this in the next release.

Assuming they made this change, could you repeat the drift test with the two tied together?   

I have been running the BB60C and still trying to hunt down a problem where it will hang.   I am not liking how the software would crash with my laptop.  Even if there were buffer overflows, underruns, bad cable.... I would expect the software to handle it.  With the new PC I still see some strange problem where the Spike software looses connection with the BB60C.   This is VERY rare.   Looking in the device manager, I can see the BB60C and all seems fine with it.  Green LED is active.  I can restart Spike and it will not find it. Windows does not see the device disconnect/reconnect.    There are two ways to recover.  I can power cycle the BB60C.  This should never happen in any case IMO.   The other is to restart the driver in device manager. 

The company has offered to provide a new cable as they suspect this is where the problem is.  I am able to bend the cable and such and it will not cause the problem.  I doubt this is the problem.   

By VERY rare, I can run it for 8 hours and may not see it happen.   At first I thought it had something to do with the audio decode but this does not seem to be the case.   I suspect it is some sort of overflow problem.  Have you ever seen this sort of problem with yours?




 

[Ivan7enych]

Assuming they made this change, could you repeat the drift test with the two tied together?   

Here is 4GHz signal from tracking gen.

1. 10MHz reference from TG44 is not connected to BB60c, ~100Hz offset
2. 10MHz reference is connected, BB60c uses external reference, frequency calibration is turned off - zero offset
3. 10MHz reference is connected, BB60c uses external reference, but frequency calibration is still on - offset is 1.5KHz

The company has offered to provide a new cable as they suspect this is where the problem is.  I am able to bend the cable and such and it will not cause the problem.  I doubt this is the problem.   

Does your BB60c have enough voltage? My unit shows me 4.6V and I had no issues on 2 computers.

[Ivan7enych]

Catched some avia communications with BB60c + antenna at 129.8MHz, AM modulation.

Center frequency of one signal has some offset, may be due to doppler effect of a flying plane?

[joeqsmith]

Assuming they made this change, could you repeat the drift test with the two tied together?   

Here is 4GHz signal from tracking gen.

1. 10MHz reference from TG44 is not connected to BB60c, ~100Hz offset
2. 10MHz reference is connected, BB60c uses external reference, frequency calibration is turned off - zero offset
3. 10MHz reference is connected, BB60c uses external reference, but frequency calibration is still on - offset is 1.5KHz

The company has offered to provide a new cable as they suspect this is where the problem is.  I am able to bend the cable and such and it will not cause the problem.  I doubt this is the problem.   

Does your BB60c have enough voltage? My unit shows me 4.6V and I had no issues on 2 computers.

No drift as I would expect.   Now what happens if you run #2 and repeat the amplitude drift?

Mine shows 4.8V and is using version 7.   

[joeqsmith]

It's been a month, time for a little update.   I now suspect the computer I bought has one port that causes problems with the BB60C.   It's a Dell 5810 and it has several USB 3 ports.   There is one port that is on the back of the PC that is soldered to the MB that when the SH is connected to it, it will very rarely hang.  I still thing that the firmware in the SH needs some help just in how it behaves when it hangs.   Having to pull the cable or power cycle  to recover is something I would expect from a high school science project but not from a professional group.   I hope over time they will continue to improve it. 

I don't like having to pull the power on the hub to turn the thing off.  Again, this just seems stupid to not have a way to put the thing into a least some sort of very low power mode, if not just to turn the thing off. 

I am still up in the air on the whole headless thing.  It saves room but the PC and monitor is not near the bench where I normally work.  It's not been a problem to run it from the hub with the long USB cables.   But having to turn the PC on, wait for it to boot, run the program.... Its a bit of a pain.  I wonder how long USB 3 will be supported and the thing gets tossed in the trash when Windows 15 comes out, or they just don't support it.   Time will tell.   

Overall I still like it for my home hobby sort of work.  Been having some fun with the digital side of life. 

[joeqsmith]

Nope, not the port.  It hung again yesterday. Something in the USB stack on the BB60C I suspect in how it hangs.   

Attempting to do use it to decode some very low 2FSK data rates in the 915MHz.  Seems this may be a problem.  The best I seem to be able to do is oversample the data but then there is no way to decimate it in the digital domain.  So rather than say "010"  you would have "00001111000" for a 4X over sample.  Then it really limits the amount of data it can collect. 

Wishing now it had more complex triggers as well. For pulsed RF work, they have what they call a video trigger where you set the power level you want to trigger.  But you can't seem to search for a pattern after this. 

Depending what SH comes back with for the slow data rates, I may attempt to decode the raw data in LabVIEW.   Anyone else run into this?  Is  there a work around?

[joeqsmith]

I posted my suggestions about adding a more complex trigger.  Maybe with some luck we can get that in there.   When I asked about the slow BAUD rates using 2FSK, it sounds like this is a software limit.  Dumping the raw IQ data and looking at it with LabVIEW, it seems to all be there.  Oversampling like I was doing was one of their suggestions for using the demodulation as well.   I tried to collect a full packet by oversampling the data and using their copy to clip board.     It does seem to work but     Nice thing is that it looks like you can get the raw digital data this way rather than doing your own demodulation.   

What's funny is you can crank it up with a much more complex modulation than binary, sending a lot more data and it does pretty good.  But slow binary data, it falls apart.   

More I play with it, the more I like it.

[Bobaganoosh]

Does anyone here have experience with Aaronia products? I'm awfully tempted to spend a bit more vs the BB60C and have a poke around inside, but the relative lack of reviews makes me nervous.

I was wondering this same question, which led me to this forum as others have been wondering too. I just thought I'd share my experience here, hoping that the next person who looks will be able to start off where I left off. I thought the features they were touting over the BB60C would warrant the price difference and I found one for sale at a distributor for ~$4200 (less than the $5700 quote they gave me) so it was looking pretty doable.

I can't find any semblance of a review from people online (not posted to their website) that indicates people like their equipment or have had good luck. Apparently, the old V2 and V3 units were "kids toys" or "garbage" according to various sources I've found and those people generally indicated that the V4 specs looked considerably better and the software has improved drastically, but they are now afraid to purchase the new SA's. I couldn't find anybody who had actually purchased or tested a V5 8060 USB RTSA...I found one discussion on a WISP forum where one guy tried to convince everyone else that Aaronia doesn't lie on their specifications, while everyone else there tried to indicate that they must be lying because what they were claiming is not possible. I honestly don't know who was correct.

What I do know is this:
Aaroniausa.com indicates that they partner with Kalman Creations LLC as the only authorized US distributor.

The website says they no longer do demo units ("too many models available now"), they offer a 30 day return policy (for your money back, minus a 15% restocking fee, plus shipping, plus any repairs required (scratches, etc.)) and a 10 year warranty, but the warranty is only good if you buy it directly from them...there goes that $4200 quote I received...

So I emailed them for information and a quote and asked Kaltman if they do demos (before reading the cancelled demo policy on the website). He said yes, but I had to have an exec. at my company fill out this form (attached) that seems like a pretty ridiculous request. If there's so much as a scratch, we'll be billed to repair or replace the entire unit, if it's late, they'll charge us 10% the total price per month it's late. if it's not working when you receive it, but don't inform them within 4 days of receiving it that it is not working, you bought it.

I have demo'd oscilloscopes and analyzers much more expensive without signing anything.

Further, I have seen various posts online about their customer service being, to put it politely, lacking. The guys at dlsreports.com were particularly stern when warning against using products from these guys.

Long story short, I have now been entirely scared away from even getting a demo unit, let alone purchasing one. I have talked to and heard from many people about how great the BB60A, B and C are and what great experiences they have. I was thinking I'd get a little bit more from a more expensive unit, but I'm no longer convinced. I think there's a BB60C in my future.

[KE5FX]

I have no affiliation with either company, but I'd say go with Signal Hound if you need a spectrum analyzer, or Aaronia if you are in the market for an "electrosmog meter."

A company that's reluctant to lend demo units to credible prospective customers is a company that may have something to hide.  A company that sells woo to the tinfoil hat crowd has nothing worth hiding in the first place.

[bson]

Aaronia if you are in the market for an "electrosmog meter."

Wow, what?!  "Electrosmog"?  

[joeqsmith]

The website says they no longer do demo units ("too many models available now"), they offer a 30 day return policy (for your money back, minus a 15% restocking fee, plus shipping, plus any repairs required (scratches, etc.)) and a 10 year warranty, but the warranty is only good if you buy it directly from them...there goes that $4200 quote I received...

So I emailed them for information and a quote and asked Kaltman if they do demos (before reading the cancelled demo policy on the website). He said yes, but I had to have an exec. at my company fill out this form (attached) that seems like a pretty ridiculous request. If there's so much as a scratch, we'll be billed to repair or replace the entire unit, if it's late, they'll charge us 10% the total price per month it's late. if it's not working when you receive it, but don't inform them within 4 days of receiving it that it is not working, you bought it.

I have demo'd oscilloscopes and analyzers much more expensive without signing anything.

Long story short, I have now been entirely scared away from even getting a demo unit, let alone purchasing one. I have talked to and heard from many people about how great the BB60A, B and C are and what great experiences they have. I was thinking I'd get a little bit more from a more expensive unit, but I'm no longer convinced. I think there's a BB60C in my future.

If it's like the Signal Hound, I assume it was targeted towards hobbyist and why they don't offer demo units.  I can believe you would have a much easier time working for a large company and getting a trial on an $80,000 scope for a month than a hobbyist trying to get a demo on a low end analyzer.  They may not get a lot of requests from real companies and want to make sure they are legit.   Signal Hound was basically the same deal.  Buy it and return it with a restock fee.  There was some sort of time limit for the return but I was upfront about what I was doing with the unit and they told me that they would not hold me to the allotted time.   

Down side IMO of the BB60C,

Uses USB
No way to power down the unit except unplug it
Custom supplied cable is short and requires two ports
You are at the mercy of their development group to enhance their software which needs some work
Runs fairly hot and will drift with the temp but GPS to the rescue
Not a fan of the SMA on it but so far it's not been a problem
The requirements for the PC forced me to tack on the cost of a new PC as my old i7 laptop would not run it reliably.  Crashes and lockups are pretty rare with the new PC but they do still happen. 

Plus side,
It's affordable for the hobbyist wanting to play with a real time SA
Spec's are decent for the price
Hardware seems reliable
They offer a way to develop third party tools
You can save the raw IQ data for post processing
It takes up very little space 

If they offered to let me return it today for a full refund, including shipping,  no questions asked, I would say forget it. What I can do with it far outweigh the few problems I see with it. 

[joeqsmith]

Taking the plunge into the BB60C's API.  Anyone else on here writing custom software for it?   

So far, most of it seems fairly straight forward.

[KE5FX]

Taking the plunge into the BB60C's API.  Anyone else on here writing custom software for it?   

So far, most of it seems fairly straight forward.

I added support for their original SA44 model to SSM a couple of years ago.  Not sure if the API is backward-compatible, but if it is, SSM might work on the BB60C.

At the time their software was very clunky and didn't have its own spectrogram recording option.  But Spike is a lot nicer, and there's probably no upside in running SSM.

[rs20]

Taking the plunge into the BB60C's API.  Anyone else on here writing custom software for it?   

Yeah, I wrote some software that takes in IQ samples and from that displays the phase of a given input tone (obviously this requires the tone to come from a TG locked to the same reference*, or else the little frequency drift would produce garbage.)

Got it nicely displaying the phase shift caused by time-of-flight difference by moving antennae towards and away from each other, and fixed antenna bouncing signals off a moving reflector pot lid. It goes without saying that this is probably not the most impressive thing a BB60C has ever done!

Some really gross, not-at-all commented code is here.

I did this on a Linux box, and encountered an issue in that the TG code didn't work at all. Managed to decode most of the protocol, while the Signal Hound folks simultaneously kindly 'hinted' the protocol to me in a funny way, which I enjoyed:

I don’t know how long it will take us to get to TG-linux compatibility, but if one were to snoop the USB bus, one would see setting internal reference out sends: ‘R’, 0x02, 0xC0. One might recognize 0xC0 as serial line internet protocol from way back.
Then one might set amplitude to maximum and observe ‘A’, 0x00, 0xC0.
One might finish by snooping to see that outputting 10 Hz sends ‘F’, 0x01, 0x00, 0x00, 0x00, 0xC0. One might then try 2560 Hz and see ‘F’, 0x00, 0x01, 0x00, 0x00, 0xC0.
One could then decide whether to wait for Signal Hound to implement Linux TG compatibility, or to forge on ahead.

The code link I gave earlier contains a basic (fixed freq CW only) standalone linux "driver" for the TG, based on the hints above.

I anticipate that the Windows drivers would work just fine on its own!

* And I actually had the TG output a 10 MHz ref, and got the BB60C to lock to that. Oddly, trying the reverse arrangement did not achieve a phase lock as far as the IQ stream was concerned.

[joeqsmith]

The range finder is really slick. Did you ever do a write up on it?

I just started to look at the API yesterday and so far the only real problem I ran across was with the bbGetDeviceType always returning a 0.  I've tried eleven other functions so far and they seem fine. 

[joeqsmith]

My BB60C in action. 

https://www.youtube.com/watch?v=f1UtLpSZ2W0&feature=youtu.be

[3db]

My BB60C in action. 

https://www.youtube.com/watch?v=f1UtLpSZ2W0&feature=youtu.be

Interesting video Joe.
Thanks for taking the time and effort to produce it .

3DB 

[bson]

Great video!
 

[joeqsmith]

Thanks! 

I may put together nice looking, more generic program for it.  First thing I want to do is get a new PCIe 3.1 USB board and see if that tames it.

[joeqsmith]

I bought a GIGABYTE GC-USB3.1 and it's been running with the extension cable and hub, not to mention I have a TV tuner on the hub that has been running non-stop as well.  It's too early to tell if it's improved. 

The one thing I noticed as soon as I installed the board and drivers is that Microsoft's USBVIEW program will no longer run.  It starts and closes.  Strange, will check into that later.


UPDATE

Downloaded the latest SDK for Windows 10 and the USBVIEW is running again.

[joeqsmith]

15 days and a lot of hours on the SH using this new card and not a single hiccup so far.   It's been long enough to say the card has made a difference.   

[joeqsmith]

I had switched  over to a different PCIe USB 3.0 card last weekend and have been using the SH daily.  Still not so much as a single problem.  It's really looking like the Dell's USB ports are just not up to snuff.   They work fine with the 2.0 peripherals.   

I have started to look at the LabVIEW communications library.  There are some things I don't like in general. Debug is off for example.  Nothing major. 

First picture showing the SH sniffing the CEM meter which was attached to the function generator.  I went through all of the decoding for that meter and there does not appear to be a way to detect when it is in AC+DC mode.  Other than that, looks like they support everything.

Second showing the raw IQ data.  You can see the 2FSK and below is using the comm lib to FM demodulate the signal.   I was doing this with my own code but this is pretty nice.  Certainly fast enough.   The spectral and eye diagrams are also in this library along with a lot of other features.   

[D3f1ant]

Does any body have a recommendation for a shipping/purchasing company in the USA that can export one of these? Shipitto will do it but their commission is a bit high (ends up around 12% Trying to avoid the rediculous markup the Au dealer puts on it.

[rs20]

I don't have any immediately helpful advice as I was fortuitously on holiday in the USA when I picked mine up. But I will point out that even with a US address to deliver the device to, they wouldn't accept a credit card with an AU billing address, so I had the wire the money instead (old school)! So just be prepared for that minor bother.

Also, be aware of import duty. I was fully prepared to pay it when I entered Australia with the thing in my bag, but the guy asked if I was ever going to sell it and when I said "no", he waived the duty. Not sure if that's actually correct, but the mail people don't tend to be so lenient.

I'll edit this post soon, I have a friend who had a US remailer that is far  cheaper than that and reliable as far as past experience is concerned.

[D3f1ant]

Yeah, that's why I was thinking of trying shipito. They have a service that handles the entire transaction, ie they buy on your behalf and then arrange shipping but they do charge as bit, although nowhere near as much as the ausi reseller. I don't understand why the resellers are adding so much margin, or why signal hound don't sell direct, they would sell a lot more product.