Author Topic: Spectrum Analyzer - Rigol DSA815  (Read 615623 times)

0 Members and 2 Guests are viewing this topic.

Offline NukerDoggie

  • Contributor
  • Posts: 42
Re: Spectrum Analyzer - Rigol DSA815
« Reply #50 on: May 03, 2012, 03:30:57 am »
I have started a new thread for the EMC stuff:

https://www.eevblog.com/forum/projects-designs-and-technical-stuff/the-trials-and-rewards-of-emi-emc/

I hope people will follow my progress there - I'll try to make it worthwhile to read.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Spectrum Analyzer - Rigol DSA815
« Reply #51 on: May 03, 2012, 04:36:44 am »
Well, I've gone on the Rigol-USA waiting list, hoping to nab a DSA815-TG in about two weeks when they expect a shipment to arrive from the factory. Right now everyone is sold out of the instrument!

FYI, Rigol have promised me a DSA815 for review maybe this month, but yes, they seem to be hard to get at present.

Dave.
 

Offline cmatten

  • Newbie
  • Posts: 6
Re: Spectrum Analyzer - Rigol DSA815
« Reply #52 on: May 03, 2012, 10:42:36 am »
Well, here's the response from Rigol on the DSA815:

See attached pdf from Rigol.

Looks pretty decent to me - what do the rest of you think?

The attached pdf shows the following (@ 750MHz)
RBW        Noise
100Hz     -94dBm
1K           -82dBm
10K         -73dBm
1M          -54dBm

Allowing for the 10dBm input attenuation, these figures are around 10db lower than expected.
Stated specifications
0 dB RF Attenuation, RBW=VBW=100 Hz, Sample Detector, Trace Average ? 50
1MHz to 1.5GHz  <-110 dBm+6 x (f/1GHz) dB, typical -115 dBm

Spec -110dBm + 6 x (.75/1.5 ) = -113dBm

Am I missing something?
 

Offline NukerDoggie

  • Contributor
  • Posts: 42
Re: Spectrum Analyzer - Rigol DSA815
« Reply #53 on: May 03, 2012, 03:52:04 pm »
Hi Dave - that's great news about getting (eventually) a DSA815 for review and teardown!

My DSA815-TG is supposed to ship today from Tequipment.net. They have stock and the item has been pulled from stock for shipping. Its ground - so I should have the unit by Tuesday or Wednesday next week!
 

Offline cmatten

  • Newbie
  • Posts: 6
Re: Spectrum Analyzer - Rigol DSA815
« Reply #54 on: May 04, 2012, 01:19:29 am »
Well, here's the response from Rigol on the DSA815:

See attached pdf from Rigol.

Looks pretty decent to me - what do the rest of you think?

The attached pdf shows the following (@ 750MHz)
RBW        Noise
100Hz     -94dBm
1K           -82dBm
10K         -73dBm
1M          -54dBm

Allowing for the 10dBm input attenuation, these figures are around 10db lower than expected.
Stated specifications
0 dB RF Attenuation, RBW=VBW=100 Hz, Sample Detector, Trace Average ? 50
1MHz to 1.5GHz  <-110 dBm+6 x (f/1GHz) dB, typical -115 dBm

Spec -110dBm + 6 x (.75/1.5 ) = -113dBm

Am I missing something?

Hi

Looking at PDF,Reference level is 0dBm
We have to consider,what if decreasing reference level?

I want to see the measurement result,like SSG -110dBm sinewave
reference level -60dBm
span 10kHz
other setting auto

Noise floor level is not flat anyway,so high frequency's DANL is getting worse,I think.

Can someone with a DSA-815 check what they see? Make sure the input is terminated with a 50ohm dummy load.


I'm thinking about getting one of these, I'm a little concerned about the DANL.

My 20 year old HP8921A shows -119dBm @ 1KHz RBW (RBW is fixed based on span, for spans less than 1MHz RBW is 1KHz) with AVG set to 10. Other spans/RBW show expected values.

UPDATE, I think I found the reason for the DANL being higher (worse) than expected, the detector is set to peak (top right hand corner under Rigol /Status). I think this should be set to sample or rms. I think trace averaging is also turned off. The user guide on page 1-17 has AVG 1 shown just above the grid in the centre. For the screenshots provided by Rigol, there is no mention of Averaging either in the text or on the screenshots.

The spec also says Sample Detector and Trace Average of >50.

Looking forward to the review.
« Last Edit: May 04, 2012, 01:52:52 am by cmatten »
 

Offline cmatten

  • Newbie
  • Posts: 6
Re: Spectrum Analyzer - Rigol DSA815
« Reply #55 on: May 04, 2012, 08:40:34 am »
Soon somebody up the good info,but at the moment,only this site's measurement result 49.5MHz,-129.68dBm is predictable the performance a bit.

http://www.rigol.com/prodserv/DSA800/

Looking at this site,I decided to order DSA815-TG.

The setting of measuring 49.5MHz,-129.68dBm is

reference level -80dBm,average 3
span 5kHz
RBW 100Hz
VBW 1Hz
Sweep time 50s

Sweep time 50s is acceptable for me,merely to measure the small sinewave signal for me.
If VBW changes 10Hz,a bit noise floor raise,but it will be -130dBm like that.
Considering the price,I think it is not bad.

That looks a better. I thought is was due to the 10db attn, using peak detector (note they use PAvg in this example), trace averaging.

 

Offline cmatten

  • Newbie
  • Posts: 6
Re: Spectrum Analyzer - Rigol DSA815
« Reply #56 on: May 04, 2012, 09:55:47 am »
This example,Attn 0dB.
But,even if the internal attenuator is set up 10dB,almost noise floor is same,because the ordinary SA's IF amplifier automatically compensate it.
My R3361B can't set up 0dB attenuation actually.only from 10dB.
It is rather safe.

I don't know the difference of noise floor between peak-detector and sample detector,sorry.
I usually only use R3361B's normal detector,I think it is peak detector.
Even though R3361B is old one,but it has quasi-peak detector up to 3.6GHz,but
I am HAM,so I only use as my hobby experiment.
If DSA815-TG is no problem,I will release R3361B.It is too heavy.LOL

Now,I am waiting for DSA815-TG.

Everybody maybe notice that,TG level is only from -20dBm to 0dBm.
Usually TG level of SA  is from-50dBm to 0dBm.

My concerns is rather Tracking generator's frequency accuracy,whether we can manually change TG's frequency to just center frequency,or TG's frequency is no problem,once calibrating.

Anyway,when I receive it,I will measure my homebrew 7MHz BPF(BW 100kHz,-6dB,insertion loss 5.5dB).
This BPF is already get the results,so I can check the performance  of DSA815-TG.

I'm also a ham (VK2KCM), for the price the DSA-815-TG seems to be the best SA I've come across. I was looking at the getting a second hand HP unit, however I like the large colour screen, lan access, being able to save screenshot to usb that this unit has.

My HP8921A has 40,20 and 0db input attn. The noise level increase by 20 and 40db when the attn is set to either of these.
The units noise level is still the same, however it need to add the input attn to the reported level.

For the different detector type, check the DSA-815 userguide on page 2-25.
« Last Edit: May 04, 2012, 10:37:21 am by cmatten »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Spectrum Analyzer - Rigol DSA815
« Reply #57 on: May 04, 2012, 10:20:02 am »
Hi Dave - that's great news about getting (eventually) a DSA815 for review and teardown!

My DSA815-TG is supposed to ship today from Tequipment.net. They have stock and the item has been pulled from stock for shipping. Its ground - so I should have the unit by Tuesday or Wednesday next week!

Mine will take some time because they have to find a "spare" unit. I guess they can't just take one out of stock and ship it?
Anyway, they have promised that, and a DS4000 sig gen "mid to late May".
Then one of the new 2000 series scopes when they are available.

Dave.
 

Offline olsennTopic starter

  • Frequent Contributor
  • **
  • Posts: 993
Re: Spectrum Analyzer - Rigol DSA815
« Reply #58 on: May 04, 2012, 12:59:52 pm »
Quote
Mine will take some time because they have to find a "spare" unit. I guess they can't just take one out of stock and ship it?
Anyway, they have promised that, and a DS4000 sig gen "mid to late May".
Then one of the new 2000 series scopes when they are available.

Dave.

I think this just goes to show, Dave, that even though you (and your forum) are largely responcible for forcing Rigol to sell their DS1102E for the price of the DS1052E, you have given Rigol more good publicity than any advertisement on a commercial website.

I can't wait to see what my DSA815's insides look like... and by that I mean I can and will impatiently wait to see what your DSA815's insides look like.
« Last Edit: May 04, 2012, 01:06:47 pm by olsenn »
 

Offline olsennTopic starter

  • Frequent Contributor
  • **
  • Posts: 993
Re: Spectrum Analyzer - Rigol DSA815
« Reply #59 on: May 05, 2012, 10:33:21 pm »
Another nice thing I just wanted to point about about this thing which i hadn't realized at first, is that it includes a 1Hz resolution frequency counter functional up to its full bandwidth of 1.5GHz.
 

Offline cmatten

  • Newbie
  • Posts: 6
Re: Spectrum Analyzer - Rigol DSA815
« Reply #60 on: May 06, 2012, 07:30:20 am »
Another nice thing I just wanted to point about about this thing which i hadn't realized at first, is that it includes a 1Hz resolution frequency counter functional up to its full bandwidth of 1.5GHz.

Lock the SA to a GPS reference using the 10MHz input and you have a very accurate frequency counter.
 

Offline olsennTopic starter

  • Frequent Contributor
  • **
  • Posts: 993
Re: Spectrum Analyzer - Rigol DSA815
« Reply #61 on: May 06, 2012, 09:29:23 pm »
Quote
Lock the SA to a GPS reference using the 10MHz input and you have a very accurate frequency counter.

I actually own a 10MHz rubidium frequency reference, so I'll just use that if I need a precise frequency reading
 

Offline NukerDoggie

  • Contributor
  • Posts: 42
Re: Spectrum Analyzer - Rigol DSA815
« Reply #62 on: May 10, 2012, 02:19:45 am »
My DSA815-TG arrived today from Tequipment.net! It looks pretty good sitting there in my lab. In a day or two I'll warm it up and start getting familiar with it. My first impression after holding it in my hands and doing a physical examination:

This is a very good quality instrument, quite heavy, very well made, attractive - professional. By no means a "cheap" imitation of a quality western-made instrument. Rigol - hats off to you! And many thanks to Tequipment.net for snagging one for me. (they're not paying me anything or giving me any discount for promoting them, by the way).
 

Offline Attorney

  • Regular Contributor
  • *
  • Posts: 69
Re: Spectrum Analyzer - Rigol DSA815
« Reply #63 on: May 11, 2012, 03:19:30 pm »
Getting in way late on the discussion.  Someone earlier mentioned the Signal Hound.  I looked at it and it would probably do fine for my purposes since it's got good RBW characteristics and I do a lot of experimental oscillator work.  Where it falls short is in SPAN.  Really falls short if you need a window for wideband measuements greater than about 200 kHz.  Beyond 200 kHz, it goes into a special "fast scan" mode that disables the actual I/Q receiver and uses just an IF power detector -- good up to 200 MHz (with a 1.5 s scan time...).  As you go past 200 MHz, the RBW becomes worsens (5 MHz) and the scan time increases more.  The Signal Hound is a true direct RF-sampled SDR receiver and is probably better characterized as a narrow-spectrum analyzer.

I need RBW better than about 30 Hz.  I looked at the high-end Rigol which has 10 Hz RBW.  Looks like a great analyzer but with TG, price is about USD $6K.  I went back & forth whether proceed with a used Agilent (N1996A) with nearly identical specs and features -- or the Rigol.  I ended up purchasing a 6 GHz Agilent since I'm unfamilar with Rigol's long-term after-sale support, and the Agilent does have a slight edge in some performance areas like screen detail where it's XGA, versus SVGA resolution.  Now, in terms of actual sampling limitation, the screen may be a non-issue.   
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: Spectrum Analyzer - Rigol DSA815
« Reply #64 on: May 13, 2012, 01:41:07 pm »
(The condition,when I checked the noise floor)

Thanks for posting the information and observations.

Initially it wasn't clear to me if VSWR was a built in feature or if it came with the VSWR bridge kit. Updates to vendor web sites indicated the VSWR option is $545 and you need a bridge ($705 for their VB1020), hopefully the firmware option would work with other bridges.

For that kind of money this http://sdr-kits.net/VNWA3_Description.html has much more utility over a bit more limited frequency range.

I would imagine 50/75 ohm is a built in feature for use with an impedance adapter (as supplied in the utility kit). When you select it do the numbers or trace move by a couple of db?

Rigol seem to have copied Agilent's X 2000/3000 pricing policy with a low base price and expensive firmware options. The AMK+EMI+VSWR options more than double the price of the unit.

I will be getting an 815-TG soon.
 

Offline olsennTopic starter

  • Frequent Contributor
  • **
  • Posts: 993
Re: Spectrum Analyzer - Rigol DSA815
« Reply #65 on: May 13, 2012, 02:15:49 pm »
Quote
About demodulation function...
I tried to use it as FM radio with FM antenna,but I can't receive even the strongest signal station here.
Usually I can receive FM radio with R3361B,setting up RBW and VBW etc.
The sensitivity is quite bad,or my function is broken?LOL
Maybe some made-up FM strong modulated signals,like set up by DG4000 series something can be decoded,but as FM radio,it has problem,I think.
If anybody can receive FM station with FM ANT,plz let me know.

I have no problem picking up FM stations in my area (NS, Canada) with just a plain piece of wire as an antenna. Make sure you set the center frequency to the desired station and adjust the span to 10KHz or whatnot, and also, just in case you forgot, make sure to turn the headohone output on and the volume way up.

In reference to the 50/75 ohm selection, the input is fixed at 50 ohms, so changing the software to denote 75 ohms only adjusts what's on the display to match what WOULD appear if it was in fact 75 ohms... this is useful if you are using a 50 - 75 ohm imedance adapter. Changing this setting is very similar to changing the probe attenuation factor on an oscilloscope.

I would highly recommend picking up a good 20db - 30db attenuator to leave attached to the input... you don't want to overpower this thing!
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: Spectrum Analyzer - Rigol DSA815
« Reply #66 on: May 13, 2012, 03:41:51 pm »
I don't know 3.0,but I am using VNWA2.6,and It can measure S21 etc,and of course,
connecting ANT,and it can calculate S11,and other directly put out VSWR etc info.
Now only complete one (not kit) is sold,so a bit expensive,but VNWA is also so educative.
it is worth to get it.

The 3.0 is not a big improvement over previous versions. It costs $20 more than the DSA815 VSWR firmware option so I don't think it is expensive.
 

Offline olsennTopic starter

  • Frequent Contributor
  • **
  • Posts: 993
Re: Spectrum Analyzer - Rigol DSA815
« Reply #67 on: May 14, 2012, 01:12:27 am »
Quote
And maybe it is important for you and me...There is no PC software in the original CD.
I expected it has PC software...but no PC software.
I think,we can use Ultra Spectrum,but not sure.

You can obtain IVI drivers and LabView drivers/examples here: http://int.rigol.com/support/ivi/more.html
VISA commands are also accepted by the device and can be found in the programming guide/manual on Rigol's site as well.

If you hear anything new about UltraSpectrum please share it!
« Last Edit: May 14, 2012, 01:17:18 pm by olsenn »
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: Spectrum Analyzer - Rigol DSA815
« Reply #68 on: May 16, 2012, 10:08:54 am »
(The condition,when I checked the noise floor)

Result from mine with the same setup
 

Offline cliffyk

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: us
    • PaladinMicro
Re: Spectrum Analyzer - Rigol DSA815
« Reply #69 on: May 16, 2012, 07:05:38 pm »
Hello all,

I am a newcomer to your group and like several of you have just purchased the DSA815-TG spectrum analyzer, for me a replacement for a 7+ year-old Instek GSP-810; that I was fortunate enough to be able to sell on eBay for US$1300, greatly subsidising the cost of the DSA815.  I find it to be a very capable instrument, far superior to the Instek which cost $2500 when new.

Also like another poster I too have a rubidium standard and thought I would post my findings.

Here's a shot of the test setup:



And the 10 MHz reference at 2 kHz span:



You can see that the peak marker reads 9.999973 MHz, 27 Hz low, however in calculating the instruments marker uncertainty according the for formula presented in the specifications:

± frequency indication × frequency reference uncertainty + 1% × span + 10% × resolution bandwidth + span/(# points -1)

where:
   
    frequency indication = 9999973 Hz
    frequency reference uncertainty = 1e-6 (it's brand new)
    span = 2000 Hz
    resolution bandwidth = 100 Hz
    # points = 601 (from the instrument specs)

I get: 9999973 * 1e-6 + 0.01 * 2000 + 0.1 * 100 + 2000 / (601 - 1) = ± 43 Hz, making the -27 Hz error well within spec.

I tested several other span/rbw combinations and found the marker readouts to all be within spec...

=================================================================

Swallowtail333, I would be very interested in how well Ultra Spectrum works with the '815; please let us know.

I have also contacted Rigol to ask if it would be possible to add an "inverse" display function in any upcoming firmware update as I feel a "black on white" display would be much easier on the eyes than the black screen--which I find to be an odd emulation of one the technological limitations of CRT technology.

This would be much easier on my eyes:



It reminds me of the control software provided with a DDS signal generator I obtained some time back, the publisher went so far as to create virtual legacy style frequency range selection buttons--on a fully digital instrument?:

« Last Edit: May 17, 2012, 01:49:58 am by cliffyk »
-cliff knight-

paladinmicro.com
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: Spectrum Analyzer - Rigol DSA815
« Reply #70 on: May 16, 2012, 08:32:41 pm »
    frequency reference uncertainty = 1e-6 (it's brand new)

Measure the 10MHz ref out and you will know. Mine looked 20Hz out although I would have to check the counter calibration to be sure.
 

Offline cliffyk

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: us
    • PaladinMicro
Re: Spectrum Analyzer - Rigol DSA815
« Reply #71 on: May 17, 2012, 01:21:59 am »
I had meant to do just that however I got sidetracked.  Here is the 10 MHz rubidium standard (Input A) and the 10 MHz clock output of the DSA815 (Input B) being compared (A - B) via an Iwatsu SC-7202 counter.



You can see the DSA815 is running 23.5 Hz (2.35 ppm) faster than the standard--very close to your observation, and slightly higher than the aging rate though no "factory calibration" specification is available, that I can find.  Correcting my calculation results in:

9999973 * 2.3e-6 + 0.01 * 2000 + 0.1 * 100 + 2000 / (601 - 1) = ± 56 Hz

Here is the clock error, again using the Iwatsu counter but with the inputs reversed; A is the DSA815 internal clock out, B is the standard; displayed as A / B:



Correcting the marker display for the known clock error = 9999973 * 1.00000235 = 9999996.49**, leaving 3.5 Hz to processing issues within the DSA815.


So now I'm on a roll, let's use the rubidium standard as the clock input for the DSA815 (which BTW worked quite nicely with the standard's 0.55 VRMS @ 50? output, split through a Mini-Circuits -3.1 dB unit [0.38 VRMS]), and compare the TG output @ 10 MHz to the rubidium standard:



The 10 MHz standard is feeding Input A and the DSA815 clock in, the DSA815 TG is feeding Input B, the counter is set to display A - B with a 10 s gate--you can see there is 0.0 Hz error.

In the last here is the spectra of the 10 MHz rubidium standard, captured using the same signal as the clock input to the DSA815--note the peak marker frequency of 9.999996 MHz.  Curiously close to the the corrected marker display** calculated above is it not?



« Last Edit: May 17, 2012, 01:43:20 am by cliffyk »
-cliff knight-

paladinmicro.com
 

Offline cliffyk

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: us
    • PaladinMicro
Re: Spectrum Analyzer - Rigol DSA815
« Reply #72 on: May 17, 2012, 04:22:49 am »
I don not mean to be boring, however as compared to my old Instek GSP-810 this is fun:

Here's a trace of my pirate FM station, with the 19kHz stereo pilot only, grabbed with the DSA815, and using the rubidium standard as the clock source:



Here is a similar shot from the GSP-810 app I developed over the years:



I hope to create a new instrument interface for the GSP-810 application so that I might use it with the DSA815...
-cliff knight-

paladinmicro.com
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: Spectrum Analyzer - Rigol DSA815
« Reply #73 on: May 17, 2012, 08:32:54 pm »
You can see the DSA815 is running 23.5 Hz (2.35 ppm) faster than the standard--very close to your observation, and slightly higher than the aging rate though no "factory calibration" specification is available, that I can find. 

Measuring its own 10MHz out with 100Hz RBW & VBW the peak function always put a marker on exactly 10.000000MHz. The marker counter function mostly showed 9.999999MHz occasionally jittering -1 to +2 Hz.

My counter and a function generator with TXCO were calibrated against GPS (a Thunderbolt) a few months ago, the counter thinks the function generator is 0.06Hz high at 10MHz so I think the cal is probably still good. The counter thinks the DSA ref out is 20.2Hz high.

One thing I noticed, turning on the TG shifts the display and all measurements by 150Hz - that looks bugged to me.
 

Offline cliffyk

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: us
    • PaladinMicro
Re: Spectrum Analyzer - Rigol DSA815
« Reply #74 on: May 17, 2012, 11:55:38 pm »
I have not yet measured the DSA815's internal clock out fed back into its input, my tests were comparing the TG output with the rubidium standard's output--with the standard being applied to the DSA815's external clock--sounds like yet another interesting test though. 

I also observed a shift when the TG was switched on, however far less that the 150 Hz you recorded, the shift on my instrument was only in the order of 2-3 Hz...
-cliff knight-

paladinmicro.com
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf