Author Topic: RF Architecture of the Rigol DSA-815  (Read 1240 times)

0 Members and 1 Guest are viewing this topic.

Offline mcckevin

  • Contributor
  • Posts: 5
  • Country: us
RF Architecture of the Rigol DSA-815
« on: March 12, 2017, 12:34:44 AM »
Hi --

Does anyone have any information about the RF architecture of the Rigol DSA-815 series of spectrum analyzers? In particular, I'm looking for the frequency of the the first IF.

The longer story is that I'm teaching an RF course. In the linearity lecture, I discuss several ways to use non-linear effects to jam receivers. Most spectrum analyzers have an unfiltered front end and upconvert to the first IF. To jam them, you can apply a strong tone to the analyzer at half the first IF frequency which will cause the non-linear elements in the system to generate the second harmonic i.e. a constant tone in the first IF regardless of the tuned frequency. The result is the analyzer's noise floor generally raising and other kinds of havoc.

I used to be able to do this with the old HP analyzers but I'm not sure if the architecture of the newer equipment (e.g. the Rigols) exhibits this problem.

Looked around the web but didn't see anything useful.

Thanks.
Kevin
 

Offline mmagin

  • Frequent Contributor
  • **
  • Posts: 553
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #1 on: March 12, 2017, 06:08:42 AM »
Just to satisfy my curiousity I spent some time skimming the user manual:
http://beyondmeasure.rigoltech.com/acton/attachment/1579/f-0503/1/-/-/-/-/DSA800_UserGuide.pdf

While it's not a bad user manual and the price on these is quite low for a spectrum analyzer, it's pretty darned disappointing to not see at least a single page explaining the architecture of a receiver when it's sold as a piece of test and measurement equipment.  :palm:
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 297
  • Country: us
  • Older New "New Player" Player Playa'
Re: RF Architecture of the Rigol DSA-815
« Reply #2 on: March 12, 2017, 06:28:59 AM »
I did find this "performance and verification" guide. Perhaps some of the values listed here are hints toward what you want (or anti-hints: perhaps the specified frequencies to test/verify at are as far away as possible from internal IF/etc spurs).

 

Offline jjoonathan

  • Regular Contributor
  • *
  • Posts: 150
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #3 on: March 12, 2017, 06:57:05 AM »
I've looked high and low for my notes but can't find them. I can't measure again either, since I sold my DSA815 and replaced it with a higher performance boatanchor. From what I recall the first IF was around 5ghz. There were three sweeping VCOs (with LEDs to show you which one was operating if you had the RF shield off -- very cute) and a fixed PLL of 4-ish ghz to bring them up into the interleaved finger bandpass. My primary interest was to see if Rigol had left room to hack it to a higher bandwidth, but once I saw that the VCO voltage at 1.5GHz went almost all the way to the rails I decided it was unlikely.

I didn't characterize the attenuator, LNA, 1st mixer, and switches, but I wouldn't be surprised if they had enough low-pass between them to frustrate a IF/2 attack. Best of luck, be sure to let us know how it goes!
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 2175
  • Country: gb
Re: RF Architecture of the Rigol DSA-815
« Reply #4 on: March 12, 2017, 08:29:53 AM »
I can remember that there was an Eevblog teardown of the 815 a few years ago and the PCB layout suggested (to me) that the first IF is going to be about 2GHz. i.e. the hairpin filter used in the first IF looked to be the right size for an IF1 of about 2GHz. But that is just based on a guess wrt a typical PCB dielectric constant.
 

Offline mcckevin

  • Contributor
  • Posts: 5
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #5 on: March 13, 2017, 12:18:22 AM »
Thanks all --

The 5 GHz IF makes sense, especially considering Rigol makes a 3.2 GHz version of the same analyzer. They wouldn't have to change the basic conversion scheme to produce either analyzer.

I also viewed the analyzer tear down and noted the hairpin filter. Wasn't sure what it was connected to though.

I'll certainly post pictures of the screen if I'm successful! So easy to get screen captures with the Rigol.

Thanks again,
Kevin
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 2175
  • Country: gb
Re: RF Architecture of the Rigol DSA-815
« Reply #6 on: March 13, 2017, 02:42:04 AM »
If it had a 5GHz first IF and supported both the 1.5GHz and >3GHz variants the first LO would have to run from 5 GHz to over 8GHz. But if you look at the artwork for the three VCOs the highest frequency VCO only looks like a design that would run up to (maybe) 4GHz. Also the artwork for the printed LPF for the first LO would have to support >8GHz operation but the physical dimensions suggest <4GHz here. If the first IF was about 2GHz (i.e. to support just the frequency range of the 815 model) then this first LO would only have to run up to approx 3.5GHz and the artwork for the first LO LPF seems to agree with this?

The total metal length in the resonators in the hairpin filter used in the first IF is about 1.5 inches. This suggests a ballpark of 2GHz for the first IF if you assume a sensible dielectric constant for the PCB material. If the first IF is 5GHz then where is the 5GHz IF1 filter in any of the teardown images?

Note that the big (2GHz?) IF1 hairpin filter will have a re-entry mode up at around 4 or 5GHz where it will look like a 5GHz BPF but it would be lossy and probably have lots of ripple and a poor match. I believe that the job of the lumped LPF just before it is to filter away these high frequency re entry modes.
« Last Edit: March 13, 2017, 03:40:10 AM by G0HZU »
 

Offline mmagin

  • Frequent Contributor
  • **
  • Posts: 553
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #7 on: March 13, 2017, 04:51:36 AM »
Thanks for the idea, tried this out with my HP 8562A.  The effect is quite clear.
With the noise floor as the slightly lower-intensity saved trace, sweeping the full lower band, 0-2.9GHz, ref level 0dbm, everything else "auto":
First picture I input 1.95535 GHz at 0dBm .
Second picture increase to 10 dBm.  YOW!
Third picture 1GHz at 10 dBm.
Fourth picture 2GHz at 10 dBm.

Now, I suppose if you are dealing with a controlled situation and you want to make sure this situation doesn't mess up your measurements, you could increase the input attenuation to where linearity is restored, look at the full span, determine that there is this undesired interference, add a filter, etc.

But if you wanted to make a (quite expensive, but self-contained) instrument to cope with this condition without having to go for a much narrower bandwidth frontend architecture, would the most sensible thing to do be to just have two (or three!) separate frontends with different first IF frequencies and switch between them as needed (or automatically with software)?
« Last Edit: March 13, 2017, 04:58:57 AM by mmagin »
 

Offline mcckevin

  • Contributor
  • Posts: 5
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #8 on: March 13, 2017, 07:30:20 AM »
mmagin --

Yup. That's what I remember seeing on the old HPs too. The noise floor rises no matter what frequency you're tuned to. Thanks for the pictures. Would love to get the students to duplicate it in class!

G0HZU --

You've obviously walked a lot further into this forest than me. You're right that the IF is probably in the 2 GHz range somewhere. I was thinking it would be easier to build a 1.5 and 3.2 GHz version with close to the same hardware if they both went up to somewhere beyond 5 GHz but I clearly didn't think it too far through. Thanks for the info!

Kevin
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 2175
  • Country: gb
Re: RF Architecture of the Rigol DSA-815
« Reply #9 on: March 13, 2017, 08:06:22 AM »
Note that I can only guess at all of this based on the teardown images. So I might be wrong here. The best way to find out would be to play with the hardware itself. I've never had access to one of these analysers so I can't help in this respect.
 

Offline mcckevin

  • Contributor
  • Posts: 5
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #10 on: March 21, 2017, 05:10:59 AM »
Hey mmagin --

Mind if I use the pictures from your HP8562A for class? I'm guessing the first IF is 2x1.95535 GHz.

No luck on the architecture of the DSA-815A. May resort to exhaustive automated search.

Kevin
 

Offline mmagin

  • Frequent Contributor
  • **
  • Posts: 553
  • Country: us
Re: RF Architecture of the Rigol DSA-815
« Reply #11 on: March 21, 2017, 08:24:32 AM »
Hey mmagin --

Mind if I use the pictures from your HP8562A for class? I'm guessing the first IF is 2x1.95535 GHz.

No luck on the architecture of the DSA-815A. May resort to exhaustive automated search.

Kevin

Certainly, go ahead.  I think the most useful diagram, albeit such a poor scan, is from the service manual, part number 08562-90062 on page 442 (12-25/12-26 in the print numbering), I've extracted just that page -
 (attached)  The part at play here is just the low-band path of course.
 

Online ted572

  • Frequent Contributor
  • **
  • Posts: 302
  • Country: ca
Re: RF Architecture of the Rigol DSA-815
« Reply #12 on: May 15, 2017, 06:41:56 AM »
DSA815-TG Architecture:
For some useful DSA815 information for those interested in repairing a DSA815 RF or IF issue, or just understanding the RF, IF Frequencies, LO Frequencies, RF/IF active components, etc of the SA sectionhttp://www.eevblog.com/forum/testgear/spectrum-analyzer-rigol-dsa815/msg1059060/#msg1059060

This will answer questions about the IF Frequencies, Mixers, LO Frequencies, the on board LEDs that tell you which LO is currently active, etc.  This architecture is valid for all versions of the DSA815s with the different Main Boards.  I didn't provide a  schematic or block diagram (I could have) just because I didn't need any except for my personal hand sketches I made while tracing it out.  You will be able to construct it yourself if needed, as I described it clearly in order of signal flow from SA Input through the Last IF (DSP Input).

You will find all of the Front-End through the last IF (DSP) active components, diodes, and mixers (with Manufacturer and P/N) for the the DSA815 with Main Board up to v. 5.  And 'mostly all' for those with Main Board v. 7 and above.
Edit: In the last sentence above, after (DSP) I changed the word 'active' to 'active components'.
« Last Edit: May 17, 2017, 09:20:55 AM by ted572 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf

 

http://opalkelly.com/