Spectrum Analyzer - Rigol DSA815

[videobruce]

How about UHF?
Also, coverage up to 3GHz?

[G0HZU]

The same applies really...

A suitably designed RL bridge will give the best performance over the widest bandwidth and this applies for designs that operate at many GHz too. It has the advantage of better isolation from load to source as well when compared to a coupler. This consideration is likely to be more significant with the Rigol as I doubt the TGen output will have an accurate 50R source impedance.

A coupler is a very versatile thing to have in the shack. So is a RL bridge. However, with the Rigol you can only measure scalar quantities like return loss or VSWR.
For more serious use where you want to know the actual impedance you are measuring across a swept frequency range then you have to look at using a vector network analyser in place of the Rigol.

However, for most home users a low cost home made coupler or RL bridge will be fine for measuring return loss with the Rigol. You might have to make two or three in order to cover the whole frequency range of the Rigol but that's no big deal. The ZFSD coupler used by Jim covers the whole range in one go... 

[jwrtiger]

I also have the Rigol DSA815 TG and I am just a hobbyist.  I noted Jim using the directional coupler and GOHZU mentioning it was fairly decent at around $85.00.  I just went to eBay and an auction was just ending so I put in a bid and won.  The auction was for two new Minicircuit couplers with female N-connectors (ZFSD-20-5).  I got the pair for $41.00 USD.  I was wondering if anyone has a recommendation for a source for short (around 1 meter) N-connector test cables for these units?

[jimeagle]

I thought I'd provide a look inside of the Mini-Circuits unit that I am using.  Specs are available on their website and I looked for something that covers to 2GHz in order to ensure reasonable performance to 1.5GHz.  I might add that the basic concept here was taken from application notes that were available on the HP website (and probably also on the Agilent web site).

I've been in the RF business for some time working variously at SBE, KLM Electronics, Identronics/IDX, Inc (one of two pioneer RFID companies), Allen-Bradley (again, RFID), RF Technologies, and Savi Technologies (purchased by Lockheed-Martin a few years ago).  This included HF through 2.4 GHz stuff but mostly low powered devices in the VHF and UHF region up to 915 MHz.

I am semi-retired but still working at Telewave part time since their plant is only a few miles (7 minutes) from my home in San Jose, CA.

Overall I would place the DSA815-TG in the Two-Way Radio Service Monitor level of capabilities though not quite as portable but with display,  print out and screen shot saving features superior to many of them. 

Noise wise, the unit is on a par with many of the lower end ($5K) Service Monitors available today but not as good as some of the $15-30K models and not specifically set up with all site management features found on the higher end products.  Several of those that I have used over the years were more expensive than the DSA815-TG but not as good in terms of their display, screen shot resolution, and even sensitivity.

This is clearly not a unit designed with aerospace measurements in mind but it is far superior in many ways to some of the Spectrum Analyzers I've used in my 40 year career as a Technician/Engineer. 

If you've got the room (I don't), spend $1-2K more for a good used HP analyzer then you can do exotic PLL work, broadband matching, etc.  Or if you really need a Network Analyzer, you will need to spend a bit more than that.  However, be prepared for a large, heavy, and mostly non-portable unit with poor, if any, modern flash drive or usb-based control capability.  Most will require IEEE 488 bus I/O which is almost always extra.  The DSA815-TG has this option available as well.

Jim

[videobruce]

However, be prepared for a large, heavy, and mostly non-portable unit with poor, if any, modern flash drive or usb-based control capability.

But, those are deal killers. Don't forget to add, they are power hungry and have the issue of parts availability and service & shipping costs if you can't get it done yourself locally. 

Regarding the interior shot of that coupler, are those odd drilling in the four corners for tuning purposes?

[PA0PBZ]

Regarding the interior shot of that coupler, are those odd drilling in the four corners for tuning purposes?

I think it's just for the screws, looks like they get a length of aluminium and hack it in pieces to get the boxes.

[olepr01@gmail.com]

Possibly a stupid question, my apologies:

I do not fully understand how the 10Mhz ref in works, and I really cant find much about it in the documentation of the unit. If I were to connect a low noise OCXO (say Wenzel ULN, for the sake of the argument), how would the parameters of the SA change? Would for instance the phase noise figures improve? Or would it only affect frequency accuracy?

Thank you..

[G0HZU]

As I don't have access to a Rigol DSA815 I can't give you a definite answer.

But I can give you some theory if that helps?

The 1st local oscillator in the Rigol probably tunes something like 2.1 to 3.6GHz. The 10MHz reference is obviously at 10MHz.

If you crudely model the synthesiser used in the 1st local oscillator as a basic multiplier from 10MHz up to say, 2500MHz you can calculate the phase noise contribution from the  reference as 20 * log(2500/10)

= +48dB.

So the very close in phase noise of the 1st LO could be influenced by the quality of the 10MHz reference. On a lab grade analyser this is very much the case for offsets very close to the carrier eg <2kHz

So if the standard Rigol reference had a phase noise of (this is just a guess) -145dBc/Hz at 10kHz offset then its contribution to the LO phase noise would be -145 +48 = -97dBc/Hz at a 10kHz offset.

Based on this figure there would be little or no benefit in upgrading the reference in order to get lower phase noise at a 10kHz offset. This is because the phase noise of the Rigol 1st local oscillator at 10kHz offset is typically -87dBc/Hz so something else dominates the noise here. Most likely it will be due to a noisy VCO used in the LO synthesiser.

You could ask Rigol what the phase noise profile is of the standard 10MHz oscillator? eg at 1kHz, 5kHz, 10kHz and 100kHz offsets ? The figure above of -145dBc/Hz at 10kHz is just a guess at a typical low cost 10MHz oscillator. I would hope the Rigol oscillator is better than this but it probably doesn't need to be better than this

[olepr01@gmail.com]

Thanks a lot, that helped! 

[olesik]

I am thinking about buying the DSA815-TG for the sole purpose of characterizing the filters and frequency response of my circuits. I do not do any RF (yet). I would like to plug the TG to the input and look at the display to see what frequencies are contributing to the noise in my circuits, this would be to identify which(if any) IC is contributing to noise at the output. Some of these circuits are basically operating at DC levels with slow changes (for example monitoring the force of a cell contraction using a optical based force/displacement sensor) in the order of 10 to 100ms.
When you use the SA do you see the clear spike at 60hz and its harmonics ?
Thanks
Jonathan

[jwrtiger]

Think you need to look at an audio spectrum analyzer that can go down to low frequency.   The Rigol DSA815 has a 9 kHz to 1.5 GHz Frequency Range.   Someone here may be able to give you a better suggestion.

[G0HZU]

Yes the Rigol would be a poor choice for use down at low frequency because it only goes down to 9kHz and also the RBW only goes down to 100Hz and also the phase noise is so poor.

For analysing stuff down at 60Hz to a few kHz then you should really consider using a decent PC soundcard and some software that allows a spectrum analyser display via the soundcard's MIC socket using the power of the PC to do the DSP.

You can get >100dB dynamic range and resolution bandwidths below 1Hz and fast update rates as well. Many soundcards can now boast a 24 bit ADC and sampling at 192kHz allowing operation up to maybe 80kHz. So by using the soundcard you get an analyser for free if you download a free program that offers this function. There are plenty to choose from.

Or you could buy some really powerful (audio) spectrum analyser software eg from Soundtechnology.com but this won't be cheap.

http://www.soundtechnology.com/buster.html

[Rufus]

from 00.01.05 to 00.01.06 .
Did someone try to do this??
Rigol technician said it's very safe procedure because "boot" will not be touched.

Firmware updating is straight forward and the technician is probably correct saying it is safe.

The difficulty is knowing if there is any firmware update available and getting it. I updated from .04 to .05 do you know what .06 fixes?

[Rufus]

Rufus can you please explain update procedure?

You put the .sys update file on a USB stick. Find and highlight the file in the storage screens then the storage menu has a 'Sys Update' button.

[olsenn]

So does anybody even know what this 1.06 firmware fixes?

[Dongulus]

I really hope this is a quality product. $1500 for a SA is unheard of.

[olsenn]

I really hope this is a quality product. $1500 for a SA is unheard of.

I own one, and I really like it a lot. You won't find a better SA for under 2X the price

[Rufus]

       Optimize the EMI radiation.                               
     -----------------------------------------------------------
       Solve the problem of option expiration can also continue 
       to use.                                                   

What does that chinglish mean?  Sounds like trial option licenses could be abused - maybe not such a useful update.

[thekubiaks]

I like everything about the Rigol 815 but I need to go up to 3ghz and I need a tracking generator.   I primarily will use this SA for antenna design and filter design.  I'd really like to stay in this price range,  or do I have to start considering something like the Rigol DSA1030A?   I found a IFR AN1820 on ebay that might work, does anybody have opinions on either idea??  Thanks

http://www.ebay.com/itm/IFR-AN1820-Color-Spectrum-Analyzer-9-kHz-to-2-9-GHz-with-Tracking-Generator-/140917618505?pt=LH_DefaultDomain_0&hash=item20cf583749

[RCMR]

I ended up buying one of those Micronix BK Precision 2650 SAs (3.3GHz)  that have been going cheap on eBay.

Monochrome display, no tracking generator, pretty basic but it seems to work just fine and dandy for what I'm doing.

They originally cost $6K (yeah, right!) but I picked mine up for about $700.

I would have loved a Rigol, with its color LCD, TG and other niceties -- but I do a lot of work on 2.4GHz so the 815 was simply out of the question and the more capable Rigol SAs were too rich for me.

The BK Precision one also has battery operation which is nice for portable use and the LCD is very legible even in full sunlight.

[thekubiaks]

I ended up buying one of those Micronix BK Precision 2650 SAs (3.3GHz)  that have been going cheap on eBay.

Monochrome display, no tracking generator, pretty basic but it seems to work just fine and dandy for what I'm doing.

They originally cost $6K (yeah, right!) but I picked mine up for about $700.

I would have loved a Rigol, with its color LCD, TG and other niceties -- but I do a lot of work on 2.4GHz so the 815 was simply out of the question and the more capable Rigol SAs were too rich for me.

The BK Precision one also has battery operation which is nice for portable use and the LCD is very legible even in full sunlight.

Have you ever seen the RF Explorer from seeed?? 

http://www.seeedstudio.com/depot/rf-explorer-3g-combo-p-1266.html?cPath=174

I have it and it is pretty good but I'd like to have that BK2650 you picked up, especially at that price...you scored !

Anyway, I am considering the BK2650 or the HP8594 A or E model with tracking generator.  I want to test my FPV gear as well as design some antennas and band pass filters.

I hear from seeed that they are going to make some kind of tracking generator for the RF Explorer.  Perhaps I should wait before I blow huge amounts of $$$.  What do you thinK??

btw, I love your youtube videos, looks like you are enjoying the BK2650 

Cheers

[fqahmad66]

Hi everyone,

This may seem an odd question..but where ever I look at DSA815 screen shoots of RF peaks, its like side slopes concave outside..I mean like a dagger pointed upward ...rather than a sharp pointed ones with side slope concave invard. Is it normal?

Regards

[fqahmad66]

I got it....

http://rf-test.tm.agilent.com/2013/01/shape-factor-benefits-of-digital-filters/

[videobruce]

Have you ever seen the RF Explorer from seeed?? 
http://www.seeedstudio.com/depot/rf-explorer-3g-combo-p-1266.html?cPath=174
I hear from seeed that they are going to make some kind of tracking generator for the RF Explorer.  Perhaps I should wait before I blow huge amounts of $$$.  What do you thinK??

That model is out of stock, maybe becasue of the proposed tracking generator.

[fpliuzzi]

Have you ever seen the RF Explorer from seeed?? 
http://www.seeedstudio.com/depot/rf-explorer-3g-combo-p-1266.html?cPath=174
I hear from seeed that they are going to make some kind of tracking generator for the RF Explorer.  Perhaps I should wait before I blow huge amounts of $$$.  What do you thinK??

That model is out of stock, maybe becasue of the proposed tracking generator.

Hello,
The following URL may be of some help (lists network of worldwide distributors for the RF Explorer).

http://micro.arocholl.com/index.php?option=com_content&view=article&id=54&Itemid=66

I noticed that the US distributor "Nuts About Nets" shows that they have the RF Explorer-3G Combo in stock.

About the tracking generator...

Here is a direct quote by the RF Explorer developer Ariel Rocholl that I found on March 31...

"Yes, there will be a wideband RF Generator + Tracking generator but not as a module, it will be a separated unit. We will keep everyone updated as the project moves on."

I have owned my RF Explorer-3G Combo for about 6 month now and I too will be very interested in looking at the tracking generator when it arrives on the market.

Regards,
Frank