SA's: Anritsu MS2711/12E vs R&S FSH3 vs Agilent N9340B: Input wanted

[Mr Simpleton]

GOHZU, thanks for all your input on this.

(which hopefully ends the debate once and for all)

I'm not sure that it was a debate, at least certainly not on my part - I don't know enough about the subject to debate it . My interest is in trying to understand the mechanism by which it happens. My (one and only) textbook on PLLs certainly doesn't discuss it - hell, it barely has a couple of small sections on general noise considerations in the loop. I guess I'd better broaden my reading...

There is more than one way to skin a cat...    Still frequency generation, PLL, DDS and such are fun things, and when we noise-nuts get a chance we tend to jump right in head first.... Hate to admit I ended up playing with ADIsimpPLL simulation program past midnight just to see how G0HZU got his "hump".... Guess most of us do use the spectrum analyzer to look at how noise affect our signals hence we do end up in these topics, with good intentions I might add!

Back on track, choosing a protable SA, having an absolut demand for built-in battery operation, narrow things down to a few, and consensus sees to be; stay away from the FSH3. Given this track record I would be a bit hesitant to spend dollars on R&S, so why not contact Agilent for a demo??

[videobruce]

The FSH3 has been ruled out some time ago thanks to all the input here. I'd like to have a side by side of the 2712E & the 9430B. 

[videobruce]

Can someone post some screen captures from the 9340B?
Does that and the 2711/12E have the ability to invert  colors when saving a screen capture so it can be printed out w/o using a ton of black ink??

[videobruce]

The Electrician;

Another thing I notice is that the numbers for sweeps/second on the Anritsu don't make sense to me.  I set it up to look at a local TV station (a digital signal here in the U.S.) at around 600 MHz.  Using a 5 MHz span and 10 kHz RBW, the instrument reported a sweep time of 900 mS, but the update rate on the display is much faster than that; I would estimate about 50 mS.  I see this behavior constantly; the unit is much faster than it indicates.  They must mean something different than what I would expect by the term "sweep time".  The FSH4 doesn't behave like this.  It reports a sweep time that is consistent with what you see on the display.

I just had the chance to audition a Anritsu MT8212E (I believe). Surely not what I was looking for, but it gave me a basic feel. Among other things, one thing I was looking for was the lowest sweep speed it could do. Not having a manual for this (the rep was suppose to have a 2712), we (both of us) were perplexed with the readout of sweep speeds. I had the unit in manual RBW, VBW & what I thought was a manual sweep setting, but it still had a mind of it's own. It let me enter slower or faster speed speeds, but it would then display what it wanted to. usually some odd number; 83, 167ms etc.   
On top of that, when trying to enter in another number, the display went into some other setting, what appeared to be a 1 sec sweep even those the readout was around 100ms.
I'll report back on mine and his findings as he did contact Anritsu for this and other questions.

One thing I surely don't like is their manuals, they are terrible. They do not show a complete Menu tree. Only for the "File" and "System" menu entries. Nothing for all the other 'important' operational entries. You know, the ones that are used all the time. 

The fastest speed it did show was 50ms depending on settings.

[ericha]

The Electrician;

One thing I surely don't like is their manuals, they are terrible. They do not show a complete Menu tree. Only for the "File" and "System" menu entries. Nothing for all the other 'important' operational entries. You know, the ones that are used all the time. 

There are several different manuals for this instrument, due to the very large number of functions it has. You are probably looking at the "User Guide". What you want is the "Spectrum Analyzer...Measurement Guide"; here's a link: http://www.anritsu.com/en-US/Downloads/Manuals/Measurement-Guide/DWL8487.aspx.  You can find all the manuals for Anritsu products on the Library tab of the relevant instrument web page.  Here's a link to the page for the MT8212E: http://www.anritsu.com/en-US/Products-Solutions/Products/MT8212E.aspx

[ericha]

The Electrician;

I just had the chance to audition a Anritsu MT8212E (I believe). Surely not what I was looking for, but it gave me a basic feel.

The Spectrum Analyzer in the MT8212E is essentially identical to the MS2712E.  The main difference is that the MT8212E also has a "Cable and Antenna Analyzer", which is basically a simple Vector Network analyzer.  The S332E (mentioned in another post) is also similar--same Spectrum Analyzer with Cable and Antenna Analyzer, but the MS2712E and MT8212E both have options for measuring digital modulation, with the S332E does not.[/font][/size]

[ericha]

The Electrician;

Another thing I notice is that the numbers for sweeps/second on the Anritsu don't make sense to me.  I set it up to look at a local TV station (a digital signal here in the U.S.) at around 600 MHz.  Using a 5 MHz span and 10 kHz RBW, the instrument reported a sweep time of 900 mS, but the update rate on the display is much faster than that; I would estimate about 50 mS.  I see this behavior constantly; the unit is much faster than it indicates.  They must mean something different than what I would expect by the term "sweep time".  The FSH4 doesn't behave like this.  It reports a sweep time that is consistent with what you see on the display.

I just had the chance to audition a Anritsu MT8212E (I believe). Surely not what I was looking for, but it gave me a basic feel. Among other things, one thing I was looking for was the lowest sweep speed it could do. Not having a manual for this (the rep was suppose to have a 2712), we (both of us) were perplexed with the readout of sweep speeds. I had the unit in manual RBW, VBW & what I thought was a manual sweep setting, but it still had a mind of it's own. It let me enter slower or faster speed speeds, but it would then display what it wanted to. usually some odd number; 83, 167ms etc.   
On top of that, when trying to enter in another number, the display went into some other setting, what appeared to be a 1 sec sweep even those the readout was around 100ms.
<<snip>>

The fastest speed it did show was 50ms depending on settings.

This is a fairly complicated subject, but I think people reading this blog would be interested.
First, there are 2 different major modes that a Spectrum Analyzer has--zero span and non-zero span.  Let me talk about zero span first, as that's the simpler case.
In zero span, the x-axis is longer frequency, but time.  The LO in the instrument is parked at one frequency, and what you get is like an oscilloscope, except that you can tune it to different frequencies.  You see the amplitude of the signal versus time.  In zero span, the sweep time of the Anritsu handheld analyzers is very accurate, and can be as fast as 10 microseconds across the screen.

Now for the non-zero span case.
Before we get into what's going on with the instrument, let me point out that for most uses, you don't need to adjust the sweep time when the span isn't zero.  If the instrument sweep time is in it's Auto setting, the instrument just sweeps as fast as it can (for the selected "Sweep Mode", which I won't get into here).  There are some relatively uncommon cases where you do want to adjust the sweep time, however. 
The first is where you want to look at pulsed signals with a "pulse spectrum"--where the RBW is wider than 1/PRF.  In a traditional analyzer, the pulses on the screen show the PRI (Pulse Repetition Internal or 1/PRF) of the pulsed signal.  The Anritsu handheld analyzers aren't very good at this particular setup--due to the sweep time being fully synthesized (and sometimes done by FFT, but we won't get into that either here).  There's more about "fully synthesized" below. 

Fortunately, you can easily see the pulse timing by using zero span, and can see the spectrum by using narrower RBW (at the expense of "pulse desensitization").

The second time you want to adjust the sweep time in non-zero span is where you want more averaging of the signal than the VBW control can give.  The VBW is limited to 1 Hz at the bottom end, but when you are using RMS/Avg detection, you can get more averaging than this by extending the sweep time.
So while the analyzer can seem to have a "mind of its own" for sweep time sometimes, fortunately this usually isn't much of a problem for most measurements.  One can wonder, however, why the instrument works that way.  Let's talk some more about that.

In a traditional (analog) Spectrum Analyzer, there is an analog ramp generator that tunes the LO to different frequencies across the span.  As you might imagine, this isn't very accurate.  Over the years a number of clever things have been done to improve on this, including "lock and roll", first done I believe in the HP 8566/68 analyzers.  In this case, the start (or possibly other) frequency is synthesized, and this is used as the beginning voltage/current for the sweep.  As I remember in the 8566/68, narrower spans had the frequency stabilized by using a discriminator, but I don't remember the details.

The Anritsu handheld Spectrum Analyzers take a different approach.  Every frequency point is fully synthesized.  This makes the frequency accuracy exceptionally good.  But it makes the sweep time very complicated.  This is because the time it takes for a synthesizer to lock up & settle is not precisely known beforehand--it depends on the frequency and phase of the VCOs in the system when the process starts, something that is somewhat random.  On top of that, the fastest sweep time is a function of many more things than you might imagine--not only the RBW & VBW as in traditional analyzers, but also the detector settings and the frequencies being used.  One of the reasons for this is the complexity of the L.O. needed to be able to tune over a wide frequency range (GHz) with very fine frequency resolution (effectively Hz), as well as really good close-in phase noise.  The 1st L.O. in these instruments is much more complex than the simple single-loop structures that have been discussed elsewhere in this thread.  Another reason for the minimum sweep time being a complex function is something called "spur avoidance".  There are certain frequencies where spurs would be bad for the way the LOs normally work; at those frequencies the 1st & 2nd LOs can be tuned to slightly different frequencies than would be normally used, making the tuned signal appear at the right place in the IF, but making the spurs go outside of the IF bandwidth.  This is pretty clever, as it allows wide IF bandwidths (which tend to cause more spurs), but it takes time.

Another complicating factor is that because the RBW and VBW are done digitally, there is somewhat limited control of exactly how fast the sweep can be.  The instrument can dwell at a particular point to get 2 or 3 points out of the RBW filter, but you can't get 2.5 points.  So some sweep times just aren't possible given the way the instrument works.

To further complicate this, the method of dealing with this variability of the sweep time in the Anritsu handheld analyzers has changed somewhat over time.  To have a precise discussion, we need to know what version of firmware for the instrument we are discussing.  The latest firmware for the MT8212E is V1.53, introduced just last month.  In this version (and several earlier ones), when you enter a sweep time, the instrument uses a model of how long the sweep will take for a particular instrument setup.  This is characterized for thousands of setting combinations, and the closest one is picked.  Then it calculates the closest multiple of that time to the user-selected sweep time, and extends the dwell time at each frequency point by that factor.  That estimated value is put on the instrument display, preceded by a "~".  The instrument then actually measures it's own sweep time, and displays the actual value.

Now we have to talk about the definition of "sweep time".  On many analyzers, if you have it set up to tune across a wide frequency range, if you use a stopwatch to time the sweep, you will find that the actual sweep time doesn't match what's displayed on the instrument.  How can this be?  There are points during the sweep when the instrument isn't sweeping, but doing something else; the most common of these is switching bands, at least for microwave analyzers.  So for these analyzers, the "sweep time" isn't the total time, but rather the time the instrument is actually sweeping.  In the Anritsu handheld Spectrum Analyzers, however, the measured sweep time is the time for the total sweep, including all of that non-sweeping time.

Another point worth commenting on is the need for sweep times faster than about 100 ms when the span isn't zero. Two points come to mind about this.  The first is that for many setups, you won't be able to get anything near this time.  This is because larger spans and smaller bandwidths take longer--that's just physics.  There are some tricks that can help (like using an FFT), but they only go so far.  When you have 1 Hz RBW, it's going to take at least 1 second, no matter what you do.  If this span is large with a narrow RBW (like 1 Hz), it's going to take many seconds to get a complete trace.
The second is that the human eye can only see so much--update rates much faster than 10/second are nice, but it's hard to see much of what's going on.   There are cases where faster sweeps can be helpful, such as when trying to capture a brief pulse--but there may be other ways to do that--but that's probably a topic for another thread.

I hope that helps clarify what's going on in the Anritsu handheld spectrum analyzers, as far as the Sweep Time is concerned.

[videobruce]

ericha;
Thanks fore the links, but the 1st one is dead. Also, watch your quotes, the close quote got misplaced, you quoted your whole post. 
Their "library" tabs have too long of a list, hard to find anything quickly. I'll look closer.

[ericha]

ericha;
Thanks fore the links, but the 1st one is dead. Also, watch your quotes, the close quote got misplaced, you quoted your whole post. 
Their "library" tabs have too long of a list, hard to find anything quickly. I'll look closer.

Sorry about the link--there is an extra period at the end that shouldn't be there.  Here is the correct link: http://www.anritsu.com/en-US/Downloads/Manuals/Measurement-Guide/DWL8487.aspx

Thanks for the hints about quotes:).

[Marcel]

Hello all,

Rohde Schwarz FSH-3 and 6 power on problems?

Replace NDC651N N channel fet or replace the NDC651N with the SI3454ADV-T1-E3 (pin 6 bent up and over to pin 5)
 
Best regards,

Marcel