I have one of the original analyzers, and I'm quite satisfied with it. Note
that there is a set of drivers so that the device can be used as, say an
SDR.
There's also a nice USB power meter available from Mini-Circuits for about
$700, although I do not have one.
Don

----- Original Message -----
From: "Geraldo Lino de Campos" geraldo@decampos.net
To: time-nuts@febo.com
Sent: Friday, December 10, 2010 6:13 PM
Subject: [time-nuts] Software defined spectrum analyzer

It's neat -- not a substitute for the big HP iron, but still a very capable
tool.  The first few they sold were less expensive, but I think it's still a
nice piece of gear for $1K.  Good bang:buck ratio.

-- john, KE5FX

Hi, John,

Do you have any experience with it? Not too long ago I was in the need
to do some conducted EMC measurements in the range of 30Hz to 100MHz,
with a quite low maximum requirements, and I found that my current probe
and HP8566B combination was not good enough to conduct them (due to
filter resolution in the low side and to noise factor in the high
side...). At the end, we contracted an external laboratory for doing
these measurements usign an R&S ESU-40 measurement receiver. Do you know
how good would this unit perform as a measurement receiver?

Regards,

Javier

El 11/12/2010 04:21, John Miles escribió:

--

Javier Herrero                            EMAIL: jherrero@hvsistemas.com
Chief Technology Officer
HV Sistemas S.L.                          PHONE:        +34 949 336 806
Los Charcones, 17                        FAX:          +34 949 336 792
19170 El Casar - Guadalajara - Spain      WEB: http://www.hvsistemas.com

Hi Don.

I'm ultra curious of how they did it. How did they covered the
full BW they have.
Did you looked inside, or did they sent a block diagram?
(...I was born curious... it is not a new symptom!)

Luis Cupido.
ct1dmk.

Don Latham wrote:

Luis Cupido wrote:

the manual is on the website..
some clues

Intermediate frequencies of 2.9 MHz and 10.7 MHz are used for all
frequencies. An RF input signal near these frequencies may generate
spurious responses and degrade the performance of the image rejection
algorithm.

Chapter 5 of the manual has the whole block diagram.. basically it's a
240kHz wide sampler, with a single/double conversion front end and a
bunch of switches.  They do some clever signal processing in software
too, by stepping the LO and combining FFTs...

Hi Luis:I, too, am curious. But I haven't opened it yet. I suspect
something like an FPGA feeding a fast a/d somewhere in the 50-70 MHz
range. That is, a decomposition.
There may be a synthesized LO and mixer to get to the 50 MHz.
The info on the website says or used to say something about an sdr
modified to act as a spectrum analyzer.
One of the things I like about the device that played a part in my
purchase is the availability of a 10 MHz reference input. as a good
time-nut that shows some serious engineering attention to accuracy.
Don

Luis Cupido

--
"Neither the voice of authority nor the weight of reason and argument are
as significant as experiment, for thence comes quiet to the mind."
R. Bacon
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Page 27 of their manual at http://www.signalhound.com/manual.pdf has a block
diagram.  The first LO for VHF and up is the ADF4350 wideband synth (138
MHz-4.4 GHz, -86 dBc/Hz @ 10 kHz at 3.3 GHz).  Baseband processing is done
with an AD9864 IF digitizer.  The ADF4007 is used as a secondary divider for
the ADF4350 output, as part of their image-rejection scheme.

-- john, KE5FX

I don't have a lot of experience with it, having bought it mostly for
amusement and to see what was inside, but it does live up to its specs as
far as I could tell.  If your measurement was somehow limited by the 8566B's
10 Hz resolution bandwidth or its highpass response above 100 Hz, then the
Signal Hound would be an improvement.  It's a very competent measurement
receiver for narrowband signals.

-- john, KE5FX

https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts

--

Javier Herrero                            EMAIL: jherrero@hvsistemas.com
Chief Technology Officer
HV Sistemas S.L.                          PHONE:        +34 949 336 806
Los Charcones, 17                        FAX:          +34 949 336 792
19170 El Casar - Guadalajara - Spain      WEB: http://www.hvsistemas.com

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Good to know. Yes, for that measurements the required limits were quite
low (21dBuA from 10MHz to 100MHz), and in that case the noise factor of
the 8566B were limiting (well... one could scan that range with narrower
bandwidths but would end taking ages). For the 30Hz frequency, the limit
was higher (I don't remember the exact value), but the 0Hz spur leakage
through the 10Hz filter was high enough to be over the measurement limit
(the EMC probe was a Solar 6247-1 clamp that has an horrendous
attenuation at that range, like all probes of that kind). Since I also
had not too much time available for experimenting, we ended conducting
the tests in a nearby laboratory, where we also conducted the EMC tests
for CE compliance. We were also near the limit of their ESU-40 and their
FFC clamp probes... but at least we had some dBs of margin to measure.

Next time, it would be the perfect excuse to buy one of these toys for
amusement also :)

Best regards,

Javier

El 12/12/2010 01:25, John Miles escribió:

--

Javier Herrero                            EMAIL: jherrero@hvsistemas.com
Chief Technology Officer
HV Sistemas S.L.                          PHONE:        +34 949 336 806
Los Charcones, 17                        FAX:          +34 949 336 792
19170 El Casar - Guadalajara - Spain      WEB: http://www.hvsistemas.com

Don Latham wrote:

I'm going to guess much simpler inside.. Essentially a programmable
receiver like the Icom PCR1000, but with better bandwidth, running into
a pair of 192kHz ADCs, for which easy USB interfacing is available.

The question is really what does the 4GHz LO in the front look like?
It's almost certainly some sort of VCO PLL scheme, probably with
multiple VCOs. But, do they use a DDS?  Probably not... They probably
have a fast PLL that settles quickly so they step through in 100 kHz
steps or so.

On 12/12/2010 02:31 AM, jimlux wrote:

Two VCOs covering 1-2.2 GHz could through mixing and selecting between
sum and difference output filters (i.e. lowpass and highpass) cover that
without too much difficulty. Except for the highest end, there is a
degree of freedom in setting the oscillators to their individual
frequency which naturally could be used to avoid "bad" frequencies to
leak into the IF for instance.

Cheers,
Magnsu

Tks, folks.
I saw the diagram on the manual, albeit a bit too simplified
but ok, indeed we can have an idea.

For LO I suspect that might be one of those clock generators IC with VCO

• PLL plus a a lot of programmable dividers and really would match the
  simplified diagram they have....
  As for the digital part I have no idea, but would not be surprised of
  that very simplistic single chip approach sound card chip and an USB
  micro to feed the control bits to the PLL chip...
  It may well be a very very simple thing hardware wise.
  ...hence, I'm still curious ;-)

Luis Cupido.

jimlux wrote:

Magnus Danielson wrote:

Yeah.. but there's a whole raft of inexpensive PLL chips with multiple
VCOs on them to cover wide ranges (e.g. all cell, mobile data, and WLAN
bands).. I'll bet they're using one of them.  If a chip exists with VCOs
that does the job, that's going to be cheaper and easier than any sort
of add/subtract/mix/divide scheme.

This is an example of a new class of lab instruments.. you pay for the
hard part (the RF design and performance) and software takes care of the
rest, and since software has almost zero reproduction cost....

For today, there's probably a significant "value added" in the software,
so it might not be free... but over time, that will change.

And, if it works out nicely, standardized interfaces (like 50 ohms or
resistor color codes) will evolve, so the same software will work with a
$500 inexpensive widget or a $10,000 high performance widget from Agilent.

There will always be a market for software tailored to a specific
market/need (like standards or regulatory compliance) that will cost,
but for generic functions (like a power spectrum) that will probably be
free, or close to it.

I wonder if there's an interface for this for my iPad?

El 12/12/2010 03:41, jimlux escribió:

And the hard part is also less expensive - older signal generators used
impressive attenuators since the output accuracy depended heavily on
them, newer ones uses not-so-good ones whose characteristics are
compensated by the software (compare the attenuator in a R&S SMFP with
the one inside a Marconi 2945)

I hope so :)

Not sure about. As we are seeing, Agilent, Tek and others are trying
hard to make their instrumentation to be obsolescent, unmaintainable and
unusable in timescales as short as they can, to force replacement. Those
nice instruments running windows are basically door stops if they fail
once the manufacturer has decided to stop maintaining them. And they are
not exactly unexpensive. They will avoid as hard as they can that you
can use and old software with a new instrument, and viceversa.

In the Signal Hound web page they mention the availability of a free
API... this is also what has attracted me, because of the posibility to
write custom software for those specific  needs.  So perhaps an
interface for your iPad could be written ;)

Regards,

Javier

--

Javier Herrero                            EMAIL: jherrero@hvsistemas.com
Chief Technology Officer
HV Sistemas S.L.                          PHONE:        +34 949 336 806
Los Charcones, 17                        FAX:          +34 949 336 792
19170 El Casar - Guadalajara - Spain      WEB: http://www.hvsistemas.com

Jim: Indeed. I've a very small collection of USB instruments so far, but
it's definitely the way of the future for me. I have an older scope from
link instruments, the signal hound, and will construct one of the very
simple component bridges ( uses a dual op amp and a couple of resistors and
the sound card). As I mentioned, there's a good power meter from
Mini-circuits, and the ham radio QEX periodical has some designs for a VNA
albeit not at microwave frequenies...
All use the computer as most of the software and display. Note  that a
simple EEpc used is now about $250 and can run most of this stuff.  I'd
really like to have some '80's Hp stuff, but the buy it now folks have
sucked up all the bargains, and for me time is important.
Don

----- Original Message -----
From: "jimlux" jimlux@earthlink.net
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, December 11, 2010 7:41 PM
Subject: Re: [time-nuts] Software defined spectrum analyzer

On Dec 11, 2010, at 7:07 PM, Javier Herrero jherrero@hvsistemas.es wrote:

Then manufacturers will rise up to meet the need.
Think of all the smallish RF equipment companies out there.  Arbiter for type 4 power meters, maury, etc...

I'm a bit cynical about that...  There's been quite a few inexpensive SDRs out there that were planning a useful API, but it never materialized.

But yes, it is interesting.  As is the USRP...

On Dec 11, 2010, at 10:14 PM, "Don Latham" djl@montana.com wrote:

I think it would be possible to build a microwave front end for, say the tentec tapr VNA.  A nice quiet synthesized LO, etc

Am 12.12.2010 07:43, schrieb Jim Lux:

I have this one:  http://groups.yahoo.com/group/VNWA/

excellent to 500 MHz, quite ok to 1300. 12-term calibration,
Includes "opt. 10", TDR and generates crystal models that work.
Size is 1.7 * 4 * 2.5 inch and lives on USB power if you don't
need all 4 S-params without manual DUT-reversal.

I had of course to add a 10 MHz ref input.

regards, Gerhard

It is pretty impressive for 10 oz of weight. Thats also a reasonable price
if you did not have the big iron big power units.
Regards

On Sun, Dec 12, 2010 at 1:43 AM, Jim Lux jimlux@earthlink.net wrote: