Agilent E4406A Vector Signal Analyser

[G0HZU]

In the case of version 12 of the 89600 SW you just download it and run it and select the option of a 14 day trial and it will self start the trial immediately even if you are offline. You don't need to get authorisation or a key from Agilent to self start the trial.

If you connect an instrument via LAN or GPIB to the PC it should auto detect the scope/analyser and if it's a supported instrument then you are good to go

Agilent dropped the E4406A support after version 10 but you can run a registry fix that means it supports the E4406A up to V12.

There is no licence required to be inside the E4406A analyser. The SW is licensed inside the PC and it should work with any hardware that is supported. I think the one licence covers all hardware because the trial doesn't know I was going to try connecting an E4406A analyser. If I had a different model analyser or scope here then it will detect it and run with it if it is in the list of supported instruments.

I can't comment on what scopes the 89600A SW supports but I can imagine it would be a very powerful tool with the right scope

[bingo600]

While I had the covers off I had a closer look at the block diagram and the various coax cables between the modules. There is the ability to feed a signal directly into the digital IF at baseband if a cable is disconnected.

The appeal here is that the system phase noise should be very low. Much lower than any of my mid range lab grade sig gens can achieve so they are useless as a test signal here. So I quickly tried a homemade 7.5MHz free running LC oscillator with a loaded Q of about 30. This should be capable of <<-150dBc/Hz at 10kHz offset.

I tried this (via an anti alias LPF) in the E4406A and the E4406A limited out at -136dBc/Hz at maybe 3-10kHz offsets. Not bad! The idea would be to use an external LO and mixer to make a fairly decent phase noise measuring capability if a coax relay was fitted here. Note: My E4406A has the 14bit IF module which will help here.

The external LO could be something like an AD9912 or AD9914 DDS that can deliver very low phase noise. This is still way off what something like a E5052A SSA can achieve in terms of phase noise floor but it's a big step up from a decent lab grade analyser in this respect.

I also had to play inside the advanced FFT menus to get the last few dB of performance when so low down in signal level.

The plot below is on 15dB/div not 10dB in order to get it on the screen. The Err logo on the screen is because the analyser is not happy that I broke the signal path to allow this signal to be fed into it. So it failed to complete its regular self test signal routine.

G0HZU

Could you elaborate some more on this setup ?

I have a E4406 and this is quite interesting.

Have you seen
http://www.ke5fx.com/gpib/readme.htm
http://www.ke5fx.com/gpib/pn.htm

I seem to remember that John mentioned -140dB wo. modifications in one of his posts.

/Bingo

[G0HZU]

G0HZU Could you elaborate some more on this setup ?
I have a E4406 and this is quite interesting.

All I am doing is injecting a signal at the final IF (7.5MHz) direct into the digital IF. This has a 14bit ADC clocked at 30MHz.

My E4406A has the same analogue and digital IF modules (and CPU module) as the £££ Agilent PSA. But the phase noise performance is limited by the quailty of the local oscillators ahead of this final digital IF.
So by bypassing the front end converters I can get very low phase noise performance.

Obviously, I need to provide my own LO and mixer because I don't just want to look at 7.5MHz... So I can use a high performance DDS as an external LO with an external mixer and feed in direct at the 7.5MHz digital IF via a screened coaxial relay switch.

Look at my earlier plot. The phase noise 3kHz from the carrier is about -136dBc/Hz. This is about 24dB better than my legendary HP8568B at a similar offset. (or maybe 55dB better than a Rigol DSA815 at this offset!)

The standard E4406A can achieve -140dBc/Hz performance but only at offsets of several hundred kHz.

[edavid]

It's worth mentioning that some E4406As have the B7C baseband I/Q input option, so they don't have to be modified for this (or they are pre-modified).

[G0HZU]

That's a good point...

The B7C IQ baseband option is a highly desirable (but rare and expensive) option that requires a complete new card/module inside the analyser.

The B7C IQ baseband option is much more powerful than the simple fudge I am doing above because it presumably has dedicated ADCs in the addon card for each of the front panel I and Q baseband inputs. These two BNC inputs are usually seen as being blanked off in the top left corner of the front panel of a regular E4406A. I would really like to have this option but it is very rare and an E4406A with this option usually commands a much higher price

[davorin]

The screenshots from the Agilent Windows Software....is this freely available and just connects via TCP/IP or GPIB to the E4406?
Or does it need something else?

[PA0PBZ]

The screenshots from the Agilent Windows Software....is this freely available and just connects via TCP/IP or GPIB to the E4406?
Or does it need something else?

The link is in the thread, see reply #17. It's a 14 day time trial but there's ways...
I used TCP/IP to connect to the E4406A

[davorin]

Must have missed that one from all those nice screenshots (o;

So would you recommend this instrument for analyzing/Debugging MiWi, WLAN and other Radio signals?

[PA0PBZ]

I'm using it to play around with the various signals and modulation techniques and I'm still learning so I'm not qualified to do any kind of recommendation. However, since both the hardware and software are quite old I guess you should try to find something more modern if you are going to use this in a business, like the TEK RSA306.

http://thesignalpath.com/blogs/2014/11/04/tektronix-rsa306-usb-real-time-spectrum-analyzer-review-and-experiments/

[davorin]

Well I'm planning to use it privately....and secondly...the Tektronix is way out of Budget (o;

Would fit perfectly with my other "old" Equipment ;-)

[KJDS]

Well I'm planning to use it privately....and secondly...the Tektronix is way out of Budget (o;

Would fit perfectly with my other "old" Equipment ;-)

YHM

[davorin]

Hmm...people mention here that the SA part span is limited to 10MHz....
But on this Youtube video I see it set to 100MHz span....

[Wuerstchenhund]

Hmm...people mention here that the SA part span is limited to 10MHz....
But on this Youtube video I see it set to 100MHz span....

That's with the W-CDMA option. Other options and Basic mode are limited to 10MHz.

[davorin]

Makes perfectly sense then....the "W-" (o;

Still a nice and fast device...seems it is also running HP-UX 10.x?

[Wuerstchenhund]

Still a nice and fast device...seems it is also running HP-UX 10.x?

HP-UX/RT, an embedded (cut down) variant of HP-UX 8.x or 9.x if I remember right. The firmware console is similar to later HP9000 PA-RISC workstations, i.e. the 712 Series. pr the 744 VME controller (which is basically what the controller in this thing is).

It's the same as Agilent's old high end Spectrum Analyzer series PSA (the E4406A is essentially a cut down PSA).

[davorin]

Just made a little contact with HP-UX today on my 16702A (o;
Bad it requires to boot from another HP-UX machine...

The onscreen drawing updates are quite fast for an older machine....dunno about the remote software tool...
But seems a better choice than using my CRTUs for analyzing modulation....

[Wuerstchenhund]

Just made a little contact with HP-UX today on my 16702A (o;

HP-UX was great, I spent a lot of time with it on the larger servers (K-Class, R-Class) and supercomputers (V2500/V2600). It had a nice utility called SAM (System AdMinistrator) which made the basic setup pretty simple.

The onscreen drawing updates are quite fast for an older machine....dunno about the remote software tool...
But seems a better choice than using my CRTUs for analyzing modulation....

Well, both E4406A and CRTU were designed for cell phone testing but they are different devices used in different scenarios. The E4406A is a great signal analyzer but a pretty bad spectrum analyzer. The CRTU is a better spectrum analyzer, plus it's a RF generator, Protocoll Analyzer (under Windows2000) and automatic tester, but it's not a great signal analyzer.

[davorin]

Well would be a great combination to get a hand on a E4406A as well....maybe I sell one of my 2 CRTU-RUs even (o;

Hmm..I was more back then the SunOS 4.1.3 guy....configuring SunISDN card and had a hard time to find a provider who supported HDLC protocol for dial-up internet connections...and at workplace at CompuServe Switzerland (remember CompuServe? ;o) I had a Sun E250 as foot warmer under the desk (o;

But I still like the X-Motif interface on the 16702A.....not sure if the E4406A can be controlled remotely as well via a X11...

[Wuerstchenhund]

Well would be a great combination to get a hand on a E4406A as well....maybe I sell one of my 2 CRTU-RUs even (o;

It's a nice device, but for the great fun you'd also need the Agilent 89600A VSA software 

Hmm..I was more back then the SunOS 4.1.3 guy....configuring SunISDN card and had a hard time to find a provider who supported HDLC protocol for dial-up internet connections...and at workplace at CompuServe Switzerland (remember CompuServe? ;o) I had a Sun E250 as foot warmer under the desk (o;

I remember the E250. It was a nice box but I never managed to get one for some reason, so I ended up with E3000s, E3500s and E4500s (the E3x00s were nice deskside machines, and not too loud). But then, I was more a HP-UX and AIX guy anyways, plus a fair bit of IRIX and ReliantUNIX 

But I still like the X-Motif interface on the 16702A.....not sure if the E4406A can be controlled remotely as well via a X11...

I never tried but I can't see why not. And in theory it should be possible to connect an external SCSI hard drive (the E4406A has an SCSI port) and a SCSI CDROM, and either install HP-UX/RT to the disk or install a full HP-UX 9.x variant for the 744 on it, and copy the VSA software across.

[davorin]

Hmm...the Agilent 89600A VSA software seems an expensive bastard...but read it could be bypassed in a way?

Hmm..Irix....always loved those small SGI espresso machines back then...but back then I switched to Mac...as they had 68k inside and I never liked Intel CPUs with the strange memory layout (o;

[G0HZU]

The E4406A is really just a signal analyser that is designed to be used under very controlled lab conditions. It isn't really meant to be used as a spectrum analyser because of the crude nature of the frequency plan in the RF downconverter section. It has no image rejection and has areas in its 7MHz to 4GHz frequency range that will have poor spurious performance.

However, when used as a signal analyser for known signals it is a very powerful tool. I use mine with the 89600 SW and I also use it with an Agilent vector sig gen that has a dual arb generator built in.

eg it's possible to capture (offair?) signals using the E4406A and the 89600 SW and then export the data to the supported sig gen to be able to replay it back in a continuous loop in the sig gen. This can be at a different frequency or power level and you can also modify the captured data before sending it to the sig gen.

In my case, my fairly old Agilent vector sig gen has the required dual arb generator but it isn't supported by the 89600 SW so I created a Windows app to convert the captured data to the format expected by my sig gen and I can then upload it to my sig gen to fully replicate the captured offair signal in the sig gen.

[davorin]

What model of vector signal generator do you use?

Sounds interesting...

[G0HZU]

I've got two Agilent vector sig gen models with different options but they are both quite old and rapidly becoming obsolete in terms of their usefulness as research tools.

The one with the dual arb generator option is the Agilent E4433B ESGD sig gen. But it isn't supported by V12 of the 89600 SW so I created an app to convert and upload the data to the sig gen in the format it is used to. i.e. it expects to see separate I and Q data files. It works up to something like a 35MHz modulation BW.

If I had the Agilent E4438C sig gen I think the 89600 SW could generate data files that can be sent to it directly.

Note that the alternative (quick and dirty) way to do the above is to do it with a modern SDR Rx and Tx combo to capture and replay stuff.
 

[davorin]

Is it also possible to capture/display longer periods up to several 100msecs?

This signal below was captured with a Signalhund SA44B, though display time is limited a little over 100msec....

[G0HZU]

Is it also possible to capture/display longer periods up to several 100msecs?

I think it depends on the signal and how you set up the system. eg for narrowband signals you can capture many seconds of data until you hit the memory limit.

You can also look at signals on the 89600 SW via the LAN interface and this runs continuously although it isn't really capturing all the data although it is still useful in this mode as long as you don't mind missing some data. For many applications this doesn't really matter and you can still demodulate many waveforms in this continuous mode.

I finally got round to swapping the RTC battery over in mine today. It didn't take that long and I needed an excuse to try out a cheapo Worx cordless screwdriver anyway

There are loads of Torx screws to remove and some care is needed when removing both the digital IF, SCSI and the CPU cards.

But it went smoothly and the only tricky bit was pulling out the CPU card as you have to pull quite hard yet you also have to maintain a lot of control in order to keep the card moving slowly (and level) to avoid snag damage. There's a warning label on the rear panel about this and also in the battery replacement  instructions.

It's a shame that battery access isn't simpler especially if it only lasts a few years per battery.