Spectrum analyser choice: Anritsu MS2661G or HP E4411B ?

[peter-h]

I am looking at these two. Both similarly priced, ex rental units.

The HP has options A4H A4J B72 049 (GPIB+Centronics, 50 ohm, 10MB memory, colour LCD).

The specs are hard to compare because almost none of it is directly comparable, but the obvious bits look very similar.

The HP unit spec is here: http://cp.literature.agilent.com/litweb/pdf/5989-9556EN.pdf
Anritsu one is here: http://www.anritsu.com/en-us/downloads/brochures-datasheets-and-catalogs/datasheet/dwl1582.aspx

The "G" looks like a typo but I have a photo, and it is actually listed on some Japanese websites...

I am sure both will be fine for my requirements (up to about 500MHz), so I am really looking for comments on quality, plus some non-obvious technical aspects.

I will be doing some amplitude modulation, about 20-30kHz deviation, so need to be able to check that sort of thing.

The screen image quality doesn't look too great on the Anritsu photos I have seen, but maybe they had the brightness turned way up, and the unit does more than the HP one when it comes to controlling it from a laptop, printing off screenshots, etc.

[Rudane]

I have used three HP spectrum analyzers, and I've loved them all. I've only tested an Anritsu spectrum analyzer once. I used the Anritsu for maybe 10 or 15 minutes, but I was very discouraged at the fact they used Windows on their unit. I'm sure the unit works just fine, but I much prefer the unix OS on my spectrum analyzers. This is just my opinion, so take it at that. I noticed that this Anritsu doesn't seem to use Windows, so maybe they have moved on from that phase.

[nctnico]

If your frequency span requirement is limited then KJDS from this forum has some HP E4406A Vector Signal Analysers for sale.

[KJDS]

The Anritsu I would avoid unless it had the narrow RBW option fitted, otherwise the minimum RBW is 1kHz whcih is too big to look at a lot of modulation formats. It does have better phase noise but not significantly so. If it is a similar display to that used in the MT8801B radio test set then I've seen a few of those go fuzzy over time.

I'd be amazed if you could ever get a schematic for the Anritsu, they don't tend to release them. The schematics and CLIP aren't available on for the E4411 on the Agilent website, but there is a service guide to provide a starting point to repairs should they be necessary.

If a 10MHz span is sufficient, then I can supply an E4406A VSA at an excellent price.

If there are any particular specification areas that you need help on then ask and I'll try and decode what the manufacturers really mean.

[G0HZU]

At work we have a few of the 26GHz E4407B analysers that look very similar.
However, I think the RF performance in terms of noise and linearity is probably a fair bit better than the E4411.

But what I can say is that these E44xx analysers from HP/Agilent are very nice to pick up and carry. Very nice chunky feel and nice touchy, feely rubber surrounds. The display on the 4407B is quite nice and the user interface is pretty good and the analyser has a nice feeling of quality.

The ones we have do take a while to start up as they do all kinds of internal LO and cal checks. Also they have an annoying self cal routine that make a faint chirpy noise. A bit like a tiny bird has just hatched from an egg and is letting out tiny chirps from deep inside the analyser 

What I don't like about the E4411B in your link is the 1kHz RBW limit and the very poor phase noise. Also the IP3 and 2HI specs are a bit low.

The overall level accuracy is reasonable but my impression overall is that this analyser will look and feel great to use but the RF performance will be 'adequate' for field service work or casual stuff. Not really a lab grade analyser.

The Anritsu seems to have the edge spec wise but it looks horrible and the display is not pleasing. Also, it may be harder to support long term as spares will not be as plentiful.

At a guess, the thing that will bite you first with the HP will be the 1kHz RBW limit. Then maybe the below par linearity. But if you really don't mind about these issues then I'd take the HP over the Anritsu purely on the ease of use and especially if you want to lug it around in a car or around a building. The rubbery shock mounted feel of the HP is very reassuring and it does look the part when it is up and running

[peter-h]

Thank you all for your input.

The Anritsu unit (have just noticed in a photo of the back) comes with options 01 02 05 07 which are

reference crystal oscillator, 1x 10e-7/year ageing
narrow resolution bw 30 100 300Hz
mysterious - nothing on google on opt 05 and nothing in the data sheet
am/fm demodulator (internal speaker & earphone socket)

Option 05 comes up as a rubidium clock, in a different Anritsu context... hmmm! But it's not a typo:

Here is a piece of the photo I have

and I would be looking for a better display. Maybe they have just misadjusted it because that one looks too bad, and a lot worse than some online.

What attracts me to this unit is that it is practically totally as-new and unused and at a very very good price. The HP is twice as much and being on the rental circuit for a few years it will have been heavily used (probably).

Of the two boxes I am looking at (the other is a 2024 sig gen) the spectrum analyser will not need to be carried around, so that's OK.

What I don't like about the E4411B in your link is the 1kHz RBW limit and the very poor phase noise

What would the RBW figure be a problem with? Would it be a problem if looking for say a 100Hz AM on a 50MHz carrier?

Edit: another screenshot:

It looks better. I reckon the background ought to just be black...

Edit: Option 05 is TV trigger. Option 07 is channel power measurement.

[rfbroadband]

not sure whether TOI or dynamic range is important to you. Often R&S SA in similar classes have higher TOI than Agilent or Anritsu.