SA's: LPT-3000 new vs 2399A demo?

[videobruce]

I'm replaceing a less expensive Spectrum Analyzer and I'm looking at these two models;
http://www.lptech.com/LPT-3000.html
http://www.aeroflex.com/ats/products...399C~120.html# (A version)

I'm primarly using it for OTA (over the air) digital TV signal, measuring and some 2-way radio and Wi-Fi spectrum monitoring.

The LP Technologies and Aeroflex/IFR are both out of Kansas with Aeroflex being the older and more established comapny.

LP Technologies LPT-3000
$6,000 (new) including a Tracking Generator, Preamp, DTV mask & NIC. USB port. Optional software $200
Aeroflex 2399A
$5,000 (demo) including Tracking generator, Hi Stability Timebase, Distance to Fault software, read only software & Quasi Peak Detectors & filters. Floppy drive (believe it or not), Preamp $415 option, but I'm not sure it is available for this version.

Both units are around the same size & weight, have a 6 1/2" Color LCD screen & coverage to 3 GHz. The 'C' version of the 2399 has a USB port, but it is $8,300 new (probably less). Both are available directly from the manufacture.

What I was looking for is input regarding the companies themselves and how important is having a SA with a preamp? (I do know the purpose, but never thought much of amps at the receiver, only to make up for line loss).

[Al]

Well, I've got a 2399B, purchased new a year or two ago. Mine has the preamp option. Since the noise figure of most SAs is grossly poor (15-25 dB or higher) without some kind of preamp for low level signal work, I need the preamp for noise figure measurement with a calibrated noise source and viewing spurs and low level stuff directly from an antenna. But if you're viewing only relatively high level signals, switching the builtin preamp off is necessary so the analyzer doesn't lie to you due to spurs and intermod products generated by preamp overload. The preamp, when switched in, lowers the SA noise figure to 10-13 dB, low enough so that a 15 dB excess noise ratio source is easily measureable and viewable.
The 2399 was designed in Asia, optioned in the UK, and is sold by Aeroflex in Kansas. Though I've not had any trouble with mine, the documentation leaves something to be desired in terms of technical details of the instrument, and unless you can wrangle the calibration software out of Aeroflex(I wasn't successful) and have a shopfull of T&M equipment(I do), they want you to send it back to Kansas for calibration.
To be fair, for a low-priced SA it does a pretty good job, and I've checked accuracy and it's within all the published specs.
The builtin speaker on the audio demod is crappy, both in terms of level and sound quality. If you're going to use the builtin demod for anything but signal intelligence identification you may want to plug a pair of headphones into the jack on the front panel.

[videobruce]

Thanks for all of that input.
I have spent quite some time scouring through the owners manuals of both SA's and these two are very close to one another in terms of specs and features. I will say, I didn't see any real lack of documentation in the Aeroflex manual. It's no worse (or better) than most of the other manuals I have read through including Agilent and Arusta.

Regarding calibration, isn't that the way it is with all of these manufactures?? I don't have any equipment to calibrate it, so software or not, this isn't a issue. Just how far off does the calibration get year to year anyway? I'm not doing FCC proof of performance tests.

You mentioned the poor noise figure for these SA's. Poor compared to what? Is it all the internal processing that creates the noise?

[Al]

The noise figure is poor on *any* SA without some sort of preamp, be it Agilent, IFR(Aeroflex), or other. Because of intrinsic losses in the first mixer, input attenuator, internal cabling, etc. most modern SAs designed for maximum utility include an internal preamp ahead of any lossy components that's switchable inline or out.
The documentation that you saw may be newer than mine, but I found that, for example, the options are not well documented: I have the option that allows +12VDC use, and the connector pinout charts do not include data on this DIN connector.
The preamp option is not well integrated into the analyzer's basic functions. For example, when you tell it to save the current instrument state, and then recall it it forgets the complete state(i.e. preamp state) on powerup and defaults to 'off'. A save state function should save all user settings and parameters to work properly.

There's almost no documentation in the technical/service manual on the digital RBW filters.

Add to this list a few minor operational software bugs with regard to saved marker frequencies.

I don't think I'm being too critical in asking that the things I've mentioned work properly and be well documented. Like you, I looked long and hard at the Agilent CSA series before purchasing this unit. I'm not sorry that I decided on the IFR(Aeroflex).

[videobruce]

We seem to be on the same page.

My 1st SA was a no name demo cheap-o Atten 5010. This is also sold by B&K and MCM Electronics (among others). It was a older design (still current), but the price was right. $750 2 1/2 years ago. The major problem was it drifted all over the place. I timed and measured the drift rate and I found it took up to TWO HOURS for it to stabilize enough to use it. By 'lab standards' it still wouldn't be stable enough. The first 5 minutes you could actuall see the drift if you preset the scope to a stable signal (anm analog TV visual carrier in my case).

I don't know if it was just this 'demo' or all of these models, anyway it went back. I now see this same Chineese company has a newer model the 6010 amd 6011 w/ tracking generator). It now has a TXCO which tells me the 5000 series does have a 'stability issue. The price of the 6011 is $1350 or $950 direct from China. It still toips out at a Gig, but for the price, it's hard to beat for non critical use. Frankly, if I had a budget that low or a interest to match, I would give it a try. AFAIC, I would go that route over a 20+ year old 50 or 75 lb boat anchor/ 300+ watt power hog for the same price.

The 2nd SA I had for almpost 2 years, which I just traded it in, was from Instek, the GSP-810. Also sold under other names including LP Technologies. This had the tracking generator. The documentation was not that great, the procedure for setting the two markers was confusing, but I got use to it. It wasn't bad. It was a digital sampling scope as opposed to a pure analog model as the Atten, which was a plus for monitoring brief signals which these sampling scopes do do well (at least in this price range). The support was not good.
They have a software program that allows monitoring of the display, saving and printing out traces, but no remote control of the scope. For $200, ythey will provide a 'key' to activate the 'full' version. Not worth it to me. I had a problem with the software on a newer XP nForce5 based PC and was told by their support the software wouldn't work at all without purchasing it. Mind you, I have already used this on a 2k PC with a older nVidea chipset. Add to that the software is called "tryit" right in the name!

The replacement was a Instek GSP-827 (also sold by LP Technologies) with coverage to 2.7GHz and a LCD display. Two main reasons for the upgrade. Well, you think the Aeroflex is bad, this is worse.
1. The markers are a disaster. Apparently, the processor 'chocks' on having all the markers active. When you change frequencies, the markers still show on the screen even if they are outside the span. Also, part of the screen goes blank when you change modes or try to change markers frequencies. I took screen shots of this and sent themn to Instek. of course, they never heard of this before.
2.On top of that, the software won't even open (on two different PC's, both AMD based), which tells me they never tested it on a AMD based machine. It returned "Run time error, out of memory" messages that they couldn't tell me what the message meant.
3. The sensistivity of this model was worse than the 810 it replaced.

After much effort and hassle on my part, it was returned directly to Instek. I received a full refund. So much for Instek. After looking at the above mentioned HP/Agilent & Tektronix monsters, I narrowered it down to these two in the 1st post.

[videobruce]

I should of added (since I can't edit or view my submitted posts yet), I'm still in the somewhat 'early' learing stage of these. SAchool of hard knocks. I thought about this 827 when I bought the 810, but couldn't justify the price. It turns out that was a good move as it gave me more learning time with a more basic SA.
Now add to my learning curve this unhappy experiance with this 827 POS that LP Technologies admits isn't that great and the reason they only have it is the fact it has a internal battery option the newer LPT-3000 doesn't. Even Instek admits the replacement for the 827, the GSP-830 is a much nicer unit. The 830 is in the same ball park as my choices above, but at $2k more. Clearly not worth it.

I might add, I did look at the Adventest U3741, but at a $10k+ list, it was way out of my price range. Much nicer I will admit. The current Aeroflex 2399C (with a faster sweep time and a USB port instead of that damn floppy) lists at $8300, but still out of my league.

[videobruce]

I went with the 2399A demo. Here is my review of the unit:

This was $1k cheaper than the newer LPT-3000 and $2-4k less than the Adventest 3741 (never really had a low price on this), which was the main factor. All three units appear to be in the same league. If this demo wasn't available and the only choice was the 'C' version (which is current), I would of gone with the LPT-3000.

This is made by Goldstar (LG if you perfer), in Korea! Confirmed by inspection of the circuit boards. I was really surprised here.

Pros:
I was pleasently surprised with the addition of the preamp. I wasn't orginally going with this option, but it was included with this demo. This lowers the noise floor by around 12 db which is a huge improvement. My portable Signal Scout has a preamp that I don't use since it makes little difference if it is on or off, so I wasn't expecting much improvement.

All three of these have a color LCD, but the viewing angle is very narrow. Viewing it above or below a 5 degree angle affects contrast severly. There is a 'white mode' that helps reduce the difference making it usable at angles up to 30 degrees above and below. The 9 markers are in different colors. (the LPT-3000 has this feature also)

Sweep time goes down to 20ms and is manually adjustable, unlike the Instek GSP-810. RBW and VBW are independably adjustable also.

LCD display, no more bulky CRT's. Another reason why I didn't want a 15+ year old 'boat anchor".

3 GHz coverage, ability to switch from 50 to 75 ohm input calculations, able to rename files withn the SA, carrying case included, AM/FM demod, high stability timebase, flip down stand, lightweight,

Cons:
The biggest drawback so far is the very limited room a floppy has to save screen shots unable to do this at all with the 810). I tried saving them using the jpeg format, but the compression was too great. Using the bitmap format, you can only save 4 images each time.
The software is a letdown. The GSP-810 from Instek (also sold by LP Tech, but with different software) had a far better software program. There isn't any way to transfer those files from the SA via the serial port (as far as I know).

No printed manual. For equipment in this price range, this should be included. The 810 at half the price had the manual. While I'm on the subject manuals, it seems to be a contunued trend for these Asian manufactures along with their American sales arms to contunue to supply poorly written user manuals. It seems the basic, simple stuff is in detail, but any advanced operation is brushed off.

The eight 'bumpers' (which seems to be common) take up alot of room, especially if you have limited shelf or rack space). I had to remove them so the unit would fit on the shelf. Unfortunately, there are 'studs' that are pressed into the cover that can't easily be removed that I had to protect by using small rubber grommets.

Not all settings can be saved. The Preamp, white mode and auto align functions had to be reset at every power up. I understand the preamp defaulting to off to protect the input stages, but the other two would of been nice to be saved in memory.

The markers are remembered according to the screen position, not by frequency. This seems to be another common occurence as the Instek GSP-827 (and the LPT-2750) have the same problem. Of course this isn't documentated from either company.
Example; you set markers in the frequency range of 600-700 MHz, save the screen setup, then move the sap to say, 500-550 MHz, the markers follow you to the new span, even though the orginal frequencies are outside the new range.
Yes, one can get use to this, but I perfer the way it worked with the GSP-810 where it saved the frequency of the marker, not the position on the gradicule.

Noisey fans. Another trend. The 810 and the 827 had the same problem. I insert a low wattage resistor in series with the fan leads to slow them down. Not so much to cause a overheating proble, but enough to reduce the noise level. It seems these people think all of these are used in a noisey enviroment. Hopefully, the power supply fan won't have to be replaced as that would require major surgery.

Overall, I'm happy with the purchase.


From my admittly short experiance with Spectrum Analyzers, for anyone considering purchasing one, I would take a serious look at the following (all under $10k) in order of price:

Atten AT6011 (w/ tracking generator), If pruchased stateside around $1350, if purchased from Hong Kong aorund $850. Yes, $850 WITH a TG. Coverage to 1 GHz.
This replaces the 5000 series which were very unstable (B&K still sells these). The drift was terrible taking TWO HOURS to stabilize enough to use it.

Instek GSP-810, With TG, it's well less than $3,000. Coverage to 1 GHz, two markers and softare available. Really decent lower cost model if you don't need greater frequency coverage.

LP Technologies LPT-3000, $6,000 w/ TG, preamp, ethernet port and software. Sold directly though the importer.

Aeroflex 2399C, then the Aventest U3741 for $2k more