EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: implor on November 14, 2019, 05:13:25 pm
-
Maybe this has been discussed but if not I think it's an interesting topic.
I have been looking for an low end spectrum analyzer on Ebay for ~6months to learn more about RF. Mostly for antenna designs and EMI testing (and fooling around ::)).
But I have almost given up finding anything decent (a.k.a working) without going broke. The prices are Insane!
>25 year old gear that's untested is going for $1000-$5000!
Why do ppl buy them when you can get a New Siglent/Rigol w. tracking gen for ~$1000-$1500? I'm I missing something?
-
There's some nonsense, there's some reflection of the unit's original cost, and there's some reflection of the relative rarity/desirability of them.
It's going to be hard to find a working, tested SA for that cheap.... $1000 is a reasonable ballpark, but especially older models can be found to around $500 in good condition from some makers. A lot of the offerings on the used market are higher bandwidth than the Rigol/Siglent units you're talking about... and their base price is on the high end of that estimate, especially if you want a TG.
Honestly, though, the Rigol SAs were some of the earlier inexpensive SAs available on the market, and they were launched at a fairly disruptive price point, so the whole point of their price point was to be a capable alternative to the used market and a new-low entry level to the new market. Siglent, to a lesser extent Rhode & Schwarz, and some other manufacturers are trying to take a shot at this new lower priced RF gear market segment. To some extent, it has actually pushed the prices of the used market down a bit, but there isn't anything that cheap with wide input bandwidth (which is becoming more and more important with 5G and EMI checking high frequency devices), there's plenty of bias against Chinese made equipment still around, and there are some options/performance characteristics that the new options haven't yet offered (though usually not that close to the $1000 price point on the used market).
There's a used SA buyer's guide thread floating around that could give you some ideas of brands to look at/units to consider that are less expensive. Also keep in mind that there are sellers on ebay that will list at a price because it once cost a lot or because someone else did and there wasn't anyone undercutting the market when they listed.... so there is plenty of gear that is inflated 5-10x a price that someone would actually buy it at.
-
Interesting view.
For me the big question is who pays this amount of money? hobbyist? companies?
I would never buy used gear without some serious warranty to my company and >2k is to much for most hobbyist.
Will try to find the buyers guides you talks about :-+
:::OFFTOPIC RANT:::
At work we develop NFC, BT, Wifi and LoRa systems and still my manager is to cheap to by a spectrum analyzer. An "Unnecessary expense" he say... :palm:
-
I didn't pay anywhere near that for either my 8568B or my 8566A. I did, however, make friends with the guy I got them from and after having done business with him several times, I trust that he's not sending out junk.
I would not pay that much for an old spectrum analyzer unless it's coming with a warranty from a reputable refurbishment house.
-
Where's your sense of adventure? :-DD
I bought my hp 8568b as separate "non-working" units for $300 and another $100 for the interconnect cables. The RF unit needed a new lamp in the encoder and the display unit needed some tweaks on one of the boards. Shipping prices are making that harder, I think I paid a total of $60 for shipping both units. At any rate, I have a fully functioning 8568B for under $500.
That's what a hobbyist does ;D I do that partly because I enjoy fixing test equipment, partly that I get a kick out of having $44K of equipment on my desk. If I just wanted to get on with it, I probably would have gotten the Rigol. I see lots of older SAs that are no longer supported in the multi kilobuck range; no way could I justify that. I can't see a business buying those either unless they were collecting parts to maintain a test cell.
-
Where's your sense of adventure? :-DD
:)
For me shipping would be $200-250 for something that size + tariffs and import taxes. So it would end at ~$900 total. For that amount I could buy a DSA710 with warranty :-+
I have been looking base station portable units. They don't have TG but still. $300-500 + cheap shipping.
-
For boat anchors in general shipping now seems to make many of them uneconomical. SA's and sig gens are generally still commanding high outright prices despite the extra burden of shipping. For now I'll stick to my scope's FFT and my SDR dongle. I see USB based SA's are starting to appear on the usual sights at the sub $200 mark so maybe these will be viable options in the near future.
-
For me the big question is who pays this amount of money? hobbyist? companies?
I think it's mostly hobbyists. If you have some disposable income and are into the hobby, a spectrum or network analyzer facilitates a lot of extra measurement and characterization for radio frequency stuff that is tough to get without, so people save up and have it as one of their more expensive bits of equipment - at least that's my assumption.
That said, it probably doesn't feel so out of reach for a lot of hobbyists - if you're a voltnut you're easily spending the same ballpark on a 7.5 digit meter... if you want a working 8.5 digit meter it's going to be hard to find under $2-3k, and something like a 3458A is going to run you $4000 or more usually - $2500 will get you a decent parts unit.
Part of the appeal of the unknown/known bad equipment for a hobbyist is that it's a project, too. Working inside expensive gear gives you a good understanding and appreciation of it, and it feels very satisfying to finally get something fixed - in some cases it's also not terribly difficult, so since the used and working market is still pretty high value, the non-working market stays not too far back because there are people who want one who are willing to invest the time/effort to make it work. Not really worth it on an hourly pay scale or sometimes just by the cost of the replacement parts, but for a hobbyist doing the work for free, over a long time, and chasing that fancy bit of kit for the centerpiece of their bench... yeah I can see why they'd still be pricey.
I think if you take a look at the total volume of instruments sold by general type, there are probably 100 multimeters sold for every oscilloscope, and easily 100 oscilloscopes sold for every full fledged spectrum analyzer - and while they are sold every day on ebay, it wouldn't be surprising to me if it was single digit quantity per day a lot of days of the year - they are still pretty specialized and somewhat rare.
-
Rigols and HP (for example) are world apart in how they are constructed. Mine are all from HP and I've been very happy with them.
Low end HP8590x and HP8591x has been coming out on eBay in well below US$1000 range. They are semi-portable.
You seem to be outside of US - that puts you in disadvantage as most of them are from US vendors. Even with guarantee, you would end up losing a lot if it has to go back. My first SA was a dud. Worked out a deal with the seller, sold it to a local lab for parts, and bought a refurbished unit from them.
Since then, I bought dozen or so SA but they are often mixed bags. It can be tricky, too. I usually do over-night power-on test after I'm satisfied with safety. I've had a few that YTO lose lock or some other issues show up.
If you are looking at eBay, be sure to look at SOLD prices, not the listed prices. I doubt no one is buying old junk for multi-thousand dollars.
Another way to go is to learn to fix them yourself - but that will require quite bit of experience and working set of instruments.
-
Just posted this in another forum....
https://www.banggood.com/custlink/DK3GeqL6ZG (https://www.banggood.com/custlink/DK3GeqL6ZG) its a 4.4Ghz $60 USD USB SA!
R.
-
I own quite a (un)reasonable amount of test equipment, including several spectrum analyzers and I am a hobbyist.
Some spend their money gambling, some in cars, others in luxury items. I like to have technology. So much for a reason.
There is another reason: I like to learn about technical/engineering issues and found that I learn best when I have the respective equipment to explore.
Examples: when I wanted to really learn how to use an oscilloscope (not the nonsense I was taught at university), I bought myself a second hand unit. For the sole purpose of playing with it. This was cheaper than paying for any course plus I got to keep the device.
I did the same with my spectrum analyzers. I know enough about them to allow me to make the measurments I need/want and I actually developed a software that implements a spectrum analyzer on cheap chinese devices (more info: click on my signature).
Why are second hand spectrum analyzers so expensive?
Here a few random reasons:
- They are solid built and will last more than a lifetime
- They have much better performance than new low-cost models: lower noise floor, higher sweep speed, etc.
- They have higher bandwidth
- They can be repaired if some/a lot experience is available
- They have "personalities"/extra software for specific measurments
- They have higher precision
- They have higher repetibility
- It is documented how to perform a calibration/compensation
- Great manuals exist
- They can be integrated in measurements cells (GPIB and third party software)
Price-wise you have these options:
50-100 Euro: Chinese SMA/NWT/D6/LTDZ - read my blog about them
1000-1500 Euro: Low-cost Siglent or Rigol
>5000 Euro: something better
Buying a 20-30 year old premium spectrum analyzer that was sold for >100.000 Euro for a mere 1000 Euro sounds like a great deal!
Especially if you can get hold of two broken ones for 500 Euro each and fix them to one working unit.
A spectrum analyzer is a dream come true for radio amateurs: suddenly you can literally SEE the whole RF band! 1000 Euro is nothing compared to the price of a proper transciever, amplifier, antenna, etc.
Everything in life is relative.
Cheers,
Vitor
-
On used SA, other reference in this forum ...
https://www.eevblog.com/forum/testgear/used-spectrum-analyzer-buyer_s-guide/ (https://www.eevblog.com/forum/testgear/used-spectrum-analyzer-buyer_s-guide/)
-
Hello all, and especially Spectrum Analyzer afficanados!
In my career, I used HP, Tektronix Analyzers since 1968, and designed real time audio analyzers before and after the FFT revolution.
Most Important specs are bandwidth, resolution and spurious free dynamic range. Check especially the input noise and maximum peak and RMS power, a blown mixer is bad news, expensive and difficult to repair.
Suggest good wideband attenuators and a DC block as many SA have inputs that can be damaged by DC.
The 1970s-1990s Tektronix scope plug-ins for 5000 and 7000 series mainframe scopes are available at $200-800.
I can personally reccomend Tek plug-ins 7L5 low band 5 MHz, and 7L12,13 1200, 1800 MHz.
A matching tracking generator is a valuable accessory, see the Tek wiki140
Beware of repair, these are Very complex electronic and mechanical design, many with exotic YAG vco, only the experts need attempt it! Find the complete service manual on any used unit you contemplate buying, some manuals are 2-3 inches thick!
Finally a very low cost alternative is the SDR - Software Defined Radio, SDRplay in UK has an excellent USB unit, I think 2 ghz for £150, also see Leo Bodnar, in UK. These have a limited input voltage range but fine panoramique spectrum display. Nice to learn about SA.
Good luck and have fun!
Just the ramblings of an old retired EE
,
Jon in Paris
-
R,
Do you have one of these devices . . . and if so, what has been your experience with it? I purchased one (seemed like a cheap experiment). Software installed fine. Found the com port wherein it resides and the software seems to be talking with the analyzer in civil fashion. The only 'instruction's I've found so far are more like a two page brochure of features cited in two languages! The screens suggest a lot of capability but not very intuitive. I've not been able to detect/plot a hand-held transceiver's output . . .
Are you aware of any tutorials or more lucid instructions? I've heard some mumblings on the 'net suggesting that there are third party originators of software for this critter.
I have a low cost, MiniVNA-tiny analyzer that has proven capable and quite useful. I'm still hoping to expand my stable of RF trick ponies with the Geekcreit product.
Thanks!
Bob, K0DYH since 1956
-
R,
Put my question on hold . . . found Vitor's stash of publications on this device. I'll see if his generous work-product opens the right doors in my gray matter.
Thanks!
Bob . . .
-
Interesting view.
For me the big question is who pays this amount of money? hobbyist? companies?
I would never buy used gear without some serious warranty to my company and >2k is to much for most hobbyist.
I've never cared one bit about a warranty, if something sold as tested working is DOA I'll return it, but beyond that I usually find it's easier to repair a piece of equipment myself than send it back under the warranty.
Hobbyists cover a wide range of people with a wide range of financial situations, there are lots of people out there for whom $2k is a trivial amount of money they'll spend on a whim, I'm not one of those but I did spend ~$500 on one of my scopes which I knew had some issues, nevermind the lack of a warranty but you're not gonna find a new 1GHz 4 channel scope anywhere near that price range. I don't have a SA because for what I do I can't justify the cost but if I were going to buy one and the price were the same I'd take the older HP as-is vs a new Rigol with a warranty any day. I like the feel of older high end gear, the fit & finish of equipment that cost tens of thousands when new is much better than modern low cost Chinese gear.
Lots of expensive hobbies out there though, sometimes if you wanna play, you gotta pay. Take up sailing or flying airplanes and $2k for a hobby toy will sound like peanuts. I think we have it pretty good with electronics these days, there's a ton of expensive gear out there that would have been completely out of reach just 10-15 years ago that is now relatively affordable.
-
I was never a fan of these 2 piece deals; Laptop & USB attachment (dongle) imitation SA. Two different pieces of electronics and software that is suppose to work together. Then add the problem of M$ and their crappy USB drivers.
The price is tempting, but it's downhill from there (unless you are lucky). besides, it's still two pieces of equipment one has to lug around (unless the intention to be bench only), one of which is easily damaged or lost especially if it is one of those "sticks". ;)
-
I thought using an rtl sdr dongle would be a quick and easy way to solve a need I had for a project. It did provide incentive to get the minor problems with my HP8568b resolved.
If you look closely you will see that most sdr programs report signal strength in dB/Hz not in absolute signal strength. Some documentation conveniently skips mentioning this at all. Some dongles do not allow gain or AGC to be controlled so getting a calibrated result is very difficult (I won't say impossible, but I got tired of chasing it). The dongle I have using the rtl_power_fftw software reports values of around -20dB/Hz for signals that range from -65 dBm to -54 dBm. Spectra for signals around 70uV look very similar to 700 uV. OK for qualitative purposes, maybe, but if you really need an SA best to actually get one.
I can get more useful readings using the SA output on an old Wavetek SAM hooked to an old scope (which, if you want a basic SA, is not a bad way to go since you can find those meters for <$100, you just need X/Y input on the scope. At least it will show actual signal levels. Of course you have to go back to the dark ages and read the graticule on the scope)
-
SDR sounds good on paper, but that's about it when it come to using one for test equipment.
-
:::OFFTOPIC RANT:::
At work we develop NFC, BT, Wifi and LoRa systems and still my manager is to cheap to by a spectrum analyzer. An "Unnecessary expense" he say... :palm:
If your company DEVELOP RF products and do not use a spectrum analyzer, then I would never buy any product DEVELOPED by your company. If your company USES RF modules already FCC or CE certified and off the shelf pre-tuned antennas, then there is not much need for a spectrum analyzer except for doing some pre-compliance EMC tests
-
SDR sounds good on paper, but that's about it when it come to using one for test equipment.
SDRs are great for signal surveillance and demodulating stuff at a low cost. You pretty much need a calibrated instrument for serious RF work, however.
-
SDR dongles are essentially in the same category as the little toy oscilloscopes you can pick up for <$100. They can show you a rough picture of what's there and you can learn the fundamentals but they're not really useful as a serious instrument. They are a lot of fun though, I use mine mostly for listening.