Equipment Life Expentancy

[jpb]

I'm not sure if this should be in general chat or under test equipment.

I'm in the process of setting up my lab at home and keep getting faced with the choice between new cheap equipment (e.g. Rigol, Sigilent etc) and older high quality equipment.

For example, my scope is a LeCroy WaveJet 334 which new is still sold for nearly £5k in some places and I got it as old stock (never sold) for £1.5k BUT it is nearly 7 years old (built September 2006). I think it was a bargain but perhaps if it packs up in 2 years time I'll be less happy.

I'm now looking to get a signal/function generator and have the opportunity to buy a Tabor model which is probably about 6 or 7 years old (again old stock) but at around 35% of list (1 year warranty). Agilent ebay store is selling a nice one for around 55% of list but it is 3 years old and only has 7 days warranty. Similarly there is the option for ex-demo TTi with a years warranty. Alternatively there are new Rigol/Sigilent.

People seem to have fully functioning equipment that is decades old but on the other hand capacitors seem to have a limited lifetime (5 to 10 years?) even if the equipment hasn't been used.

It would be really useful to know what people's experience has been as to how well good equipment lasts. Manufacturers seem to have 3 to 5 year warranties on new which implies that they should be trouble free for that sort of length of time at least.

Of course the other factor is that technology advances so that an instrument that was cutting edge 7 years ago is now of similar specification to more entry level models but the manufacturers don't tend to drop the new prices in line. Thus the new prices are rather inflated and 35% of new list might be nearer to 60% of what new models actually sell for (the street price).

[Fsck]

I have a Keithley 220 that's like as old as my parents and a Fluke 8842a that's almost as old as I am.
Some things last forever (relatively speaking) if well maintained.

If it's only 5-8 years old, high-end tech isn't really too old and will often beat the pants off new scopes where (used price)=(MSRP), except for possibly certain factors like sample rate, memory, etc. Scopes and function generators are updated often, but some things like bench mm's and psus aren't. The hp 3458a was introduced eons ago and is still considered current AFAIK.

A lot of people/businesses still use analog scopes, which are ancient by today's standards but they may have all the functions that they need.

[Wuerstchenhund]

I'm not sure if this should be in general chat or under test equipment.

It's test equipment related so this is the right category.

For example, my scope is a LeCroy WaveJet 334 which new is still sold for nearly £5k in some places

Yes, but that is not representative as they will hardly find any buyers at that price. The WaveJet 300 has been replaced by the 300A several years ago.

and I got it as old stock (never sold) for £1.5k BUT it is nearly 7 years old (built September 2006). I think it was a bargain but perhaps if it packs up in 2 years time I'll be less happy.

It's a great scope, but I think £1500 is already borderline for such an old model which also lacks all the advanced analysis capabilities the larger LeCroy models support.

And asking £5k for it is simply insane, that's very close to what these scopes did cost when they were new.

People seem to have fully functioning equipment that is decades old but on the other hand capacitors seem to have a limited lifetime (5 to 10 years?) even if the equipment hasn't been used.

Yes, but the thing is that for older test equipment you get schematics and parts lists, and because they lack the very high integration od modern kit they are usually much easier to fix. Caps for example are generally easy (and cheap) to replace.

It would be really useful to know what people's experience has been as to how well good equipment lasts. Manufacturers seem to have 3 to 5 year warranties on new which implies that they should be trouble free for that sort of length of time at least.

This is not easy to answer, as past experience is not necessarily representative for kit produced today. And even if a certain piece of equipment on average lasts 20 years, this does not mean that your's can't be a lemon that only lasts two days.

Of course the other factor is that technology advances so that an instrument that was cutting edge 7 years ago is now of similar specification to more entry level models

That is not necessarily true. For example, the LeCroy Waverunner scope (a lower midrange scope, not even a high end model) I have and which was made in 2000 still has capabilities that you won't find in any entry level scope and even not in most midrange scopes.

The general performance of micrcontrollers, CPUs and other stuff has increased over the years but that does not mean that manufacturer don't use the advances to just give you roughly the same performance at lower costs for them. Entry level scopes are a good example, as they are mostly made for low costs than max performance. And it does bring out some really silly ideas like a scope where you can't switch off interpolation (Agilent DSOX2000/3000).

For example, I recently saw a video review of a Rigol DS4000 Series scope, where the Zoom functionality was demonstrated, and I was surprised how much slower this stuff is than on my 13 year old LeCroy. I've seen similar in many other areas.

Personally I would always buy an older 2nd hand device over a new entry level device, but that's just me.

[jpb]

Thanks for the responses.

and I got it as old stock (never sold) for £1.5k BUT it is nearly 7 years old (built September 2006). I think it was a bargain but perhaps if it packs up in 2 years time I'll be less happy.

It's a great scope, but I think £1500 is already borderline for such an old model which also lacks all the advanced analysis capabilities the larger LeCroy models support.

And asking £5k for it is simply insane, that's very close to what these scopes did cost when they were new.

Yes, £5k is insane (that is the Mouser price for the 334A model which is identical apart from the rear USB port). But I felt £1.5k was still a good price as other 350MHz bandwidth scopes, even Rigol are all over £3k and even a set of 4 500MHz probes cost over £1k. Anyway I've got it now so may as well be pleased with it but this does illustrate the conundrum of trying to assess a fair price to pay for older equipment.

I'm always amazed at the amount people are willing to bid for stuff on e-bay when there is no warranty and the history is unknown - my rule of thumb is around 35 to 40% of the new price whilst they often go for something closer to 55%. My daughter informs me that this is why I'm always outbid!

[Wuerstchenhund]

Yes, £5k is insane (that is the Mouser price for the 334A model which is identical apart from the rear USB port).

I'm pretty sure if someone really wanted a new 334A then he'd probably get a much more attractive price from LeCroy themselves, especially considering that this Series is in the process of being withdrawn (hopefully there will be a successor, either from LeCroy or Iwatsu, as otherwise the lower part of LeCroy's offerings would be Siglent-made only).

But I felt £1.5k was still a good price as other 350MHz bandwidth scopes, even Rigol are all over £3k

It's certainly not a bad price. Personally I wouldn't have paid more than £1500 but this is mainly because these scopes are not the right scopes for me anyways, and I would only have bought it because of the price and as a second scope ;-)

Anyway I've got it now so may as well be pleased with it

I'm pretty sure you will. Iwatsu made a lot of really great and very durable stuff.

And at least you can switch off interpolation ;-)

but this does illustrate the conundrum of trying to assess a fair price to pay for older equipment.

I'm always amazed at the amount people are willing to bid for stuff on e-bay when there is no warranty and the history is unknown - my rule of thumb is around 35 to 40% of the new price whilst they often go for something closer to 55%.

Yes, but that's the ebay effect (many people bid live and then often end up bidding way too much). The key is to have patience, look for auctions that end at times few people will spend time in front of the computer, and also look equivalent devices that are not in the main focus of potential buyers (anything that has 'HP', 'Agilent' or 'Tektronix' written on it usually attracts bidders like roadkill is attracting flies).

[marmad]

That is not necessarily true. For example, the LeCroy Waverunner scope (a lower midrange scope, not even a high end model) I have and which was made in 2000 still has capabilities that you won't find in any entry level scope and even not in most midrange scopes.

I don't want to start a war since I know how strongly you feel about 2nd hand equipment vs. new  , but in the interest of fairness I thought I should point out that, while your statement is certainly true, the opposite is also true. I looked through the Waverunner's manual and although, without a doubt, it has some great analysis tools - my entry level scope has a single-shot sampling rate of 2GSa/s; acquisition memory of 28M points per channel; vertical resolution of 12 bits when >= 5us/div @ 1GSa/sa; vertical sensitivity to 500 uV; time base setting to 2ns;  and up to 65000 time-stamped segments.

So it's not that cut and dried. I have one of the tiny, analog, double-insulated, battery-powered Tektronix scopes (~33-years old now) which I still use regularly, so I certainly have nothing against older equipment (and I do love having schematics/service manuals). But that doesn't alter the fact that some new entry-level gear can offer features (such as massive amounts of memory) which would be hard-pressed to find in older gear. I think it boils down to exactly what your needs are - and if they can be met with second-hand equipment, then it's probably the way to go. But good quality/condition, reasonably-priced, second-hand gear can be hard to find in some areas.

[jpb]

Yes, £5k is insane (that is the Mouser price for the 334A model which is identical apart from the rear USB port).

I'm pretty sure if someone really wanted a new 334A then he'd probably get a much more attractive price from LeCroy themselves, especially considering that this Series is in the process of being withdrawn (hopefully there will be a successor, either from LeCroy or Iwatsu, as otherwise the lower part of LeCroy's offerings would be Siglent-made only).

I think LeCroy are offering up to 55% off some scopes in the range but I don't know which ones (I'd bought my scope by then, so didn't inquire further).

I don't think there will be an update from Iwatsu as looking at their site their replacement scope is very similar. They have upped the RAM to 1M instead of 500k and have a few more trigger options and better LAN but the overall changes are so small they would have to call it the WaveJet 300B series rather than it being a new model.

[Rascal]

As stated 2nd hand equip or not - its all down to what a persons needs are.

I still use valve test equipment from the 1940's / 50's - its quite exceptable as long as the work you are doing doesnt need the precision and accuracy that some more modern equipment gives.

I often use an ex-World War II frequency meter, a BC-221 to test receiver / radio alignment - it works well. 



















 

[ben_r_]

No idea how long these later generations of scopes will last, but for me, I have a 5 year warranty on my Agilent and in five years there will more than likely be something better out that Ill want and will upgrade to and sell off my current to help fund. It seems to work that way with everything I buy. And seeing as how quickly these days technology advances and changes I imagine more and more people will have the same behavior. So for me, Ill be happy if it makes it the 5 years as by then Ill be done with it either way Im sure.

[Wuerstchenhund]

I don't want to start a war since I know how strongly you feel about 2nd hand equipment vs. new  , but in the interest of fairness I thought I should point out that, while your statement is certainly true, the opposite is also true.

I never said it wasn't.

looked through the Waverunner's manual and although, without a doubt, it has some great analysis tools - my entry level scope has a single-shot sampling rate of 2GSa/s; acquisition memory of 28M points per channel; vertical resolution of 12 bits when >= 5us/div @ 1GSa/sa; vertical sensitivity to 500 uV; time base setting to 2ns;  and up to 65000 time-stamped segments.

Well, I have the smallest Waverunner model (LT 224 with only 200MSa/s and 100k memory, but the better models go up to 500MHz, 1GSa/s and 2M memory, and the Waverunner 2 goes up to 1GHz, 4GSa/s and 8M), but it still offers 4 channels, a vertical resolution of up to 11 bits at all time base settings, a time base from 1ks/div to 1ns/div, a whooping 400 time-stamped segments (due to the relative small memory), automatic measurements (average, max, min, std deviation), analog persistence (green/color graded/heat map) with history (up to 4000 acquisitions), advanced FFT (power averaging, power density, real, imaginary) on all acquisition points, histogram (up to 2 billion events) and 18 histogram parameters, Pass/Fail (up to 5 parameters), digital filter (user configurable), Power Measure Analysis (Power Device Analysis, Modulation Analysis, Line Power Analysis), Hard Drive Failure Analysis, advanced jitter/timing analysis, daisy-chaining of up to four math functions, zoom (much more responsive than on the DS4000, apparently), GPIB/serial/parallel/VGA ports, 512MB Flash hard disk, optional Ethernet port (via a network card that is essentially unobtainium), large 8.4" TFT, really good Windows software available from leCroy for free,...

Not bad for less than half of what a new Rigol DS2072 would have cost me ;-)

But don't worry, I'm certainly not suggesting that this is a better scope for a beginner (most of the options are worthless without the appropriate understanding of what they do), or that this is a better scope for everyone. But it's an example  that 'old' doesn't necessarily mean 'slower' or 'inferior' (and gave me a chance to show off a little after you started it with your Rigol's specs ;-))

But that doesn't alter the fact that some new entry-level gear can offer features (such as massive amounts of memory) which would be hard-pressed to find in older gear.

You're right with memory of course (memory is cheap like chips these days, but in the past has been painfully expensive, especially fast memory required for sampling memory), but that's probably the biggest advantage (aside from a smaller size of a current entry-level scope). In terms of raw performance, there isn't necessarily an advantage with modern kit (like the example of the Rigol DS4000 zoom I mentioned) Of course there's other stuff like USB support which is definitely handy for a lot of things. Older devices usually require GPIB, and even a good Chinese GPIB-USB adapter (Beiming S82357 or F82357) costs a bit of money.

So it's not that cut and dried.

I never said it was, and I certainly won't suggest that 2nd hand is always better than new entry-level.

However, the thing is that nowadays beginners often only consider new entry level scopes (look at all the 'what scope should I buy' threads, where the contenders usually are Rigol DS1000/2000, Agilent DSOX2000, or some cheap Siglent/Hantek/Uni-T scopes). My point is that in addition to a new scope, it is worth at least considering the alternative of a 2nd hand midrange/high end scope (I would not necessarily recommend old low end scopes, unless the price is really good), as in many cases it offers either more bang for the buck or a lower price for roughly the same bang.

[smashedProton]

Well designed test equipment lasts a lifetime.  (Unless crt)

Well designed test equipment with sagan may only last a week

[marmad]

Not bad for less than half of what a new Rigol DS2072 would have cost me ;-)

Ok, but now you're just bragging about what a good deal you got    A cursory scan of Waverunner LT 224 prices on eBay puts them above what the Rigol DS2000 series costs - into DS4000 territory (which, I agree, if you're thinking of buying, you should be considering the used Waverunners as well).

However, the thing is that nowadays beginners often only consider new entry level scopes (look at all the 'what scope should I buy' threads, where the contenders usually are Rigol DS1000/2000, Agilent DSOX2000, or some cheap Siglent/Hantek/Uni-T scopes). My point is that in addition to a new scope, it is worth at least considering the alternative of a 2nd hand midrange/high end scope (I would not necessarily recommend old low end scopes, unless the price is really good), as in many cases it offers either more bang for the buck or a lower price for roughly the same bang.

Absolutely, I certainly can agree with this. As I mentioned before, location can play a big role in what your options are - if I was living in the States currently, it's likely I'd go for second-hand most of the time (easier to find, dirt-cheap shipping rates, etc) - but it's much more difficult in the continental EU. Time can also be a factor - if you can wait, you're much more likely to be able to scour eBay and other auction places / second-hand markets until you find the deal you're hoping for. But if you need a piece of test gear now, your only options, unfortunately, are often new equipment.

[MasterOfNone]

If buying equipment for home/hobby use, and considering whether to go new cheap or second hand, another thing to consider is Calibration.
If you work for a large enough company there will probably be someone responsible for making sure all the equipment is kept in Cal. After the company has sent equipment off for Calibration they won’t the engineers buggering up the Calibration by using some Self-Cal feature (that won’t be as good the pro-calibration).  So the high end equipment probably won’t contain a Self-Cal function.
At home you might not want to have to send your equipment off for Calibration, and not everyone will be prepared (or equipped) to do it themselves. So cheaper equipment with a built in Self-Cal function might be the best option. A lot of the new cheap Chinese equipment have built in Self-Cal features. 
But you could ask is self-calibration good enough? Will the Self-Cal  get significantly worse over time? Will the cheap equipment with self-Cal drift more than uncalibrated high end equipment?
- Who knows? 

[alm]

Self-cal will usually offer much less coverage that a calibration by a cal lab. Calibration usually means the comparison with external standards and signals, which are by definition absent during self-cal. Oscilloscope calibration involves signal with an accurate amplitude, frequency or rise time. Self-cal may only correct some offsets that are likely to drift as the temperature changes.

Companies regularly send out most equipment for calibration, but that doesn't mean that it stops working without calibration. Good DMMs, for example, will rarely require adjustment. It's even less of an issue with scopes, since accuracy is generally in the 1-3% range anyhow. Exception may be frequency accuracy with the internal counter, but checking frequency accuracy is easy.

[nctnico]

IMHO sending equipment to a calibration lab is more a matter of checking whether it works properly than whether it is still within its specifications. Most equipment is designed to stay within its specifications by design. Some equipment needs to be self calibrated because of changes in the operating temperature or may use components which age a little bit.

[jpb]

If buying equipment for home/hobby use, and considering whether to go new cheap or second hand, another thing to consider is Calibration.
If you work for a large enough company there will probably be someone responsible for making sure all the equipment is kept in Cal. After the company has sent equipment off for Calibration they won’t the engineers buggering up the Calibration by using some Self-Cal feature (that won’t be as good the pro-calibration).  So the high end equipment probably won’t contain a Self-Cal function.
At home you might not want to have to send your equipment off for Calibration, and not everyone will be prepared (or equipped) to do it themselves. So cheaper equipment with a built in Self-Cal function might be the best option. A lot of the new cheap Chinese equipment have built in Self-Cal features. 
But you could ask is self-calibration good enough? Will the Self-Cal  get significantly worse over time? Will the cheap equipment with self-Cal drift more than uncalibrated high end equipment?
- Who knows?

This is a good point, but I think it depends on what equipment one is looking at.

I've been looking at AWG/AFGs and for these the calibration of the older more expensive equipment seems to be better covered. For example the Tabor WW5061 has detailed calibration instructions most of which don't require expensive equipment (a multi-meter, a counter and oscilloscope). The TTi TG5011 is self-calibrating and requires a good counter and multimeter I think. The Rigol DG4062 doesn't seem to have a published calibration procedure (unless I missed it - I searched the manual) nor does it seem to have a self-calibration option though it may do it by itself;  yet it is very reliant on being calibrated at least in terms of amplitudes as it is compensating for amplifier roll-off. Agilent have calibration instructions for their waveform generator but they recommend rather expensive equipment.

For scopes most have self-calibration but I think with any scope if it goes out of calibration then it probably needs fixing (e.g. if the timebase is out or the bandwidth is wrong). With my scope I've satisfied myself by looking at dc gain and measuring a couple of fixed frequency points. As a hobby user it is not worth spending £250 plus to pay to have it properly calibrated.

Given fairly low cost access to a good frequency standard via GPS, for the home lab having a good multimeter that is occasionally calibrated probably gives enough to calibrate (in terms of checking) a scope and a function generator. The multimeter itself is the most difficult to calibrate (for say a 5 1/2 digit one) as it requires the most accuracy but then it is also the least expensive to have professionally done.

[Wuerstchenhund]

If buying equipment for home/hobby use, and considering whether to go new cheap or second hand, another thing to consider is Calibration.
If you work for a large enough company there will probably be someone responsible for making sure all the equipment is kept in Cal. After the company has sent equipment off for Calibration they won’t the engineers buggering up the Calibration by using some Self-Cal feature (that won’t be as good the pro-calibration).  So the high end equipment probably won’t contain a Self-Cal function.

I seems you don't really understand what Self-Cal is for. It is *NOT* a replacement for a full instrument calibration. Self-Cal is a functionality to do a quick compensation for changed environments and small variations due to age. Self-Cal is also not "buggering up" the instrument calibration.

If you move a scope from an environment that has say 20 deg C to one that has 22 deg C, you should run Self-Cal.

Most high end gear I have seen has had Self-Cal, for obvious reasons.

[Hydrawerk]

There are some people who don't like LeCroy scopes. They seem to have a confusing control panel and overall usage. But I've used only the crappy siglentish WaveAce. LeCroy has always been a brand of serious scientific instruments, I wouldn't recommend it to a beginner, unless you get it very cheap... My DSOX2002A is very easy to operate, it's a great beginner's scope apart from the price.  On the other hand, doesn't have much analysis functions... No statistics, vectors always on, no color grading, real sensitivity only 4mV/div, interpolation always on and no simultaneous display in XY and XT mode... The XY mode in GDS-2000A is much more advanced!

[MasterOfNone]

Yep I think you guys may right, when I run the self-Cal on my Rigol scope it seems to spend a lot of time doing the Vertical and Horizontal alignments checks/adjustments and there is a lot of relay clicking, so I assumed it was making a lot of adjustments. My Siglent Waveform generator also has a self-test that makes a lot of relay clicks. So I just assume these cheap devices were doing a lot of adjustments, but I don’t really know what they are doing.
At work our official test results have to include equipment serial numbers and calibration details or the results will not be accepted by QA. So I never go hunting in the menus for the self-cal feature, just in case it invalidated the Cal-Certificate. But it’s only on items like the Agilent 6000 series scopes that performing a self-cal will invalidated your Cal-Certificate and we don’t get those in very often.  Even though I’ve switched from doing mainly Hardware Design to Software at work, I still use test equipment because I do mainly low-level work.

[Wuerstchenhund]

There are some people who don't like LeCroy scopes. They seem to have a confusing control panel and overall usage.

Yes, to EEs, usually especially for those that come from analog scopes. But actually, I think their operating concept is very good, and more logical than those of many Agilent and Tek scopes.

But I've used only the crappy siglentish WaveAce. LeCroy has always been a brand of serious scientific instruments, I wouldn't recommend it to a beginner, unless you get it very cheap...

I don't think that the price should be the deciding factor, if someone can't operate a specific scope then it doesn't help when it was cheap.

But I would also generally not recommend a LeCroy scope for a beginner, unless he's looking into doing a lot of waveform analysis, has a solid fundamental mathematical knowledge and willing to spend some time learning about and understanding the capabilities of these scopes and the differences to other scopes. As you already said, one thing to remember is that LeCroy originally comes from the science corner (high power physics), not from Electronics Engineering, and this shows in their instrument design. LeCroy's philosophy is that a scope must always show the signal unchanged as default, all functions that affect the displayed waveform must be optional only (most Agilent and Tek scopes have interpolation enabled by default, often don't tell you what interpolation is used, and in case of the Agilent DSO-X2000/3000 can't even be switched off).

The Siglent-made WaveAce are a different story, though. They are as good or bad as their original Siglent counterparts, and aside from their insanely high price would make decent beginner's scopes if they hadn't some nasty firmware issues.

My DSOX2002A is very easy to operate, it's a great beginner's scope apart from the price.  On the other hand, doesn't have much analysis functions... No statistics, vectors always on, no color grading, real sensitivity only 4mV/div, interpolation always on and no simultaneous display in XY and XT mode... The XY mode in GDS-2000A is much more advanced!

Not bad for a scope of that class. How are the other maths capabilities of this scope?

[Hydrawerk]

DSOX-2002A's manual:

Math functions can be performed on analog channels. The resulting math
waveform is displayed in light purple.
You can use a math function on a channel even if you choose not to
display the channel on- screen.
You can:
• Perform an arithmetic operation (like add, subtract, or multiply) on
analog input channels.
• Perform a transform function (like FFT) on an analog input channel.
• Perform a transform function on the result of an arithmetic operation.

It's all that I will ever need... I think that the FFT is quite well implemented. There is even a special autoset for FFT (that real engineers don't use.  ) And I like the separate knobs for Math waveforms.
OK, in DS2000 there are more advanced features...

[Wuerstchenhund]

Hydrawerk,

my question was actually aimed at the GWInstek GDS-2000A ;-) 

The capabilities of the DSO-X2000 seem to be very basic.

[Hydrawerk]

According to the manual, GDS-2000A has Addition (+), Subtraction (–), Multiplication (×),
Division (÷), derivation, integration, square root, Hanning FFT, Hamming FFT, Rectangular FFT and Blackman FFT.
Not so much feature rich when compared to Rigol DS2000, but still better than my DSOX2002A.  But I love the separate Math controls (knob) of my scope...  Hardly any scope has it.