Professional Scope Purchasing Criteria

[Wuerstchenhund]

There is a simple answer to that: If I look up the DC accuracy specs(*) of the 64MXi I see that it has a DC accuracy of 1.5%. At 5.5V that is already an uncertainty of +/-82mV. As a rule of thumb you can't use an oscilloscope for really accurate DC or amplitude measurements. For such measurements a DC or AC voltmeter are much better tools. IMHO Lecroy shows too many meaningless digits in the result.

LeCroy scopes have a function where from time to time the acquisition system goes through an auto-calibration cycle to compensate for temperature changes. This function (which can be disabled) increases the precision of voltage measurements, although it sometimes can be a bit annoying if the scope re-calibrates after changes to the vertical settings have been made.

Scopes from other manufacturers don't do that. They self calibrate after power up and then that's about it.

The specs listed are warranted worst-case limits which are valid even with auto-cal disabled.

So no, the digits are not meaningless.

[nctnico]

Sorry but if the specification says there is a typical error of 1% and worst case 1.5% you can't magically make the measurement more accurate. You have to assume a reading of 10V with a 1% accuracy is somewhere between 9.9 and 10.1V in reality. You could verify your oscilloscope against a reference but that verification is only true for the exact settings per verification. Different attenuation or offset and all bets are off. And that is just DC.

[TunerSandwich]

Sorry but if the specification says there is a typical error of 1% and worst case 1.5% you can't magically make the measurement more accurate. You have to assume a reading of 10V with a 1% accuracy is somewhere between 9.9 and 10.1V in reality. You could verify your oscilloscope against a reference but that verification is only true for the exact settings per verification. Different attenuation or offset and all bets are off. And that is just DC.

...you obviously aren't reading the posts here...are choosing to ignore fact....or are simply trolling. 

My guess is you just don't know what "full scale" means.  Even though I explained it to you above.....and posted pictures as proof.  The scale represented is 160mV.

You know those little clicks you here when you are ranging a MM or a scope?

By the same token, you claim I am taking things off-topic here....but in fact it YOU that are trying to take things down the road of fiction. 

Explain to us what "full scale" means....and then explain why the pictures I posted above show accuracy, in accordance with the MM, and certainly within the 2.4mV/160mV claimed by the manufacturer.....



 

[TunerSandwich]

somehow, even at 20mV.....and through all the noise of the crap Korad PSU (which is greater than 20mV)....the LeCroy acquisition, trigger, cursor, measurement and calibration system....still seems to be able to resolve a pretty outstanding degree of accuracy.....

[Wuerstchenhund]

even for agilent? start up calibrate once only?

I've not seen a similar thing on any other scope.

[nctnico]

even for agilent? start up calibrate once only?

I've not seen a similar thing on any other scope.

My Siglent does a self calibrate every now and then during use. But given their relationship with Lecroy that is not a surprise.
Siglent does specify the accuracy very detailed. Not only is the DC accuracy specified for the vertical deflection but also the accuracy of the offset and frequency response flatness in 3 frequency ranges. An oscilloscope may perform better than spefified but you have to do the verification work yourself and be sure to get it right if you want meaningful data beyond the speficied accuracy. Hold on to your helmet if you ever need to take measurements with an oscilloscope which need to hold up in court...

[Wuerstchenhund]

My Siglent does a self calibrate every now and then during use.

Is that a SDS2000?

But given their relationship with Lecroy that is not a surprise.

Probably not.

[nctnico]

My Siglent does a self calibrate every now and then during use.

Is that a SDS2000?

It is indeed.

[tautech]

My Siglent does a self calibrate every now and then during use.

Is that a SDS2000?

It is indeed.

Auto Self Cal (actually an auto Quick cal) can be disabled to a manual Self Cal mode in the Siglent SDS2000 series.

[Wuerstchenhund]

Auto Self Cal (actually an auto Quick cal) can be disabled to a manual Self Cal mode in the Siglent SDS2000 series.

Ok, so it's a bit different than on LeCroy scopes as they do a full auto calibration, although on newer scopes (i.e. WavePro 7kA manufactured after somewhere late 2008) this can be changed so that the scope does a quick cal instead, with the loss of some accuracy. The recalibration cycles sure can pull on one's nerves sometimes.

But it does show the influence LeCroy seems to have on Siglent's developments, and I'd say a quick cal is completely fine for an entry level 2GSa/s scope.

Just out of curiosity, it would be interesting to see the same test Tuner Sandwich did with his LeCroy scope done with a SDS2000 scope.

[nctnico]

I have given the 'auto cal' feature some thought and I'm leaning towards thinking it is a crap solution to hide a poor design. I think we all know component values drift a little bit over temperature. A part of the art of designing electronics is to make sure drifting component values stay within the design limits. The reason you don't see the automatic calibration on Tektronix or Agilent scopes is most probably because they don't have to because their designs are better. Worse, if the auto cal is disabled what kind of accuracy can you expect from a Lecroy scope?

Now you could argue Lecroy tries to make their oscilloscopes better than their competitors but then tell me why they specify and accuracy of 1% and not 0.1% or even better? Or 1% with 'auto cal' off and 0.1% with 'auto cal' on? 1% accuracy just puts them in the same league as Tektronix and Agilent.

[Wuerstchenhund]

I have given the 'auto cal' feature some thought and I'm leaning towards thinking it is a crap solution to hide a poor design. I think we all know component values drift a little bit over temperature. A part of the art of designing electronics is to make sure drifting component values stay within the design limits. The reason you don't see the automatic calibration on Tektronix or Agilent scopes is most probably because they don't have to because their designs are better.

That's utter nonsense, really. I'd say LeCroy has clearly shown that they know quite a bit about proper DSO design, probably more so than Agilent who tends to win customers more by 'optimistic' (sometimes borderline dishonest) marketing than by technical progress. If you look back in the history of DSOs, it was mostly LeCroy who pushed the boundaries (i.e. deep memory scopes, advanced processing, bandwidth pushes for real-time scopes). And while LeCroy came out with a true 12bit scope, Agilent had to resort to some oversampled 8bit scope that needed to be bolstered up by marketing to become competitive. Oh, and Agilent's UI is still as shit as it was in 1998 when they came out with their first Infiniium scope running Windows95.

And Tektronix, are you kidding me? I understand some people have some fond memories from the old analog scope days, but frankly in terms of scopes their DSOs are crap. Their low-end designs are buggy, and their high-end designs are ancient (33GHz top). It's shit all around, they can't even compete with Agilent. The only sales they make are probably to people stuck in yesteryear that think the Tek brand still stands for a good product.

Worse, if the auto cal is disabled what kind of accuracy can you expect from a Lecroy scope?

Same or better than the specs say. As I said, the specs are worst-case assumptions, and remain valid even if auto-cal is disabled.

Now you could argue Lecroy tries to make their oscilloscopes better than their competitors but then tell me why they specify and accuracy of 1% and not 0.1% or even better? Or 1% with 'auto cal' off and 0.1% with 'auto cal' on? 1% accuracy just puts them in the same league as Tektronix and Agilent.

Simply because the effectiveness of a temperature-based recalibration very much depends on the environment. The improvement this brings probably also depends on other factors (i.e. vertical setting) as well. I'm sure if they were Agilent they would have made up a best-case scenario and listed the specs for that, but they're not Agilent. So why make up some artificial scenario when the worst-case specs are already as good as for other scopes?

BTW: One thing that LeCroy doesn't participate is Tek's and Agilent's 'ours-vs-theirs' comparisons where they stick some moron in front of a camera who then goes to incredible lengths to make the competing product (which is generally a lower model than their own scope model) look as bad as possible, even if this involves lying. You won't see these types of videos from LeCroy, and if any competing products are listed in some of their documents you will find that all test details are clearly specified and repeatable. And they don't try to hide if a competitor's product comes out better from the tests than their own. Such corporate behavior is a really rare thing to find these days.

Somehow I have the feeling you haven't really given much thought to it at all.

[nctnico]

I was hoping you could point me to some specifications which say what the difference is between having an 'auto cal feature' and without. As I wrote before an oscilloscope is just as good as it's specifications say it is. What I've seen in the Lecroy datasheet I pointed to earlier is a very incomplete way to specify an oscilloscope.

[tautech]

I believe you all are not making a distinction between Self Cal & Auto Self Cal

Self Cal is a Manually used feature to keep the full range of specifications within the published range.

Auto Cal is self calibratimg "on the fly" the range of settings in use at the time.

These are very different processes, the FULL Self Cal takes some time while Auto Cal is barely noticeble and incorporated for operator convenience and of course continued ongoing accuracy.
Hence we hear relays clicking etc while the scope is in use but not being adjusted if Auto Cal is enabled.

[tautech]

Just out of curiosity, it would be interesting to see the same test Tuner Sandwich did with his LeCroy scope done with a SDS2000 scope.

Siglent are about to release the Power Analysis option for their SDS2000 series.
They supply an additional USB powered PCB for Skew null adjustment between current and voltage probes.

PA is a new field for me but I hope to post some simple analysis in the SDS2000 thread.
Back to study in preparation......

[TunerSandwich]

I have given the 'auto cal' feature some thought and I'm leaning towards thinking it is a crap solution to hide a poor design. I think we all know component values drift a little bit over temperature. A part of the art of designing electronics is to make sure drifting component values stay within the design limits. The reason you don't see the automatic calibration on Tektronix or Agilent scopes is most probably because they don't have to because their designs are better. Worse, if the auto cal is disabled what kind of accuracy can you expect from a Lecroy scope?

Now you could argue Lecroy tries to make their oscilloscopes better than their competitors but then tell me why they specify and accuracy of 1% and not 0.1% or even better? Or 1% with 'auto cal' off and 0.1% with 'auto cal' on? 1% accuracy just puts them in the same league as Tektronix and Agilent.

you are, once again, bypassing what the term "full scale" means.....LeCroy is stating accuracy over a given range.....which is then temperature compensated.  ALL devices change thermally.....just because another manufacturer chooses not to take that into account, doesn't make their design "better".....

it looks like I have to provide some more tests......although you won't ever be convinced, because you know it all already 

[Wuerstchenhund]

I was hoping you could point me to some specifications which say what the difference is between having an 'auto cal feature' and without. As I wrote before an oscilloscope is just as good as it's specifications say it is. What I've seen in the Lecroy datasheet I pointed to earlier is a very incomplete way to specify an oscilloscope.

Is it? or is it just that you don't know what the specs are for?

The automatic recalibration feature offers an improvement in accuracy, nothing more, nothing less. The specs listed in that datasheet are worst-case specifications, that means that the user can expect the scope to be no worse than what's written in that pamphlet. It doesn't mean the specs can't be better, though.

Now for some dubious reasons you seem to suggest that a scope from manufacturer X which states 1% accuracy but offers no automatic recalibration is somehow better than a scope which also offers 1% accuracy but also offers a feature (which can be disabled) which improves accuracy further.

Either you don't understand the purpose of specs or you don't get what auto recal is for.

BTW, in regard to your earlier statement that LeCroy's design is somehow inferior to say Agilents (don't mention Tek, just, don't), I challenge you to highlight some of the (in your opinion) design flaws on the old WaveRunner LT. You'll find the Service Manual including all the schematics and diagrams here:
http://168.144.151.127/manuals/82.69.35.198/LeCroy_LTXXX_WaveRunner1_Digital_Storage_Oscilloscope_Service_Manual.pdf
I've choosen the WaveRunner because it's one of LeCroy's simpler designs, and the full schematics and diagrams are freely available. I wouldn't expect you to do that for something like my WavePro 7300A (which, laughably as it is, falls under ITAR anyways).

So here's your chance to show where exactly LeCroy has failed with their obviously very crappy scope design. Blind me with your expertise 

[Wuerstchenhund]

I believe you all are not making a distinction between Self Cal & Auto Self Cal

None of these terms is standardized, and manufacturers use them as they please (not always in a logical manner). What you say is mostly accurate for older/simpler scopes which don't have a hardware GND setting for the input (i.e. on my old SDS1102CML the GND setting was pure software) and thus all signals had to be removed from the inputs manually.

In terms of LeCroy's Auto Re-Calibration, this is a full self calibration which depending on the scope will take somewhere between say 3s (WR2 LT) to over 15s (i.e. WP7kA). This is done by the scope when the environmental temperature has changed by a certain amount after the initial power-up auto self calibration. As these scopes have a software-controlled input switch (1M/50ohms/GND/Cal) there's no need to remove the signals from the input.

The thing is that modern scopes do a full self cal after power-up, and as long as the room temperature doesn't change (i.e. aircon'd lab environment) then the scope can be expected to maintain the specs. If the temperature changes, however, it's well possible that the full specifications may no longer maintained. The Auto ReCal feature tries to avoid that. The price the user pays for it is that, depending on how much the temperature changes, the system from time to time stops aquisition to go into calibration, which is a bit of a pain if that happens at the time when you need to find that glitch. That's why the function can be disabled or set to quick calibrations, depending on the scope model.

[tautech]

ha i think my usage environment temperature fluctuates badly

so when it does a self calibrate, does it show a prompt? like "please save your work, autocalibration in xx seconds" or there is a manual option?

In the interest to see this thread remain OT, reply posted:
https://www.eevblog.com/forum/testgear/siglent's-new-product-msosds2000-series/msg567800/#msg567800

[Wuerstchenhund]

so when it does a self calibrate, does it show a prompt? like "please save your work, autocalibration in xx seconds" or there is a manual option?

No. If Auto-ReCal is enabled then it will just stop aquisition and display 'Calibration' (on Windows scopes also with a progress bar.

If Auto-ReCal is disabled then a manual calibration can be performed when necessary.

[nightmechanic]

I hope I am not hijacking this thread too far,
I was reading the thread and learning a lot, and i Have a question regarding the TunerSandwich's 64MXi accuracy:

you are, once again, bypassing what the term "full scale" means.....LeCroy is stating accuracy over a given range.....which is then temperature compensated.  ALL devices change thermally.....just because another manufacturer chooses not to take that into account, doesn't make their design "better".....

It seems your scope is in fact very accurate (great!) but regarding the worst case accuracy calculations, why do you ignore the DC offset accuracy specifications ?
For they scope, as far as i uderstand, it is: ±(1.5% of offset value + 0.5% of full scale value + 1 mV) (this is from the 64Xi manual, i could not find an english one for 64MXi but from the japanese one I found it seems the same)
However, in your case if I understand correctly the offset is internal to the AP020 probe you are using (?), so the accuracy spec is related to the probe and the only number i could find is the 2% attenuation (10:1 transfer) accuracy, which should at least be considered (?) assuming the internal offset generator is prefect, or included in this 2%.

Thanks!
Ran

[TunerSandwich]

I hope I am not hijacking this thread too far,
I was reading the thread and learning a lot, and i Have a question regarding the TunerSandwich's 64MXi accuracy:

you are, once again, bypassing what the term "full scale" means.....LeCroy is stating accuracy over a given range.....which is then temperature compensated.  ALL devices change thermally.....just because another manufacturer chooses not to take that into account, doesn't make their design "better".....

It seems your scope is in fact very accurate (great!) but regarding the worst case accuracy calculations, why do you ignore the DC offset accuracy specifications ?
For they scope, as far as i uderstand, it is: ±(1.5% of offset value + 0.5% of full scale value + 1 mV) (this is from the 64Xi manual, i could not find an english one for 64MXi but from the japanese one I found it seems the same)
However, in your case if I understand correctly the offset is internal to the AP020 probe you are using (?), so the accuracy spec is related to the probe and the only number i could find is the 2% attenuation (10:1 transfer) accuracy, which should at least be considered (?) assuming the internal offset generator is prefect, or included in this 2%.

Thanks!
Ran

I explained it a few times earlier on....but it must have gotten lost in the muck. 

The full scale accuracy is based on the scale represented.  This is how ALL oscilloscopes and DMM work. 

Example.  The scale is 20mV/div....8 div wide.  That is a "full scale" measurement of 160mV. 

1.5% at >10mV/div is 2.4 mV

Just like with a DMM....where the accuracy is stated as "full scale".  If the "scale" is 0-10 V the accuracy might be 0.001%.  If the scale is 0-100 the stated accuracy usually goes down an order of magnitude....for example, possibly from 0.001% to 0.01%.....

However let's say that the circuitry was in the DMM to have a "rescaling" from 90-100V.....then we could re-gain an accuracy of 0.001%, as the "scale" would be the same width....

the claims aren't for DC "offset".....as DC offset can be easily nullified etc......

so, for example...if I AC couple the input and take a reading....on my specific scope/bench the sDEV in the measurement is the same as it would be when DC coupled....

Also LeCroys claims are "warranted"....which is a worst case scenario.  Obviously to characterize the accuracy a probe must have been used for the measurement.  So if an active probe (PA) wasn't used to make those claims..... then we can only get "better".

This is why manufacturer specs are generally only guidelines....what matters is what YOU can measure in YOUR lab environment.   

P.S.
the probe I am using (ap020) has a dynamic range of 5V.....so the maximum scale I can represent with it is 1V/div....8V total....however only 5V of that scale is usable, within the stated accuracy specifications.....

[Wuerstchenhund]

this is from the 64Xi manual, i could not find an english one for 64MXi

It's on http://teledynelecroy.com/ under Support -> Tech Library.

Look for the WaveRunner Xi-A Manual (the manual for the Xi and Xi-A Series is the same).

[TunerSandwich]

As you can see the accuracy of the scope drifts as the vertical scale becomes wider.....

That is what the term "full scale" means.....the accuracy has very little to do with total DC offset....

The test is a stabilized voltage reference @ 5v (Maxim MAX6350 buried zener) .  As you can see it was calibrated using the dm3068.... (0.5ppm/°C, ±0.02% initial accuracy)

The DMM's are NOT as accurate as the scope, despite their LONGER integration times.  Because they are reading DC, and not all of the signal energy is DC.  The AC component can't really be read by the DMM's...even in AC mode, because the max bandwidth of the AC range on the DM3068 is 200Hz.....and it's unknown on the Fluke 289, because they don't state it (and I can't be bothered to actually measure and extrapolate.....that's why i have a scope).

The important part of the measurements, on the scope, to pay attention is the MEAN.  The mean is as close to the DMM's reading as the scope can get....unless I turn on input filtering and boxcar averaging.....and that would defeat the purpose of the scope....I don't want to see pretty waveforms and filter energetic peaks.....especially if I am using this as a true reference signal, or if this was a critical i/o chain.  I want to see the overshooting energy.....that is the point.  I want to know if the peak above mean, or the trough below mean is outside of the tolerance for my application.

There is a purposeful fault in the output of this voltage reference.  I built this circuit specifically to illustrate this problem.  The DMM shows it as a nearly perfect V-ref for something like a 16 bit A/D.  The scope shows it's clearly problematic and needs some better output filtering. 

The fault is in the capacitance.  I keep showing these capacitance faults, because they illustrate my point about the need for bandwidth in an oscilloscope. 

The voltage reference is being fed by a linear DC-DC converter (ON semi LM350).  I isolated the circuits, so you guys could see what is going on.  The DC-DC is on the round disc PCB. (Line Regulation 0.005%/V.....0.3% nominal stability/1000hrs.)

The voltage ref has a 4.7k ballast/load....which is a 1% 2watt resistor.  I chose this because of it's high stability.  It is the v-sense resistor for the DMM's.  The scope probe (ap020) is attached to the lowest value output cap (the proper way to measure into the scope). 

[Wuerstchenhund]

Also LeCroys claims are "warranted"....which is a worst case scenario.  Obviously to characterize the accuracy a probe must have been used for the measurement.  So if an active probe (PA) wasn't used to make those claims..... then we can only get "better".

Those specs are pure scope specs and do not consider probes. These are "at port", i.e. a fully parametrized signal is laid to the input (usually via a calibrated direct RF connection, i.e. no probes).

If you connect a probe you have to take it's specs into account as well, i.e. if you connect a probe with (just to take some figure) say 2% of warranted accuracy then your warranted accuracy won't be better than that. If however your probe has a warranted accuracy of say 0.5% then the warranted accuracy for that setup will be the 1.5% of the scope.

With newer active probes (i.e. AP-033/034, HFP Series) which have their own internal calibration system (which calibrates the probe's amplifier section) the scope basically performs an (almost) end-to-end calibration when it does it ReCal. I'm not sure about your AP-020 probe, though, which is pretty old (it's from the old 9300 Series days, but the fact alone that a pretty modern LeCroy scope still supports their nearly 20 year old probe is just great!), as I don't know if it does have self-cal functionality.

Just wondering, for your scope screenshots, why do you use a camera and not just a screen dump? You can reconfigure the printer button in the upper right corner to create a screenshot instead and save it to whatever storage you want to. Quicker and better to read. And you can still use a printer i.e. in LabNotebook.