Author Topic: recent experience shopping for hobby scope in today's market. tek, rigol, ...  (Read 6150 times)

0 Members and 1 Guest are viewing this topic.

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
I don't want to derail this thread further. I don't really understand, which feature is available on the TBS1000C but not SDS1000X-E ?

The TBS1000C has the equivalent of a dual delayed sweep which allows capture and magnification of the waveform *after* the trigger even when it is outside of its record length, which makes up somewhat for its lack of record length.  The other DSOs make up for this with long record lengths, but can only show the waveform within the limits of their record length.  This means that the short record length of the TBS1000C is not as much of a disadvantage as it otherwise would be.

So for instance the SDS1000X-E has a record length of 8 Msamples, which means at 1 GS/s it can display the wave at up to 3 milliseconds after the trigger.  The TBS1000C can do the same thing despite having only 20 Ksamples, and can go much beyond 3 milliseconds.

Both DSOs describe something called "trigger delay" in their documentation, but they are referring to very different things.  On the Tektronix (page 72), the trigger delay delays the acquisition like a delayed trigger or delayed sweep would.  On the Siglent (page 37), the trigger delay has nothing to do with a delay, and is simply the difference in time between the trigger point and what is shown on the screen.  On the Tektronix, this is called magnification.

This is another example of a Chinese oscilloscope company deliberately lying about a feature their DSO lacks.
« Last Edit: February 27, 2023, 07:22:33 pm by David Hess »
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
I don't want to derail this thread further. I don't really understand, which feature is available on the TBS1000C but not SDS1000X-E ?

The TBS1000C has the equivalent of a dual delayed sweep which allows capture and magnification of the waveform *after* the trigger even when it is outside of its record length, which makes up somewhat for its lack of record length.  The other DSOs make up for this with long record lengths, but can only do the same within the limits of their record length.

Basically this means that the short record length of the TBS1000C is not as much of a disadvantage as it otherwise would be.
Yet for the vast majority of work 14 Mpts SDS1000X-E has would swamp the capabilities of a similar Tek.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
The TBS1000C has the equivalent of a dual delayed sweep which allows capture and magnification of the waveform *after* the trigger even when it is outside of its record length, which makes up somewhat for its lack of record length.  The other DSOs make up for this with long record lengths, but can only do the same within the limits of their record length.

I don't have the TBS1000C, but I have a similar-era Tek with a similar (one generation earlier?) ASIC and I'm not clear on the 'dual delayed sweep' statement.  Mine will do a fixed delay after trigger, but it doesn't display both the A and B sweep like earlier scopes, nor do I see how it could since it doesn't have the memory available to do so.  Is the TBS1000C like this or have they done something else?

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
I don't have the TBS1000C, but I have a similar-era Tek with a similar (one generation earlier?) ASIC and I'm not clear on the 'dual delayed sweep' statement.  Mine will do a fixed delay after trigger, but it doesn't display both the A and B sweep like earlier scopes, nor do I see how it could since it doesn't have the memory available to do so.  Is the TBS1000C like this or have they done something else?

Not all dual delayed sweep oscilloscopes display both the A and B sweep simultaneously.  If they do, then it is referred to as "alternate dual delayed sweep", which was a later innovation that DSOs tended not to copy.

The TBS1000C operates exactly like a dual delayed sweep oscilloscope in runs-after mode.  So the first sweep starts after the trigger, there is an adjustable delay, and then the second sweep starts and acquires 20k sample points.  Both sweeps are completely independent like a dual sweep oscilloscope, but only one of them is displayed at a time.  On the TBS1000C, it means that a 1 GS/s aquisition can be make way after the trigger point beyond the point where these other DSOs could do it.
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Yet for the vast majority of work 14 Mpts SDS1000X-E has would swamp the capabilities of a similar Tek.

Unless you wanted to capture and display a 1 GS/s acquisition beyond 6 milliseconds, and the TBS1000C is still has modern features despite its short record length.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
I don't have the TBS1000C, but I have a similar-era Tek with a similar (one generation earlier?) ASIC and I'm not clear on the 'dual delayed sweep' statement.  Mine will do a fixed delay after trigger, but it doesn't display both the A and B sweep like earlier scopes, nor do I see how it could since it doesn't have the memory available to do so.  Is the TBS1000C like this or have they done something else?

Not all dual delayed sweep oscilloscopes display both the A and B sweep simultaneously.  If they do, then it is referred to as "alternate dual delayed sweep", which was a later innovation that DSOs tended not to copy.

The TBS1000C operates exactly like a dual delayed sweep oscilloscope in runs-after mode.  So the first sweep starts after the trigger, there is an adjustable delay, and then the second sweep starts and acquires 20k sample points.  Both sweeps are completely independent like a dual sweep oscilloscope, but only one of them is displayed at a time.  On the TBS1000C, it means that a 1 GS/s aquisition can be make way after the trigger point beyond the point where these other DSOs could do it.
This is something you can do with any DSO. Just move the screen with the horizontal position control (which equals trigger pre/post delay). This is exactly how setting the trigger delay is described in the TBS1000C user manual BTW. On some DSOs (like the modern ones from R&S) you can specify the horizontal position by entering a number. However, there is a difference between various brands. On Lecroy and Siglent you are limited to a couple of thousand horizontal divisions, on other DSOs the maximum value is typically little over 1 second. The latter makes it possible to visualise jitter on 1 PPS signals from which the edges are far apart.
« Last Edit: February 27, 2023, 09:32:07 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 7462
  • Country: hr
Yet for the vast majority of work 14 Mpts SDS1000X-E has would swamp the capabilities of a similar Tek.

Unless you wanted to capture and display a 1 GS/s acquisition beyond 6 milliseconds, and the TBS1000C is still has modern features despite its short record length.

With due all respect, no it does not.
It is utterly outdated.  Even Tek themselves didn't dare to even compare it to Rigol DS1000Z, but to an very old and very outdated DS1000E.
All comparisons are against scope designs older than 10 years.

TBS1000C has 4 operations waveform math, no statistics on measurements, no segmented capture, no decodes, no active probe interface, only 2 ch....  And yes that very short memory is significant flaw.

It is literally worth 200 USD in comparison to other scopes you can have. You can buy SDS1104X-E or MSO5000 that make it look like a cheap chinese Finrsi.
« Last Edit: February 27, 2023, 09:26:35 pm by 2N3055 »
"Just hard work is not enough - it must be applied sensibly."
Dr. Richard W. Hamming
 
The following users thanked this post: Performa01

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
On Lecroy and Siglent you are limited to a couple of thousand horizontal divisions, on other DSOs the maximum value is typically little over 1 second. The latter makes it possible to visualise jitter on 1 PPS signals from which the edges are far apart.

So of course I had to try this on the SDS2104X+.  On a 1pps signal I can delay the trigger by 1 second on a timebase up to 100µs/div, so 10k divisions or 1000 screens worth, all at a full 2GSa/s (which means that it is not, in fact, limited by record length, which surprised me a bit).  However, I can stack that with the zoom feature and dial in down to any available timebase, even 1ns/div if I like.  That would be a hundred million screens worth.  The screenshot is at 5ns/div, so 200 million divisions, with the actual delay finely adjusted as shown in about the middle of the screen.  Obviously the accuracy is limited by the clock, an external clock is a big help here on those scopes up the food chain that have it.

I don't know what the upper limit is, but I was able to set the scope up for a delay of a full hour (displayed as 3.6ksec) with a reduced sampling rate and a maxium of 10 seconds to get the full 2GSa/s, zoomable to 1ns/div.  So it appears that 10 billion divisions of delay is the limit.

« Last Edit: February 27, 2023, 10:35:07 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 
The following users thanked this post: Performa01, thm_w, tautech, 2N3055

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Yup, that's what deep memory is all about.  :-+
Simple ain't it.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Problem is that 100us/div or even 5ns/div is not very interesting for checking a 1PPS signal (which should be rather obvious...). Things get interesting for me at 200ps/div (or less). The Lecroy Wavepro 7k which I have is pretty difficult to use for these kind of measurements though (max. delay 2us; the Siglent scopes suffer from the same limitation) while the Agilent Infiniium 54845A I used to have was much better at it because it allowed over 1 second of delay. So even though the 54845A only has 64k points per channel and far less capabilities compared to 48Mpts and analysis feature of the Wavepro 7k, the first is better suited for certain measurements. I kind of regret selling it. Bottom line is: you can't go around saying an instrument with more options is alway better. The one feature you need might be missing. Quality over quantity.
« Last Edit: February 27, 2023, 10:37:22 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Problem is that 100us/div is not very interesting for checking a 1PPS signal (which should be rather obvious...). Things get interesting for me at 200ps/div (or less).

100µS is just the first step of two.  As I stated and showed, you can expand that to 1ns/div, which is the fastest available for this model.  If you want to look at 1pps at 200ps/div with any sort of BW and time accuracy that makes that worthwhile, you need a much faster scope with a much better clock.  But there is no software or other limitation with Siglent keeping you from using whatever capabilities the scope has to full advantage and you aren't limited by record length.

SDS1104X-E is the same story.  You can have up to a 10 second trigger delay with full sampling speed (that's 1000X the record length for delay) and fastest timebase available.  It isn't as convenient and you do have to twiddle knobs and the time display only goes to 6 digits and so on, but there it is.




« Last Edit: February 27, 2023, 10:45:59 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Yup, that's what deep memory is all about.  :-+
Simple ain't it.

Well, to be accurate, it isn't based on deep memory although it helps in the acquisition end.  The available delay is 1000X as long as the record length.  Now I'm wondering about the interval trigger. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Yup, that's what deep memory is all about.  :-+
Simple ain't it.

Well, to be accurate, it isn't based on deep memory although it helps in the acquisition end.  The available delay is 1000X as long as the record length.  Now I'm wondering about the interval trigger.
;D
You'll be a while exploring all the features even the little X-E has.
Here's one I did years back using the Search and Trigger features:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg1370717/#msg1370717

An old saying comes to mind....there's more than one way to skin a cat < whoever would want to, maybe only Mr Jones.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
The Siglent manual is very specific.  Is it wrong?  On page 37:

When changing the horizontal delay, the delay time displayed in the information
bar at the top of the screen changes in real time, indicating the distance between
the time reference point and the trigger point. The amount of delay range available
depends on the time/div selected and memory depth.

 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
The Siglent manual is very specific.  Is it wrong?  On page 37:

When changing the horizontal delay, the delay time displayed in the information
bar at the top of the screen changes in real time, indicating the distance between
the time reference point and the trigger point. The amount of delay range available
depends on the time/div selected and memory depth.


That statement appears to be technically correct in that the available settings 'depend' on the other parameters, but the inference that the scope can only work by continuously recording samples from the trigger point to the end of the selected delay clearly appears to be wrong unless there is a lot of extra memory in there.  The scopes apparently do have more memory than the bare minumum suggested by their specs, but I doubt it is enough to cover the length of time and sampling rates shown in the example I posted.  I was unaware that my Siglent scopes had that ability until today--I've never had reason to dial in that much delay and like you, I would have assumed I could only push the trigger point off the left of the screen by one record length.  It never occurred to me that I could push it out by a full hour!

Edit:  On the SDS2104X+, the maximum trigger delay is 10Ms, or about 115 days.  The acquisition would be 1ks/div, or about 3 hours per screen at 20kSa/s.  It does say "Slow Acquisition" on the screen.... :)

« Last Edit: February 28, 2023, 01:00:36 am by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
The (Siglent) scopes apparently do have more memory than the bare minimum suggested by their specs, but I doubt it is enough to cover the length of time and sampling rates shown in the example I posted. 
We first saw this from History captures however there's no hard and fast rule as to how much larger it is that the datasheet memory depth spec.
SDS1004X-E and SDS5000X offer nice surprises in this regard:
https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg2786800/#msg2786800
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
The Siglent manual is very specific.  Is it wrong?  On page 37:

When changing the horizontal delay, the delay time displayed in the information
bar at the top of the screen changes in real time, indicating the distance between
the time reference point and the trigger point. The amount of delay range available
depends on the time/div selected and memory depth.


That statement appears to be technically correct in that the available settings 'depend' on the other parameters, but the inference that the scope can only work by continuously recording samples from the trigger point to the end of the selected delay clearly appears to be wrong unless there is a lot of extra memory in there.  The scopes apparently do have more memory than the bare minumum suggested by their specs, but I doubt it is enough to cover the length of time and sampling rates shown in the example I posted.  I was unaware that my Siglent scopes had that ability until today--I've never had reason to dial in that much delay and like you, I would have assumed I could only push the trigger point off the left of the screen by one record length.  It never occurred to me that I could push it out by a full hour!

Edit:  On the SDS2104X+, the maximum trigger delay is 10Ms, or about 115 days.  The acquisition would be 1ks/div, or about 3 hours per screen at 20kSa/s.  It does say "Slow Acquisition" on the screen.... :)
But it is still weird that there is a limit of 10000 divisions. This seems to be copied straight from Lecroy as well (like no zooming out) for no good reason.
« Last Edit: February 28, 2023, 01:38:12 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
But it is still weird that there is a limit of 10000 divisions. This seems to be copied straight from Lecroy as well (like no zooming out) for no good reason.

Yes, there seems to be a sort of 'that's the way it happened' rather than 'we designed in a useful feature' situation.  I can't imagine they would take a design that had a maximum trigger delay of 11 days and say "gee, let's make it 10X as long just in case..."  And it's not memory, you can manually reduce the memory depth and not change the timing.  However, the 10k division limit is meaningless when you factor in zoom.  Even on that 115-day acquisition, I can zoom in up to 50µs/div, which seems to be deliberately and sensibly set at one sample per division--no point in zooming in more than that.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline argile_tileTopic starter

  • Contributor
  • Posts: 17
  • Country: us
from OP

Laught all you want I returned it ;)  Although I did like my choice.  Something came up.

Turns out the "care free warranty" bought with the scope didn't arrive with it.  4-8 wk wait due to some distributor hang up.

Without the warranty it was too risky for me.  Is it $750/yr or $75/yr?  Does their regular warranty dept cheat like some other brands?  idk.  idk.

(un-repairability and 0 warranty make scope shopping difficult:  i checked Keysight EDU:  their ps have a part number but NOBODY sells them.  ps built around a custom ic 5layer, if it dies then you need a good one to even hope to build one, and so on you all know the story.  compare to LED tv with PS in separate box replaceable at BestBuy - you know if your LED tv needs a PS then your not out a gambled $500, is the answer there)
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
@argile_tile: you are way to worked up about warranty and repairability. In modern test equipment failures are extremely rare. There are not much parts in it and nothing runs really hot. And if something fails, it is usually power related which is easy to fix. Even without a schematic as power supply stuff is as generic as it gets.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Without the warranty it was too risky for me.

The 5-year warranty from Tek wasn't enough?  The odds of it failing in the first decade are really low and the odds that Tek would renege on their warranty is even lower.  Unless they were acquired by Keysight, which seems unlikely at this point.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 6105
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
As others have said, repairability is not really a differentiation factor among modern equipment, thus one needs other criteria to make a decision. The sudden realization of the immense gap in important features highlighted in this thread probably caused the OP to have a change of heart.

At any rate, get one oscilloscope that meets your current needs and, if in the future you feel the need for something else, pivot and make a different decision then. Aspects such as resale value and sheer durability are too volatile over the lifespan of such products to add any deterministic weight to the decision. 
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline zrq

  • Frequent Contributor
  • **
  • Posts: 341
  • Country: 00
@argile_tile: you are way to worked up about warranty and repairability. In modern test equipment failures are extremely rare. There are not much parts in it and nothing runs really hot. And if something fails, it is usually power related which is easy to fix. Even without a schematic as power supply stuff is as generic as it gets.

I would politely suggest otherwise. More or less modern test equipments do fail in irreparable ways. I heard some KS3446xA can have dying SOCs and no one in the world except KS can repair them. I also noticed there are a few users on this forum have dead Tektronix DMM4050/Fluke 8846A and even experts like TiN cannot repair some of them despite significant amount of posted work.

Of course, Chinese brands are very likely not going to be better in terms of this (except that for me, I can talk to a rep in my native language rather than English  ;)).
« Last Edit: March 02, 2023, 09:47:15 pm by zrq »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 17518
  • Country: 00
I would politely suggest otherwise. More or less modern test equipments do fail in irreparable ways.

Of course, but higher integration (single PCB, three or four big chips) is much more reliable than a box full of interconnected PCBs with backplanes and dozens of smaller chips, many of them custom made for a particular device).

 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
@argile_tile: you are way to worked up about warranty and repairability. In modern test equipment failures are extremely rare. There are not much parts in it and nothing runs really hot. And if something fails, it is usually power related which is easy to fix. Even without a schematic as power supply stuff is as generic as it gets.

I would politely suggest otherwise. More or less modern test equipments do fail in irreparable ways. I heard some KS3446xA can have dying SOCs and no one in the world except KS can repair them.
Actually these can be fixed by replacing the chip which is available commercially. But there are always pieces of test equipment that use custom parts. It has been that way since the invention of the silicon chip but nowadays it looks like less custom parts are used (except in really high-end gear) and parts are being run at much lower temperatures. Overall test equipment has become massively more reliable. Don't get discouraged by a few exceptions that affect only a small portion of the total number of devices sold.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf