New Rigol DS1054Z oscilloscope

[avvidclif]

Since the hack sites seem to be disappearing, probably due to Rigol, they may holding off shipment until the hack sites are down.  I guess we'll find out when they finally start shipping again. Or they have a defeat for the hack and want it installed as rapidly as possible and in as many systems.

From what I have read no-one has found a major enough flaw to halt shipment.

 My 2cents worth. 

[DanielS]

The DS1054Z was released a few months back -- has it had a FW update yet?   

The DS1054Z may have been released only two or three months ago but it is the same hardware as the DS1074Z/DS1104Z released nearly two years ago. The biggest differences are the software bandwidth limit and model number stickers.

[Orange]

The DS1054Z was released a few months back -- has it had a FW update yet?   

The DS1054Z may have been released only two or three months ago but it is the same hardware as the DS1074Z/DS1104Z released nearly two years ago. The biggest differences are the software bandwidth limit and model number stickers.

It was announced in September 2013 (see Dave's video, It was made available around oct/dec 2013, So that is ONE year.

Software versions were : 2.01, 2.03SP5, 4.01, and lately a very buggy beta version..............

[pickle9000]

Since the hack sites seem to be disappearing, probably due to Rigol, they may holding off shipment until the hack sites are down.  I guess we'll find out when they finally start shipping again. Or they have a defeat for the hack and want it installed as rapidly as possible and in as many systems.

From what I have read no-one has found a major enough flaw to halt shipment.

 My 2cents worth. 

I'd bet on the firmware issue. If they start delivering with a known firmware issue (even if it's minor) then both Rigol and distributors will be overloaded with return/repair requests. The main issue is that it's a known problem, not that it's major problem with the scope.

Rigol may still be making shipments to dealers willing to upgrade the firmware before shipping to customers. From a dealers perspective they would probably prefer to have Rigol do it.

[DanielS]

I bought a DS-1054Z last week and it seems fairly easy to use even though I've never owned a DSO before.

The biggest challenge to using a multimeter, oscilloscope, spectrum analyzer, logic analyzer and most other basic instruments is knowing what you are looking for. The instruments' fundamental functions usually become largely self-explanatory once you know why you need to use a given instrument for a given task since the two are directly related by physics and maths.

The most "difficult" thing going from an analog to digital scope is getting used to functions being two or three menus deep into the DSO's UI instead of having dedicated buttons or switches on the analog scope.

[tautech]

I'd bet on the firmware issue.

I'm not so sure.
Members Bud & MarkL's investigations in the jitter thread https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/ indicate the PLL might not be engineered precisely or incorrectly implemeted.
Firmware can only suppress any issue here, not eradicate it.
Thats why not all units are affected to the same degree.

[pickle9000]

I'd bet on the firmware issue.

I'm not so sure.
Members Bud & MarkL's investigations in the jitter thread https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/ indicate the PLL might not be engineered precisely or incorrectly implemeted.
Firmware can only suppress any issue here, not eradicate it.
Thats why not all units are affected to the same degree.

That's true of course, but firmware is often used to hide the limitations of hardware. While we may not like it, certain settings are often "locked out" to prevent hardware be used to it's full capacity or because of cost.

I could see a situation where the software detects this (jitter) issue and simply displays a message on the bottom of the screen "Unable to lock", is that sufficient? I don't know that many would like it but it's a valid approach to keep the unit in production. The message has nothing to do with reality but better to have it so another (more reliable) test can be performed. My feeling is that this approach would have been a better first step.

[tautech]

That's true of course, but firmware is often used to hide the limitations of hardware. While we may not like it, certain settings are often "locked out" to prevent hardware be used to it's full capacity or because of cost.

I could see a situation where the software detects this (jitter) issue and simply displays a message on the bottom of the screen "Unable to lock", is that sufficient?

Maybe IF any scope was being used in an advanced manner, certainly not for simple measurements.

Bud has even gone so far as to put his nuts on the line in this post:
https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg564533/#msg564533

Thats real guts IMO
Study the previous posts to see why he has arrived at that conclusion.

Buy a DS1054Z or not?
Information is paramount for this decision.

[marmad]

Bud has even gone so far as to put his nuts on the line in this post:

That's a bit of an overstatement (rather like the one he makes in his post): he's an anonymous poster who has been a member here since April of this year - so I'm not sure what he would lose by being wrong. He makes some valid points in his posts - but he also jumps to conclusions that are not supported by any/much evidence at the moment.

Buy a DS1054Z or not?
Information is paramount for this decision.

There is no doubt that Rigol needs to address these problems. OTOH, people have been managing to use the DS1000Z for their intended purposes for about a year now. Why? Because most people aren't using large trigger offsets or AC-coupled triggers most of the time. The reason the problem(s) have gone unnoticed/unfixed is because the usage which produces the errors isn't common. Even Dave had played around with one for awhile without noticing the problems - he found out from another member here. Heck, Dave has had a DS2000 for 2 years and never noticed the AC-coupled trigger bug - and we had even reported the bug in this forum 

[tautech]

That's a bit of an overstatement (rather like the one he makes in his post): he's an anonymous poster who has been a member here since April of this year .....

So....... we have members with significant expertise joining here everyday.
I haven't seen others give this the same investigation that it rightfully deserves.

If what has been claimed in his posts is indeed correct, IMO it displays a trend in the ongoing reduction in the quality of test quipment in general. Does this not worry us...it does me.
Are not oscilloscopes precision equipment?

Sure we have seen many products hurried to market, only to be exposed as unfinished/incomplete.
Time will tell if this is the case with DS1000 series or there is indeed a design error.

[pickle9000]

That's a bit of an overstatement (rather like the one he makes in his post): he's an anonymous poster who has been a member here since April of this year .....

So....... we have members with significant expertise joining here everyday.
I haven't seen others give this the same investigation that it rightfully deserves.

If what has been claimed in his posts is indeed correct, IMO it displays a trend in the ongoing reduction in the quality of test quipment in general. Does this not worry us...it does me.
Are not oscilloscopes precision equipment?

Sure we have seen many products hurried to market, only to be exposed as unfinished/incomplete.
Time will tell if this is the case with DS1000 series of there is indeed a design error.

It is a design error, no question. It does not matter if it's firmware or hardware related. All new products have issues, Rigol was caught with one (or two or three).

I guess the issue is that if there is a hardware design fault and you can hide it with firmware do you do it? I say yes, correct the hardware in a later board revision and continue on. Future firmware will have to account for this but such is life. Design is about money, not nice but true.

As for the Wallmarting of electronics in general, it's always there, it's just a matter of how far is too far. The lack of communication between the penny pinchers and everyone else is a big issue. There is nothing wrong with pushing hard to save money but not so hard you break the product (and therefore the company).

[marmad]

So....... we have members with significant expertise joining here everyday.
I haven't seen others give this the same investigation that it rightfully deserves.

I wasn't questioning his expertise or sincerity (perhaps he's completely right - or perhaps he's partially right) - just your statement that "his nuts are on the line". To me, that involves telling your boss something really bad, etc, - i.e. 'real-world' consequences.

If what has been claimed in his posts is indeed correct, IMO it displays a trend in the ongoing reduction in the quality of test quipment in general. Does this not worry us...it does me.
Are not oscilloscopes precision equipment?

Well, I would say that even if what he claims is NOT correct, there is definitely an ongoing reduction in the quality of a lot of equipment - both test and otherwise (I've been recently buying semi-professional woodworking machinery and I'm fairly disappointed in the quality).

And cheap DSOs as precision equipment? Sure, I suppose they can be - but I think it boils down to 'you get what you pay for'. Even with all the bugs or unfinished features in some of these modern Chinese DSOs, the amount of power per dollar is pretty astounding. There are some things they do well - some things they do badly - and some things they won't do it all - and as long as you know the limitations of the particular device (and whether the things it can do will fit your needs), you can still get a lot of use from them.

[edavid]

Since the hack sites seem to be disappearing, probably due to Rigol, they may holding off shipment until the hack sites are down.  I guess we'll find out when they finally start shipping again. Or they have a defeat for the hack and want it installed as rapidly as possible and in as many systems.

What hack sites have disappeared?  The ones I know of are still up.

Also, has there been any actual statement that Rigol has stopped shipments, or is that just a guess?

[poida_pie]

OTOH, people have been managing to use the DS1000Z for their intended purposes for about a year now. Why? Because most people aren't using large trigger offsets or AC-coupled triggers most of the time.

My Ds1054Z, running the beta firmware and no installed license codes has a horizontal timebase jitter
of approximately 75-80ns at 15 us offset.
You say most people don't need a decent trigger offset performance.
Have you looked at any decoded serial bus data without browsing forward and backwards from the trigger point along the captured (and decoded) signal? I know I do and I suspect most others do as well.

This jitter problem is not going to disappear just because you may think most people do not need to see a delayed signal. A scenario could be where an event occurs and 15 us later a 1 Mhz clocked spi data stream starts.
my DS1054Z will blur the signal to 8% of the 1 Mhz clock.
I see this as unusable.

[DanielS]

Are not oscilloscopes precision equipment?

The bulk of DSOs out there only have 8bits ADCs. You are not going to get much precision out of those unless you have a repetitive signal and crank averaging up all the way.

I would be more worried about accuracy: making sure what I am seeing is actually representative of what is really there.

[marmad]

You say most people don't need a decent trigger offset performance.

I never wrote that (nor would I)...

This jitter problem is not going to disappear just because you may think most people do not need to see a delayed signal.

...and I certainly never wrote that.

What I did write was that I thought the reason the problem hadn't been reported in the last year was that it wasn't a capability many owners were often using. The other possibilities are that the 5us jitter is something limited to the more recent HW revision(s) of the main board - or - that it is only severe in a limited number of cases.

[leppie]

OTOH, people have been managing to use the DS1000Z for their intended purposes for about a year now. Why? Because most people aren't using large trigger offsets or AC-coupled triggers most of the time.

My Ds1054Z, running the beta firmware and no installed license codes has a horizontal timebase jitter
of approximately 75-80ns at 15 us offset.
You say most people don't need a decent trigger offset performance.
Have you looked at any decoded serial bus data without browsing forward and backwards from the trigger point along the captured (and decoded) signal? I know I do and I suspect most others do as well.

This jitter problem is not going to disappear just because you may think most people do not need to see a delayed signal. A scenario could be where an event occurs and 15 us later a 1 Mhz clocked spi data stream starts.
my DS1054Z will blur the signal to 8% of the 1 Mhz clock.
I see this as unusable.

Mine peaks around 16-17us as well with the beta FW.

Can you check if it becomes good in multiples of ~33us?

A scenario could be where an event occurs and 15 us later a 1 Mhz clocked spi data stream starts.
my DS1054Z will blur the signal to 8% of the 1 Mhz clock.
I see this as unusable.

That is a piss poor use case ;p Unless of course you are not analysing any of the data and just watching continuous signals, but who does that with SPI?     (you wont really have issues with decoding and using a non-auto trigger)

[pa3bca]

What I did write was that I thought the reason the problem hadn't been reported in the last year was that it wasn't a capability many owners were often using. The other possibilities are that the 5us jitter is something limited to the more recent HW revision(s) of the main board - or - that it is only severe in a limited number of cases.

..which is certainly possible. My 1074Z (spring 2014) does NOT exhibit the n*5 us trigger issue. And I have used the trigger offset extensively when decoding (and especially troubleshooting SW implementations of) SPI and I2C signals...

[DanielS]

..which is certainly possible. My 1074Z (spring 2014) does NOT exhibit the n*5 us trigger issue. And I have used the trigger offset extensively when decoding (and especially troubleshooting SW implementations of) SPI and I2C signals...

Some people have noticed jitter worse than 200ns, others around 100ns, yet more under 50ns and the lucky ones report none. There clearly is a wild variance and luck-of-the-draw component behind it; maybe an issue with the ADC clock generator PLL's drift and jitter combined with the reference crystal's own and the tolerances of all the other components around the PLL/VCO.

Rigol may have rounded off one too many corners there and got wildly inconsistent yields for this particular use-case as a result.

[ruffy91]

The swiss distributor for rigol (electronic.maxdata.ch) told me that a container with Rigol instruments (including DS1054z) arrives in germany on 17. Dec. So I should have my DS1054z on 23. Dec. I'll report wheter I have the jitter or not.

[aroby]

Are not oscilloscopes precision equipment?

The bulk of DSOs out there only have 8bits ADCs. You are not going to get much precision out of those unless you have a repetitive signal and crank averaging up all the way.

I would be more worried about accuracy: making sure what I am seeing is actually representative of what is really there.

I'd asked that same question a while ago about my DS2072A (https://www.eevblog.com/forum/testgear/ds2072a-vmax/msg428068/#msg428068) which doesn't display the correct voltages, apparently because of the ADC precision.

[poida_pie]

Mine peaks around 16-17us as well with the beta FW.

Can you check if it becomes good in multiples of ~33us?

A scenario could be where an event occurs and 15 us later a 1 Mhz clocked spi data stream starts.
my DS1054Z will blur the signal to 8% of the 1 Mhz clock.
I see this as unusable.

That is a piss poor use case ;p Unless of course you are not analysing any of the data and just watching continuous signals, but who does that with SPI?     (you wont really have issues with decoding and using a non-auto trigger)

My unit peaks around 28us. See
https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg558628/#msg558628
for screen captures.

Maybe it's time to feed yours and my DS1054Z with an SPI data stream clocked around 1 and up to 8 MHz (why 8? Arduino max SPI clock speed. remember this DSO is targeted at hobbyists)
and see if it can decode at a few delay positions. It might be fine even with the large jitter.

[leppie]

My unit peaks around 28us. See
https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg558628/#msg558628
for screen captures.

Maybe it's time to feed yours and my DS1054Z with an SPI data stream clocked around 1 and up to 8 MHz (why 8? Arduino max SPI clock speed. remember this DSO is targeted at hobbyists)
and see if it can decode at a few delay positions. It might be fine even with the large jitter.

Thanks, seems your might even be worse than mine Edit: Actually not! I assume you mean repeat, and not peak from the screens. Mine repeats around ~33us.

Here are mine: https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg558863/#msg558863

I have not done anything really after the beta FW, but before I was quite happily dealing with 8MHZ SPI decoding on a delayed trigger, in single shot mode. No issues experienced. Was using a STM32L0x in my case.

[David Hess]

Are not oscilloscopes precision equipment?

The bulk of DSOs out there only have 8bits ADCs. You are not going to get much precision out of those unless you have a repetitive signal and crank averaging up all the way.

And even this will not improve linearity or settling time.  Most of the benefit from averaging or high resolution mode is in reducing noise.

[Rodot]

Hello,

I've been following EEVblog on YouTube for a while now, it's my first time on this forum as I have a few questions about this oscilloscope :

I'm from France and can't find the DS1054Z in stock anywhere in Europe, all sellers seem to be out of stock. Batter Fly are out of stock too since a few days ago, I'm waiting for an answer from arBenelux but I don't think they ship to France. Zeitech seem to have it in stock but with a 4 weeks delay (do you know if it's usually faster?).
Ideally, I would need this oscilloscope within 8-15 days. Thanks for reading !