Author Topic: New Rigol DS1054Z oscilloscope  (Read 2043117 times)

0 Members and 7 Guests are viewing this topic.

Online JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
Re: New Rigol DS1054Z oscilloscope
« Reply #2825 on: August 24, 2016, 06:21:32 am »
One thing I don't think I saw mentioned: how is the UI responsiveness with 04.04.00.07 compared with 04.03.01.05 (04.03.SP1)?  I've intentionally kept my scope on 04.03.01.05 in order to avoid the issues with SP2 (especially the UI issues), but if the UI responsiveness is the same for both 04.04 and 04.03.SP1 then it sounds like there's little reason to remain on 04.03.SP1.
No improvement noticed. Still slow as hell

***
I don't think for a second that the delay on filtering is a bug. You should know that math is calculated from data on the screen buffer.
i don't know how exactly they are filtering, probably the same fir which taps gets 'compressed' for higher cutoffs so the delay between the first input and the first output increases... before that there is no data.
« Last Edit: August 24, 2016, 06:26:37 am by JPortici »
 

Offline IAmBack

  • Regular Contributor
  • *
  • Posts: 162
  • Country: pl
Re: New Rigol DS1054Z oscilloscope
« Reply #2826 on: August 24, 2016, 05:53:27 pm »
***
I don't think for a second that the delay on filtering is a bug. You should know that math is calculated from data on the screen buffer.
i don't know how exactly they are filtering, probably the same fir which taps gets 'compressed' for higher cutoffs so the delay between the first input and the first output increases... before that there is no data.

It is a bug. Rigol used non-casual filtering algorithm (which means that response is calculated not only with "past" samples but also "future" samples - when You look at the plot, it reacts before the occurence of input signal change. If the oscilloscope's reaction is the result of data processing, plot should be truncated on both sizes of the screen.
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: New Rigol DS1054Z oscilloscope
« Reply #2827 on: August 25, 2016, 01:50:25 am »
That doesn't make a lot of sense. Yes, it is a bug. Yes, it is a _digital sampling oscilloscope_ so _everything_ you see on the screen is a result of data processing. The result of any math operation should be displayed as properly synchronized with the trace data that is being operated upon.
The easiest person to fool is yourself. -- Richard Feynman
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 2199
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2828 on: August 25, 2016, 02:25:35 am »
yeah. and it is interesting in the options you can set the start and end point of the filter. Totally cool!!!

Maybe it's me, but why would using a low pass filter set to 20 khz cause so much ripple on a 1 khz square wave? Lowering the filter smooths it out, and that makes sense. OK, maybe it makes sense, I guess, if the 20th (or so) harmonic really contributes to the squareness - I was just thinking of the discussion of how much bandwidth is required to show a square wave as a square...
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1714
  • Country: se
Re: New Rigol DS1054Z oscilloscope
« Reply #2829 on: August 25, 2016, 07:25:54 am »
yeah. and it is interesting in the options you can set the start and end point of the filter. Totally cool!!!

Maybe it's me, but why would using a low pass filter set to 20 khz cause so much ripple on a 1 khz square wave? Lowering the filter smooths it out, and that makes sense. OK, maybe it makes sense, I guess, if the 20th (or so) harmonic really contributes to the squareness - I was just thinking of the discussion of how much bandwidth is required to show a square wave as a square...
The shape depends heavily on the phase and roll-off characteristics of the filter.
A simple RC-like filter (1st order, slow roll off, phase shift) will give a "smoothed" wave, but a perfect brick-wall, no phase shift filter gives you a rippling waveform, see the attachment.
Such a filter is impossible in reality as its time-domain response extends infinitely in the future and in the past, so it can only be approximated (easier in the digital world, I would say).
For a nice animation, see Gibbs phenomenon.
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 2199
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2830 on: August 25, 2016, 03:25:24 pm »
Thanks!

I also noticed in Options there are a few fx operations, but not understanding what they are supposed to do. For example, the selection is already A+B, but changing the operation to A-B or any other seems to have no effect on the math trace (when using low pass filter as shown by alsetalokin4017)
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 9057
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: New Rigol DS1054Z oscilloscope
« Reply #2831 on: August 25, 2016, 05:25:09 pm »
See page 6-22 of the manual.

Unless you have selected a Math Operator function (e.g., Intg) in the Operator menu that uses the output of the fx Operator function in the Options menu, changing the latter won't have any effect on the former.
TEA is the way. | TEA Time channel
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16548
  • Country: us
  • DavidH
Re: New Rigol DS1054Z oscilloscope
« Reply #2832 on: August 25, 2016, 06:59:32 pm »
The shape depends heavily on the phase and roll-off characteristics of the filter.
A simple RC-like filter (1st order, slow roll off, phase shift) will give a "smoothed" wave, but a perfect brick-wall, no phase shift filter gives you a rippling waveform, see the attachment.
Such a filter is impossible in reality as its time-domain response extends infinitely in the future and in the past, so it can only be approximated (easier in the digital world, I would say).
For a nice animation, see Gibbs phenomenon.

I probably got caught by this when evaluating a Tektronix MSO5204 a couple years ago.  It has both hardware and DSP bandwidth limiting and to my surprise, they did *not* match at high frequencies with the DSP filters displaying what was probably the Gibbs phenomenon do to their steep rolloff.  At least the Tektronix engineers could not explain it either; I suspect they were not used to dealing with a prospective customer who pulls a 250ps reference level pulse generator designed specifically for oscilloscope evaluation out of their pocket.
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 2199
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2833 on: August 25, 2016, 07:35:55 pm »
See page 6-22 of the manual.

Unless you have selected a Math Operator function (e.g., Intg) in the Operator menu that uses the output of the fx Operator function in the Options menu, changing the latter won't have any effect on the former.

Perhaps they should then be grayed out, like some of the other options. It was not really clear what would be the point of those options anyway. I'm not near the scope, but aren't these settings already on the top level menu for the other math operators (and BTW, filter is under the Math menu, so assumed it is one of the math functions)?

I think someone made a cheat sheet for all the menus - a menu tree, yes?
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 9057
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: New Rigol DS1054Z oscilloscope
« Reply #2834 on: August 25, 2016, 08:31:26 pm »
Perhaps they should then be grayed out, like some of the other options. It was not really clear what would be the point of those options anyway.
Yeah, lots of equipment have interaction design issues.

Quote
I'm not near the scope, but aren't these settings already on the top level menu for the other math operators (and BTW, filter is under the Math menu, so assumed it is one of the math functions)?

The manual explains how the math functions operate and, more specifically, how the fx Operators layer into them. Check out the example they give on the page I referred to.
TEA is the way. | TEA Time channel
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2835 on: August 26, 2016, 06:04:59 am »
I probably got caught by this when evaluating a Tektronix MSO5204 a couple years ago.  It has both hardware and DSP bandwidth limiting and to my surprise, they did *not* match at high frequencies with the DSP filters

Blasphemy!
 

Offline quetzalcoatl

  • Newbie
  • Posts: 8
  • Country: pl
Re: New Rigol DS1054Z oscilloscope
« Reply #2836 on: August 26, 2016, 09:45:50 am »
I have read the SCPI commands automatically from the SDRAM.
Below is the list of commands from the firmware 00.04.03.01.05 (00.04.03.SP1).

A few commands are undocumented in the Programming Manual.
The parameters necessary to determine itself.

Peter

Edit: Undocumented commands are indented with a tab.
It was compared to the programming manual of December, 2015.
Some commands are described in the programming manual from the DS2000.

I'm currently doing researching programming interfaces of DS1054Z, including commands. Searching for strings is trivial, but that returns just parts of the commands. I'm curious - how did you manage to rebuild the full commands? did you discover some pointer tables or lists that point to partial strings? Could you share how they are structured?

I some time ago I have browsed through 00.04.03.01.05 image contents, and although I have not managed to rebuild a full command set, I have found out a few more commands than you have published now. It may mean that there are several command sets in the memory. You could try running your method on the other ones and see what else it finds.

Also, I have discovered/guessed meaning of some of the undocumented commands and spent a long time investigating curious limitations of :WAV:DATA? transfers - I've read that people have a hard time reading 24Mpts with transfer speeds like 100Kpts samples per sec, the 2015 programming guide says that max raw transfer speed is 250Kpts in one block read. :WAV:DATA? is unintuitive in a few aspects, but I'm currently successfully reading the data using >1Mpts blocks (NI-VISA drivers) and I'm quite suprised I can't find any information on it, neither here on forum, nor on the internet. But maybe just my googling skills suck?

Naturally I was going to write a blog post about all of my findings (and probably I will anyways*), but now after finding your command list it got me thinking - is there any wiki/repo/etc where we could gather&merge such knowledge and tools?

*) but looking back at my years-old and still unpublished research/tools on Hantek DSO2090.. ..it takes me a while sometimes:)
« Last Edit: August 26, 2016, 09:51:34 am by quetzalcoatl »
 

Offline freebil

  • Regular Contributor
  • *
  • Posts: 50
  • Country: gr
Re: New Rigol DS1054Z oscilloscope
« Reply #2837 on: August 26, 2016, 10:50:42 am »
Hello. I am trying to read more than 1212 points in matlab but I cant.

Code: [Select]
DSO1000Z = visa('ni','TCPIP0::192.168.1.101::INSTR');
DSO1000Z.InputBufferSize = 2048;
fopen(DSO1000Z);
fprintf(DSO1000Z, ':wav:data?' );
data= fread(DSO1000Z,2048);

Matlab reads 1212 points whatever input buffer size I enter. Any tips?
 

Offline quetzalcoatl

  • Newbie
  • Posts: 8
  • Country: pl
Re: New Rigol DS1054Z oscilloscope
« Reply #2838 on: August 26, 2016, 11:12:12 am »
Hello. I am trying to read more than 1212 points in matlab but I cant.

Code: [Select]
DSO1000Z = visa('ni','TCPIP0::192.168.1.101::INSTR');
DSO1000Z.InputBufferSize = 2048;
fopen(DSO1000Z);
fprintf(DSO1000Z, ':wav:data?' );
data= fread(DSO1000Z,2048);

Matlab reads 1212 points whatever input buffer size I enter. Any tips?

if you get max ~1200 points, it means that you are reading from the screen buffer, not from the main memory. check out the programming guide, it is explained there. In short:

- if the device is in RUN mode, only screen buffer is available to read, it contains, wel, the data on the screen, which is aggregated and trimmed down to the whereabouts of the screen resolution
- once the device is in STOP mode, you have to send :WAV:MODE and :WAV:FORM commands to pick RAW reading mode, and then you will be able to read from "internal memory" buffer and get the actual full readout

see the 2015 programming guide, not 2013 not 2014. 2015 version has much more information, including how to read the memory part-by-part
« Last Edit: August 26, 2016, 11:14:16 am by quetzalcoatl »
 

Offline quetzalcoatl

  • Newbie
  • Posts: 8
  • Country: pl
Re: New Rigol DS1054Z oscilloscope
« Reply #2839 on: August 26, 2016, 11:53:27 am »
I'm currently doing researching programming interfaces of DS1054Z, including commands. Searching for strings is trivial, but that returns just parts of the commands. I'm curious - how did you manage to rebuild the full commands? did you discover some pointer tables or lists that point to partial strings? Could you share how they are structured?

I found a table and wrote a program to read with Xojo.
I can not enough English to describe the table exactly.
I understood the fields not complete. However, it is sufficient to create the commands.

By reading out the wave data I've done experiments, see picture.

Thanks for sharing it! I looked at the table and it's close to my results, but not exactly the same.

The're many complications to explain how exactly the data-length work.. I'll try to summarize it shortly

There are at least three types of it:
(RAW/BYTE reading mode)
- full-width (blocksize=64b): 1179584 .. 1179647 *)
- half-width (blocksize=32b): 589792 .. 589823
- quarter-width (blocksize=16b): 294896 .. 294911
- **)

Depending on certain conditions, your device may happen to be in of those three states (full, half, quarter), and the single-read data-length limit will be in the range as above, eaxct value will vary depending on the START position you send to the device. So, if the device is in quarter mode, if you send one start position you can get i.e. max 294899 bytes at once from that pos, but if you send different START position, you may get i.e. max 294905 bytes at once from that pos.

It actually fluctuates in a deterministic way, but it's tricky on many levels:
- the varying max-data-length follows a certain pattern
- but that pattern shifts randomly after each measurement
- and also the 'full/half/quarter' mode of the device is quite nasty thing, because you can't easily read it (you have to check few things and deduce it), and can't easily configure it (it is selected automatically at BOOT TIME depending on the number of selected channels at BOOT TIME, and seems to be locked until SHUTDOWN -!aargh!-)

However, I've got many aspects of that analyzed already. I'll write a post about it during the weekend and write back here.

Anyways if there are any pages wikis or other threads I missed I'd be grateful for links or contact to mantainers :)

*) in your table there's a 1179648 value - I'm pretty sure it is not possible, you probably got a +-1 error when you calculated it
**) I found out only these modes, because I worked in RAW/BYTE combinations and certain settings. It is possible that there are more modes after 1/4, for example like 1/8 (147k), 1/16 (73k) etc. - I will later compare my settings with your table to see ifI can get them. However I think they don't exist, becuse the device has only 4 channels, so there's no need for more divisions than /1,/2,/4 that I already found
« Last Edit: August 26, 2016, 11:58:01 am by quetzalcoatl »
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 9057
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: New Rigol DS1054Z oscilloscope
« Reply #2840 on: August 26, 2016, 05:06:46 pm »
Some threads? That's an impressive index, PeDre!
TEA is the way. | TEA Time channel
 

Offline mnementh

  • Super Contributor
  • ***
  • Posts: 17541
  • Country: us
  • *Hiding in the Dwagon-Cave*
Re: New Rigol DS1054Z oscilloscope
« Reply #2841 on: September 01, 2016, 07:06:44 pm »
Sooooo.... has the 500uV bug been fixed in firmware yet?

*Ducks soggy ol' boot thrown in his general direction*

Guess not... ;)


mnem
*agitating-ily*
alt-codes work here:  alt-0128 = €  alt-156 = £  alt-0216 = Ø  alt-225 = ß  alt-230 = µ  alt-234 = Ω  alt-236 = ∞  alt-248 = °
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
  • Older New "New Player" Player Playa'
Re: New Rigol DS1054Z oscilloscope
« Reply #2842 on: September 01, 2016, 10:06:14 pm »
I get the same filter delay on 04.03.01.05. Unfortunately, I can't save a screenshot right now because the scope doesn't like my made-in-China thumb drive. Irony? :-DD

All the buttons stop functioning after inserting the drive. It's a 2GB formatted using FAT32 and I tried 1K, 2K and 4K cluster sizes. Oh, well. I'll have to find the one I used for firmware updates before. That one worked.

Anyway, the delay looks the same as what IAmBack posted.

This may not be useful advice, but check to make sure you have the USB option set to Computer instead of Printer.
 

Offline ceoxrad

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
Re: New Rigol DS1054Z oscilloscope
« Reply #2843 on: September 03, 2016, 07:05:40 pm »
Hi, first post here so if I'm doing something wrong please be kind  :)
I recently purchased a rigol ds1054z and I want to unlock it.
I've done some reading about it on the forum, the content in there is really a lot and I still have some doubts about it:
My scope came with SW version 00.04.03.SP2, if I hack it can I upgrade the firmware after it? (for example to 00.04.04)
If I upgrade before the hack, will it work on 00.04.04? Is the upgrade worth it?
I'm sorry to ask so many questions in my first post, and please forgive my poor English.
Thank you in advance
Marco
Sorry for any mistake in my English
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2844 on: September 03, 2016, 07:23:27 pm »
Hi, first post here so if I'm doing something wrong please be kind  :)
I recently purchased a rigol ds1054z and I want to unlock it.
I've done some reading about it on the forum, the content in there is really a lot and I still have some doubts about it:
My scope came with SW version 00.04.03.SP2, if I hack it can I upgrade the firmware after it? (for example to 00.04.04)
If I upgrade before the hack, will it work on 00.04.04? Is the upgrade worth it?
a) Yes.
b) Yes.
c) Definitely!

please forgive my poor English.

It was perfect.
 

Offline ceoxrad

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
Re: New Rigol DS1054Z oscilloscope
« Reply #2845 on: September 03, 2016, 07:31:59 pm »
Thank you very much for your response, i'm upgrading right now
Sorry for any mistake in my English
 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 513
  • Country: de
  • What is business? It’s other people’s money
Re: New Rigol DS1054Z oscilloscope
« Reply #2846 on: September 03, 2016, 08:03:43 pm »
I did just some weeks ago.
I newly received the DS1054Z, had firmware 00.04.03.SP2.
Unlocked all features (except the 500uV option, as described in the forum (quite an easy task)
Updated to new firmware 00.04.04
Works fine.
So don't worry.
You will learn something new every single day
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2847 on: September 03, 2016, 08:20:36 pm »
Unlocked all features (except the 500uV option, as described in the forum (quite an easy task)

Yes, the 500uV option doesn't work. Best not to unlock that so you can't select it accidentally.

Rigol can't ever stop these codes from working because they sell them as upgrades. They can't put out a firmware update that stops legitimate keys from working.

(Although the bandwidth isn't an option, I guess they could block that in theory. I don't think there's a way for the BIOS to know what the hardware is though - if it could decide the bandwidth based on a hardware difference then there wouldn't be an unlock. Either way, don't worry).

PS: I'd use the 'scope for a few days before unlocking it, just in case there's any problems and you need to send it back. All the features except bandwidth are unlocked anyway so you can play with them.
 

Offline ceoxrad

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
Re: New Rigol DS1054Z oscilloscope
« Reply #2848 on: September 03, 2016, 08:55:18 pm »
I did the upgrade, I hacked all option except the 500uV (DSER) and finally an autocal. Everything works perfectly and I haven't had any problems during the procedure.
As you said (I've haven't seen the post since I was doing the work in that precise moment) I should have waited to do the unlocking, but I couldn't resist.. during my inital check everything worked and now there is no apparent change.
BTW In the unfortunate event of something wrong I can revert the hack with the SCPI console of Ultrasigma, can't I?
I also quickly "checked" the bandwidth with a fast risetime generator (based on a 74AC14) and the rise time changed from approx 3.2ns to 2.4ns.
Overall I'm super happy with my new purchase,this product feels very good, it's not even comparable with my old scope (beaten up 20MHz analog).
Thank you all!

Edit: forgot the ic code
« Last Edit: September 03, 2016, 08:57:16 pm by ceoxrad »
Sorry for any mistake in my English
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #2849 on: September 03, 2016, 09:05:49 pm »
In the unfortunate event of something wrong I can revert the hack with the SCPI console of Ultrasigma, can't I?

Yes.

(Assuming it still powers on normally... :popcorn: )

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf