New Rigol DS1054Z oscilloscope

[Hydron]

Dave reverse engineered the front end, it's almost all discrete, with the only IC in the path being an op-amp for the DC component. There is a relay switching in an attenuator before the buffer, no other gain control.

http://www.eevblog.com/files/Rigol-DS1054Z-Schematic-FrontEnd.pdf
http://www.eevblog.com/files/Rigol-DS1054Z-Schematic-DiffAmp.pdf

[David Hess]

I had to watch Dave's video again and I guess you are right.  His schematic ends where the integrated programmable gain amplifier would be but in the video you can see that there is nothing between that output and the ADC so Rigol must be relying on the ADC for all of the gain switching except for the high impedance attenuator.

[metrologist]

Had to use the scope outside in the sun. It was not happening, not even trying to shade it. The display cannot be discerned at all. 

[frozenfrogz]

That’s the thing with vampires and other super human creatures: They got awesome super powers, but sunlight turns the to dust / water turns them to stone / silver or kryptonite sucks up strength / ...

It is not the brightest display out there, but at least it has proper antireflective glass on top of the LCD

Since it is not battery powered I don’t see it as a design flaw. Maybe consider this for outdoor use:

[sandor626]

rigol DS1054Z bandwidth : I measured it a risetime of 2,95 ns

[xrunner]

rigol DS1054Z bandwidth

You know that thing can save a screen shot to the USB port don't you? 

[Fungus]

rigol DS1054Z bandwidth : I measured it a risetime of 2,95 ns

Unlocked to 100MHz, I presume.

[nanofrog]

rigol DS1054Z bandwidth : I measured it a risetime of 2,95 ns

Unlocked to 100MHz, I presume.

FWIW, unlocking adds the features that are only available as 30 day trial runs IIRC (one could argue about both the memory increase as well as the scope's BW however).

[ebastler]

FWIW, unlocking adds the features that are only available as 30 day trial runs IIRC (one could argue about both the memory increase as well as the scope's BW however).

What's to argue? Unlocking codes (purchased or generated by Riglol) can be used to permanently enable all options originally incuded as time-limited trials. These include decoders, advanced triggering, and also the increased memory capacity.

The bandwidth upgrade is not available from Rigol as an official option -- neither in the trial licenses nor for purchase -- but can be enabled via a Riglol-generated key.

[nanofrog]

Unlocking codes (purchased or generated by Riglol) can be used to permanently enable all options originally incuded as time-limited trials.
These include decoders, advanced triggering, and also the increased memory capacity.

Quite true; I wasn't paying sufficient attention to the details of my post.  My deepest apologies. 

Perhaps I need to skip the single malt scotch before making any posts.   

[sandor626]

another signal hooked , rise time 2.550 ns , BW > 130 MHZ

[sandor626]

rigol DS1054Z bandwidth : I measured it a risetime of 2,95 ns

Unlocked to 100MHz, I presume.

of course , 

[David Hess]

another signal hooked , rise time 2.550 ns , BW > 130 MHZ

You hopefully need a better test signal; 50% overshoot is too much for that test and if that is the actual transient response of your hacked DS1054Z oscilloscope, then it is broken.  Greater bandwidth is available with a "peaked" transient response and some cheap probes and oscilloscopes actually do this.

I know of one example where it was done deliberately.  Tektronix made two versions of the 7704A oscilloscope; the normal one had a 200MHz bandwidth with the expected normal Gaussian response but it could be ordered with an option to extend the bandwidth to 250MHz sacrificing clean transient response.

A x10 probe which has high frequency transient response adjustments can be tweaked to do this deliberately.

[JPortici]

If i saw that signal i would be worried. what is wrong?

the way i probed? something in my setup? probe? oscilloscope channel?

or (cold sweat on the back) something in the circuit?

[TurboTom]

I've got my DS10x4Z already for a few years but only now got to check the rise time with a half-way decent square source -- my recently constructed Rb 10MHz source with multiple outputs, among them three pretty fast 74LVC drivers with a symmetrizised (sp?) square. These outputs directly connected to the DS...Z's input via RG58 and a proper terminator result in rise/fall times round about 1.6....1.7ns, single channel enabled. See attached screenshots/files for details. I didn't expect the "lil'Rigol" to be so fast. Impressive!

Cheers,
Tom

[Fungus]

If i saw that signal i would be worried. what is wrong?

It's probably a problem with probing.


(nice alliteration...)

[sandor626]

I've got my DS10x4Z already for a few years but only now got to check the rise time with a half-way decent square source -- my recently constructed Rb 10MHz source with multiple outputs, among them three pretty fast 74LVC drivers with a symmetrizised (sp?) square. These outputs directly connected to the DS...Z's input via RG58 and a proper terminator result in rise/fall times round about 1.6....1.7ns, single channel enabled. See attached screenshots/files for details. I didn't expect the "lil'Rigol" to be so fast. Impressive!

Cheers,
Tom

 

[sandor626]

If i saw that signal i would be worried. what is wrong?

It's probably a problem with probing.


(nice alliteration...)

Attention, I never said that the test signal was the one provided by the oscilloscope to calibrate the probes!  Of course I tried with an external signal

[DavidMassey]

I've had my DS1054Z for about a week now and am amazed at what it can do.  Right now I've got it set up to calibrate the head alignment of an old TEAC A-6300 reel-to-reel tape deck.  I'm feeding the Left line out from the deck to Channel 1 on the scope and Right line out from the deck to Channel 2 on the scope.  My question is, is it possible to lock the V/division of Channel 1 and Channel 2 together so if I want to adjust the gain (V/division) of the vertical I don't have to individually select Channel 1 and Channel 2 to change the vertical gain to match each other.

[JPortici]

I've got my DS10x4Z already for a few years but only now got to check the rise time with a half-way decent square source -- my recently constructed Rb 10MHz source with multiple outputs, among them three pretty fast 74LVC drivers with a symmetrizised (sp?) square. These outputs directly connected to the DS...Z's input via RG58 and a proper terminator result in rise/fall times round about 1.6....1.7ns, single channel enabled. See attached screenshots/files for details. I didn't expect the "lil'Rigol" to be so fast. Impressive!

Cheers,
Tom

Impressive indeed, but what happens if you expand the timebase so that you can see the most of the rising/falling edge? remember that this scope makes measurements based on the screen's pixels (measurements are made on screen buffer) and not on acquired data

[ebastler]

Impressive indeed, but what happens if you expand the timebase so that you can see the most of the rising/falling edge? remember that this scope makes measurements based on the screen's pixels (measurements are made on screen buffer) and not on acquired data

 

The 5ns/div time base TurboTom used (in the second and third screenshot) is the fastest timebase the DS1000Z series offers. The rise time is clearly a fraction of one division, approx. 1/3 of a division. So I don't think the fast rise time is an artefact of the on-screen measurement in any way. (More data points or a faster time base would give you a more precise measurement of the rise time, but won't make it larger than 1.7 ns.)

[ebastler]

I've had my DS1054Z for about a week now and am amazed at what it can do.  Right now I've got it set up to calibrate the head alignment of an old TEAC A-6300 reel-to-reel tape deck.  I'm feeding the Left line out from the deck to Channel 1 on the scope and Right line out from the deck to Channel 2 on the scope.  My question is, is it possible to lock the V/division of Channel 1 and Channel 2 together so if I want to adjust the gain (V/division) of the vertical I don't have to individually select Channel 1 and Channel 2 to change the vertical gain to match each other.

No, 'fraid not. I don't think this is a common function in DSOs. If you need to change vertical scales a lot (why?), scopes with individual vertical controls for each channel do have a distinct advantage...

[RoGeorge]

...My question is, is it possible to lock the V/division of Channel 1 and Channel 2 together so if I want to adjust the gain (V/division) of the vertical I don't have to individually select Channel 1 and Channel 2 to change the vertical gain to match each other.

I never heard about, or saw such a feature on any oscilloscope ever, but it can be done very easy by sending SCPI commands from a computer. Here is some doc about how to do it without installing anything, using LXI: https://hackaday.io/project/6857-master-your-rigol-from-command-line

You can use this technique to send SCPI commands from a terminal, like in this one that saves a capture of the oscilloscope screen without installing any software or drivers: https://rogeorge.wordpress.com/2017/01/10/ds1054z-tricks-less-known-or-undocumented-part-2/

For changing both CH1 and CH2 to 1V/div, that would become something like (assuming you give the IP 192.168.1.3 to your scope):

Code: [Select]

echo ":CHANnel1:SCALe 1" | nc -w1 192.168.1.3 5555
echo ":CHANnel2:SCALe 1" | nc -w1 192.168.1.3 5555

You can create many files with a batch of SCPI commands, one for each desired vertical sensitivity, and run them from your computer when you want to change the scale simultaneously on both CH1 and CH2.

[Karel]

Or via USB. No additional software or IP-address needed. In a console just enter:

Code: [Select]

echo ":CHAN1:SCAL 1" > /dev/usbtmc0
For all possible commandline options, have a look here:

http://beyondmeasure.rigoltech.com/acton/attachment/1579/f-0386/1/-/-/-/-/DS1000Z_Programming%20Guide_EN.pdf

[Fungus]

Impressive indeed, but what happens if you expand the timebase so that you can see the most of the rising/falling edge? remember that this scope makes measurements based on the screen's pixels (measurements are made on screen buffer) and not on acquired data

The 5ns/div time base TurboTom used (in the second and third screenshot) is the fastest timebase the DS1000Z series offers.

Yes, but certain people here need to hate the DS1054Z at every opportunity.