REVIEW - Rigol DS2072 - First Impressions of the DS2000 series from Rigol

[Electro Fan]

I found another new feature in FW v.3 - something we've been asking them for over the last year - and although it's a baby step by Rigol, at least it's a baby step in the right direction:

The External Trigger input can now (finally) be used for a second trigger type: PULSE
- how would you wire this up to demonstrate this?

This would help explain Teneyes' earlier discovery that the Ext. Trigger can now be used as a source for the counter: it seems to go hand-in-hand with adding pulse measuring capabilities.
- how would you wire this up to demonstrate this?

An encouraging sign... let's hope they keep adding more triggers it can be used as a source for.

Thanks

[Teneyes]

the Ext. Trigger can now be used as a source for the counter:
- how would you wire this up to demonstrate this?

Here is 1MHz on Ch1 , with  a 250KHz ,phase locked to the 1Mhz and connect to the External Trigger.
Note:  The Trigger is set to Rising edge on the External input. but the trigger point is the falling part of the Sine wave.
# 2 showing what was on the External input

[Circlotron]

Teneyes, can you show us the same test using square waves, say from a synchronous binary counter chip? If there is a difference there, that would be serious.

[marmad]

Teneyes, can you show us the same test using square waves, say from a synchronous binary counter chip? If there is a difference there, that would be serious.

What do you mean? Serious in what way?

[Circlotron]

It would mean something is way wrong. With a sinewave input maybe there is some phase shift in the external analog circuit, but inputting two square waves with known identical corresponding transitions every 4 cycles, only different in frequency, it is a very easy test to reproduce, and if an anomaly shows up, then yeah, theres problems alright.

[marmad]

It would mean something is way wrong. With a sinewave input maybe there is some phase shift in the external analog circuit, but inputting two square waves with known identical corresponding transitions every 4 cycles, only different in frequency, it is a very easy test to reproduce, and if an anomaly shows up, then yeah, theres problems alright.

There is no anomaly from the Rigol in the images he posted. It's triggering exactly where it's supposed to: on the zero-crossing rising edge of the External Trigger input (or CH2, when he made the image to demonstrate the previous one). Any phase shift is, obviously, external.

So I don't see your point: you want another test to show it operates correctly again?

[marmad]

Here's a repeat of Teneyes' demo, except using EXT. Trig as the source to a PULSE trigger.

1MHz sine to CH1 , with the AWG Sync Out signal connected to the External Trigger.
The Trigger is set for a positive PULSEs > 490ns.
Image #2 shows the same triggering replicated with CH2:

[Teneyes]

Teneyes, can you show us the same test using square waves, say from a synchronous binary counter chip? If there is a difference there, that would be serious.

@Circlotron & Others
  In my last report of a External trigger I used a sine wave to show how the external trigger was independant of the CH1 & Ch2 inputs.
 Below I repeated the test with two square waves the are Phase Aligned.
Note: I am using a DG4000 with 2 outputs. (lock)
In the display #1 you can see the 4 Cycles on Ch1 locked to the 1 cycle on Ch2
and in the Zoom part you can see the trigger  on Ch 2 works very accurately.

I then disconnected the 250KHz signal off of Ch2 input and connected it to the External Trigger input And the display #2 show the result.
CH2  in disconnected but still on.
Note that there appears to be about a  9ns offset between the "EDGE" External Trigger point and the CH1 trace.
Can anyone else verify this.  Just use a "T" connector and feed a square wave signal to both CH1 & External Trigger. Set the trigger source to EXT.

Also there appears to be about a  13ns offset between the "Pulse" External Trigger point and the CH1 trace.


I would think this trigger to trace offset could be compensated in the Firmware to display the Traces correctly when on external triggering. Is this a Bug? Is it only on the DS2000?


Please note that the External trigger requires an input >280mV to trigger and Count.
so if you see the Counter not working then the trigger will be jittering at first then NOT triggering at lower input signal.
There is not spec in the manual to the External Trigger sensitivity

[Orange]

Teneyes,

I can confirm the delay of 9nS, also I see jitter of 500nS if I trigger externally.

On old Tek scopes you sometimes had the possibility to view the external trigger signal. Not so with the Rigol DS2000 series....

 

[Orange]

also I see jitter of 500nS if I trigger externally.

I see jitter on a weak signal. < 280mVpp
What was the amplitude of the signal in your test?

I tested with Sin Wave 100mV RMS, and with Sq wave 1Vpp. I did not see any difference in the jitter.

BTW 500nS should be 500pS 

[Orange]

I tested with Sin Wave 100mV RMS, and with Sq wave 1Vpp. I did not see any difference in the jitter.

was the Sq.wave above and below 0.0 Vdc ,
What version of FW??

Symetrically at 0Volts, so -500..+500mv
FW is 3.00.01.03

[marmad]

Note that there appears to be about a  9ns offset between the "EDGE" External Trigger point and the CH1 trace.

The offset I get with a 0V External Trigger level is ~5-6ns (see image #1 - it varies a bit based on input voltage level). But since the External Trigger is a different signal path from the channel inputs, I would expect some difference. Of course, the offset changes based on the trigger level set: note the 3 images with a 0, 500mv, and 750mV trigger level. At 500mv (with 2V pk-to-pk square wave input simultaneously to CH1 and EXT) there is zero offset.

It appears the optimal setting for an External Trigger (assuming you want as little offset as possible) is between ~1/4 to 2/3 (with 1/2 between the average) of the respective-going voltage (i.e. for falling/negative edges/pulses ~half of the negative voltage - for rising/positive edges/pulses ~half of the positive voltage).

I can confirm the delay of 9nS, also I see jitter of 500nS if I trigger externally.

Symetrically at 0Volts, so -500..+500mv
FW is 3.00.01.03

+500mV/-500mV square wave - virtually zero jitter (and no offset with trigger level set to 250mV). Check animated GIF of captured segments.

[Orange]

Note that there appears to be about a  9ns offset between the "EDGE" External Trigger point and the CH1 trace.

The offset I get with a 0V External Trigger level is ~5-6ns (see image #1 - it varies a bit based on input voltage level). But since the External Trigger is a different signal path from the channel inputs, I would expect some difference. Of course, the offset changes based on the trigger level set: note the 3 images with a 0, 500mv, and 750mV trigger level. At 500mv (with 2V pk-to-pk square wave input simultaneously to CH1 and EXT) there is zero offset.

It appears the optimal setting for an External Trigger (assuming you want as little offset as possible) is between ~1/4 to 2/3 (with 1/2 between the average) of the respective-going voltage (i.e. for falling/negative edges/pulses ~half of the negative voltage - for rising/positive edges/pulses ~half of the positive voltage).

I can confirm the delay of 9nS, also I see jitter of 500nS if I trigger externally.

Symetrically at 0Volts, so -500..+500mv
FW is 3.00.01.03

+500mV/-500mV square wave - virtually zero jitter (and no offset with trigger level set to 250mV). Check animated GIF of captured segments.

Nice animated JPEG ! almost a real scope on your desktop
However I can see horizontal jitter, if you look closely, you can spot it. It is even better viewable with a fast risetime signal, and on 2nS/div....
We are talking about 500pSec. This dissapears if you change the trigger from ext to ch1 (at least on mine)

[marmad]

Nice animated JPEG ! almost a real scope on your desktop
However I can see horizontal jitter, if you look closely, you can spot it. It is even better viewable with a fast risetime signal, and on 2nS/div....
We are talking about 500pSec. This dissapears if you change the trigger from ext to ch1 (at least on mine)

I just realized that you changed 500ns jitter to 500ps in a later post - a slight difference - so, yes, there seems to be a jitter of ±500ps 250ps - i.e. one sample period.

EDIT: I think the jitter is ±250ps or 500ps in total - see attached GIF.

[jlmoon]

Just a matter of curiosity:

Has anyone determined the better of the two latest firmwares for the DS2000 series yet?  FW v2.xxxx or v3.xxxxx?
I just bought a DS2202 non A version and have the desire to unlock all the options but want to make sure I have the better (less buggy) of the firmwares in place first.
I have the HW level 1 version so I presume I will be stuck with 1M inputs which is no big deal. 

Thx

JLM

[marmad]

Has anyone determined the better of the two latest firmwares for the DS2000 series yet?  FW v2.xxxx or v3.xxxxx?
I just bought a DS2202 non A version and have the desire to unlock all the options but want to make sure I have the better (less buggy) of the firmwares in place first.

Well the v.3 FW is only a couple of weeks old, but I haven't discovered any major new bugs in it yet (and they fixed at least one from v.2). Since it adds new (IMO important) features, I'd recommend it.

[jlmoon]

marmad,

Thank you, I took the plunge and performed the v3 update.. guess we shall see.. .. Thank you. 
so far, so good! 
what a great piece of equipment for the money, very impressive!

JLM

[FrankenPC]

marmad,

Thank you, I took the plunge and performed the v3 update.. guess we shall see.. .. Thank you. 
so far, so good! 
what a great piece of equipment for the money, very impressive!

JLM

If Rigol made a DS2074 version of this scope, it would be perfection.

[jlmoon]

If Rigol made a DS2074 version of this scope, it would be perfection.

Even a higher level of perfection would be a DS2304 with room to grow!..

[Wim13]

If Rigol made a DS2074 version of this scope, it would be perfection.

Indeed i have, thats why i was looking at the ds1000, connected to a ds2000..!!, 6 channel DSO with lots of different timebases.

now i can compare two rubidiums clocks, a gps, a BBC4 and an other reference at the same time,
with different timebases for the rubidiums etc...

Fore more detail i look at the DS2000, for others i look at the DS1000

Works very well.

[Ivan7enych]

Even a higher level of perfection would be a DS2304 with room to grow!..

A few members here have 2 Rigols and connect trig out/in.
Maybe it's called the 'Combo DS1/2306'

I don't think it's a good idea.
As I wrote here, Trig Out signal from DS2072 has very bad jitter.
https://www.eevblog.com/forum/profile/?u=88871

[marmad]

I don't think it's a good idea.
As I wrote here, Trig Out signal from DS2072 has very bad jitter.

We had an in-depth discussion about the Trigger Out delay and jitter in this very thread over a year ago. But these things can be compensated for when using the Trigger Out.

BTW, in response to something you wrote 2 months ago:

It seems, that Trig Out of 2072a is not hardware, but a kind of software (from FPGA logic) and it is suitable for low frequencies only, not higher than 1MHz I think.

The Trigger Out reflects the actual waveform capture rate of the DSO; it doesn't pass through triggers it doesn't use. As such, it's frequency will NEVER be higher than ~52kHz maximum. 

Indeed i have, thats why i was looking at the ds1000, connected to a ds2000..!!, 6 channel DSO with lots of different timebases.

@Wim13 - Which reminds me - what is the situation with the Trigger Out on the DS1000Z series? Less delay and/or jitter?

[marmad]

BTW, now that the External Trigger input can be routed to the frequency counter - it makes it very easy to see the waveforms per second rate without using an external counter - or affecting the rate by turning on CH2.

Here is the Trigger Out (from a 1MHz square wave on CH1) being fed back into External Trigger In (and the frequency counter):

[pa3bca]

@Wim13 - Which reminds me - what is the situation with the Trigger Out on the DS1000Z series? Less delay and/or jitter?

Well I did some tests with a DS1072z trigger out to ext trigger in of the DS2072.
Input is a 2 MHz 10% pulse from a DG1032

From the screen of the DS2072 I measured a trigger out delay of abt 330 ns

You can see that there is jitter: about 8 ns. I experimented with several timebases, but these figures look to be constant.

If you want both scopes together to act as a 6 channnel scope the delay can be corrected for, but the jitter means that for frequencies > say 1 MHz the jitter becomes annoying....

[marmad]

If you want both scopes together to act as a 6 channnel scope the delay can be corrected for, but the jitter means that for frequencies > say 1 MHz the jitter becomes annoying....

Thanks for the tests!

But I don't think the annoyance of the jitter would be related to the frequency of the input signal (or DUT). As mentioned before, the Trig.Out frequency is always < ~50kHz on the DS2000 - and < ~30kHz from the DS1000Z). Instead, I would think it's related to the time base setting of the DSO which is being triggered from the Trig.Out - in other words, if the time base of that DSO is set to 100ns/div or larger, the 8ns of jitter is < 1/10 a division (image #1), or scarcely noticeable - but if the time base is much smaller (e.g. 5ns/div), the jitter is a much larger percentage of the display (image #2).

Of course, the most accurate way to trigger two DSOs - if possible - is to send triggering signals from the DUT to both DSOs simultaneously - although that requires more setup.