Rigol DG4062 or Siglent SDG2042X

[MoriDove]

I decided to buy a signal generator, tell me which of these generators to choose Siglent SDG2042X and Rigol DG4062
Siglent SDG2042X Sample Rate 1.2 GSa / s, Vertical Resolution 16-bit, Waveform Length 8 pts to 8 Mpts
Rigol DG4062 Sample Rate 500MSa / s, Vertical Resolution 14bits, Waveform Length 16 kpts
It seems that in all respects Siglent SDG2042X is superior to Rigol DG4062 except for the maximum frequency
But at the same time, the older model Siglent SDG5000 series is also cut down there as in Rigol, why is that?
In terms of quality and functionality, which one is better with a priority for the future?

[RoGeorge]

I don't know which one is better, but I know the Rigol DG4062 can be unlocked to generate up to 200 MHz sinus (turned into a DG4202 by software, apparently all DG4000 models have the same hardware).

I remember some Siglent were using interpolation to increase the apparent number of bits and apparent sample rate, so a direct comparison of the numbers is not fair.  I have a Rigol DG4102 and it's OK-ish, probably the operating interface could have been better.  No idea if Siglent has a better or a worst one.

DG4000 also has sync outputs for each channel (4 BNCs on the front panel), very useful, has a separate counter/frequency meter on the back, also very useful for precise frequency measurements.

[tautech]

The isolated channel SDG5000 has been discontinued unfortunately and the top model was 160 MHz.
The only isolated AWG Siglent how do is the single channel 25 MHz SAG1021I module that has to paired with one their DSO's from which it's controlled from.

SDG2042X can be improved to 120 MHz if you're into that sort of thing.

[nctnico]

to the OP:
Based on jitter the SDG2042X is the better generator but still not stellar. If you are going look at edges (with an oscilloscope) of none-sinusoidal signals the Rigol likely is going to show visual smearing due to the 500ps jitter. The Siglent has 4 times less jitter at 125ps. The SDG2042X also has way longer AWG memory. What is the budget?

[MoriDove]

I don't know which one is better, but I know the Rigol DG4062 can be unlocked to generate up to 200 MHz sinus (turned into a DG4202 by software, apparently all DG4000 models have the same hardware).

I remember some Siglent were using interpolation to increase the apparent number of bits and apparent sample rate, so a direct comparison of the numbers is not fair.  I have a Rigol DG4102 and it's OK-ish, probably the operating interface could have been better.  No idea if Siglent has a better or a worst one.

DG4000 also has sync outputs for each channel (4 BNCs on the front panel), very useful, has a separate counter/frequency meter on the back, also very useful for precise frequency measurements.

It says here after unlocking there are distortions that cannot be calibrated
Do you have such problems with your oscilloscope ?
https://www.eevblog.com/forum/testgear/dg4000-a-firmware-investigation/msg3297160/#msg3297160

to the OP:
Based on jitter the SDG2042X is the better generator but still not stellar. If you are going look at edges (with an oscilloscope) of none-sinusoidal signals the Rigol likely is going to show visual smearing due to the 500ps jitter. The Siglent has 4 times less jitter at 125ps. The SDG2042X also has way longer AWG memory. What is the budget?

budget $1000

Rigol likely is going to show visual smearing due to the 500ps jitter

It is very important ?
As stated above, some Siglents have used interpolation to increase the apparent number of bits and the apparent sampling rate.
Does SDG2042X real numbers ?

[nctnico]

Whether jitter is important depends on what kind of tests you need to do. In general more advanced waveform generators use a technique which enhances the placement of samples and make sure that as much points as possible are used from from a waveform. This translates into lower jitter and a better signal at the output. The SDG2042X uses these techniques as well. So from a signal production point of view the SDG2042X is better compared to the DG4062.

$1000 is a decent budget however I wouldn't know a better all-round generator in this price class than the SDG2042X. GW Instek has the MFG-2000 series with floating outputs (although the outputs have a common ground) and there is the TTI TGF4000 series. I tried both GW Instek MFG-2000 and TTI TGF4000 series and wasn't impressed (*) but it could be worth looking at for completeness.

* Both these generators failed the test to see if the outputs stay in sync with a large difference in frequency between the outputs. For example 1Hz and 10MHz; both showed a frequency offset (less than 1Hz but still).

[MoriDove]

nctnico

SDG2042X What AWG Memory By Numbers?
Rigol DG4062 What AWG Memory By Numbers?
And if we compare the functionality, which of the given generators has more and better functions?
I need a generator for the future where I don't know what I need

[RoGeorge]

It says here after unlocking there are distortions that cannot be calibrated
Do you have such problems with your oscilloscope ?
https://www.eevblog.com/forum/testgear/dg4000-a-firmware-investigation/msg3297160/#msg3297160

No.

[TurboTom]

I've got both a Siglent SDG6000X and a Rigol DG4000. I haven't got an SDG2000X but usability-wise and regarding the interfaces, the SDG6000X should be very similar if not identical. I only mention the "family" designation of each model since the hardware throughout the individual family is exactly the same and there are ways to morph models into each other without invasive (hardware) methods. You will find the corresponding threads if you search for them.

Both families can be considered to be "mature", so the bugs should be ironed out by now, which is mostly true. Simply due to the physical size of the user interface, usability differs a lot. I find Rigol's model much easier and faster to use and to set up, with it's direct buttons for the most frequently used standard waveforms and the controls grouped only to the right of the screen. I don't miss a touch screen on this AWG at all. In general, the U/I is much less cramped, simply since there is more than enough space on the front due to the differnet form factor. Accessing the built-in arbitraries is a real PITA since Rigol still didn't add grapical icons of the waveforms in the selection menu. Shame! The screen size and appearance (brightness, contrast) is superb.

Siglent's SDG1000X/2000X/6000X series somewhat limited by its small frontal area which requires the use of very small buttons and grouping them to the right and below the rather small(ish) touchscreen. The screen itself is mediocre at best, somewhat affected by the resistive touch panel placed in front. The touch functions mostly duplicate functions that are available as hard- or softkeys anyway. It's quite sad to see Siglent didn't use the opportunity to integrate the touch capabilities into a more modern U/I concept. The touch capabilities rather appear as a half-hearted attempt to "pimp" an already existing, non-touch U/I.

If you need some connectivity (individual -per channel- modulation/trigger inputs, sync outputs and a really capable counter with full statistics and graphical trend display), the DG4000 wins hands down. For its two channels, there are a total of four additional BNC interfaces to "talk" to other equipment while while the SDG?000X series of instruments offers only a single one. The integrated frequency counter is a joke, even basic information doesn't get displayed while in the generator screens, probably due to the limited screen size. IMO, it would have been much more reasonable to do without the counter completely and use its BNC connector to add a second AUX interface for the generator channels. Anyway, I don't know anyone who looks at the frequency counter functionality when he's in the market for a mid-range AWG.

As far as I could find out, the sample width  (14 vs. 16 bits) doesn't really affect signal fidelity of the built-in/standard waveforms. Compared to the SDG6000X, the DG4000 produces an ever-so-slightly cleaner signal (sine, harmonics-wise). But the differences are marginal. What may actually make a difference is that the output level that the generators are able to supply at higher frequencies drops on the SDG2000X less quickly than on the DG4000. On the other hand, the DG4000 can output a maximum of 200MHz while the SDG2000X series maxes out at 120MHz. Depending on the application, this means the DG4000 can be used as a makeshift RF signal generator for repairing/testing VHF aircraft radios or for applications in the 2m HAM band while the SDG2000X cannot.

There are other thoughts that may affect the decision which instrument is better suited, i.e. simply form a physical size point of view (small frontal area but higher depth or big frontal area but shallow). If there is already a Siglent oscilloscope present that supports the Bode plot function, to accompany the AWG, the the choice of a Siglent AWG is virtually obvious. On the other hand, Rigol's AWGs can directly copy a waveform form a Rigol scope to their arbitrary buffer to duplicate it (whoever needs that).

I rarely use "my own" arbitraries so there's little I can tell about the memory limit on the two instruments -- I never had a problem regarding this with any of the mentioned instruments.

In the end, the choice is broken down to the individual requirements and jobs that the AWG is required for. IMO, as a general lab instrument with good connectivity and easier/faster (to use) U/I, the DG4000 is the better choice, while in certain applications, steeper/more accurate pulse reproduction, higher output levels of moderately high frequencies, or as a source in combination with a Siglent oscilloscope, the SDG2000X is the instrument of choice.

To make the confusion complete: If a signal bendwidth of 100MHz is enough, and the connectivity to a Siglent Scope to enable Bode plot functions isn't required (and the ridiculous outer appearance doesn't put you off... ), I'ld also have a look at Rigol's DG811 and "open it up" to DG992 level. That's currently the best bang-for-buck AWG available, provided the specs meet the requirements.

If you would like me to do specific tests on the DG4000, please let me know...

[H.O]

I've had a DG4000 for 6-7 years or so and one thing that has come to bite me is how it "restarts" the waveform when you change the dutycycle of a squarewave, causing a discontinuity in the output.

I tried to use it to test a H-bridge driven in locked antiphase mode - that didn't go well because when changing the dutycycle from 50% to 49% it effectively goes to 0% for a period of time.

I do not know how the various SDG series generators handles this (or any other AWG for that matter, perhaps it's just "how it works").

Apart from that it has served my needs perfectly.

[TurboTom]

I've had a DG4000 for 6-7 years or so and one thing that has come to bite me is how it "restarts" the waveform when you change the dutycycle of a squarewave, causing a discontinuity in the output.

I tried to use it to test a H-bridge driven in locked antiphase mode - that didn't go well because when changing the dutycycle from 50% to 49% it effectively goes to 0% for a period of time.

I do not know how the various SDG series generators handles this (or any other AWG for that matter, perhaps it's just "how it works").

Apart from that it has served my needs perfectly.

That depends a lot on the firmware version used and the configuration of the channels (phase locked or independent). Right now, I'ld say none of the generators is 100% safe to use in that application. Moreover, generating antiphase signals with a fixed dead time or alternating PWM signals with 180° shift and single input pulse width adjustment doesn't appear to be possible with any of these. If any type of coupling is enabled between the two channels, seamless parameter changes aren't possible anymore. So testing of only slightly more complex PWM stages is better done with a pulse generator specifically taylored for this application. Even Rigol's DG800/800/2000 series, despite they advertised it for providing seamless parameter changes, fails miserably under these conditions. I guess, there's still some way to go for the manufacturers ...

[MoriDove]

I've got both a Siglent SDG6000X and a Rigol DG4000. I haven't got an SDG2000X but usability-wise and regarding the interfaces, the SDG6000X should be very similar if not identical. I only mention the "family" designation of each model since the hardware throughout the individual family is exactly the same and there are ways to morph models into each other without invasive (hardware) methods. You will find the corresponding threads if you search for them.

Both families can be considered to be "mature", so the bugs should be ironed out by now, which is mostly true. Simply due to the physical size of the user interface, usability differs a lot. I find Rigol's model much easier and faster to use and to set up, with it's direct buttons for the most frequently used standard waveforms and the controls grouped only to the right of the screen. I don't miss a touch screen on this AWG at all. In general, the U/I is much less cramped, simply since there is more than enough space on the front due to the differnet form factor. Accessing the built-in arbitraries is a real PITA since Rigol still didn't add grapical icons of the waveforms in the selection menu. Shame! The screen size and appearance (brightness, contrast) is superb.

Siglent's SDG1000X/2000X/6000X series somewhat limited by its small frontal area which requires the use of very small buttons and grouping them to the right and below the rather small(ish) touchscreen. The screen itself is mediocre at best, somewhat affected by the resistive touch panel placed in front. The touch functions mostly duplicate functions that are available as hard- or softkeys anyway. It's quite sad to see Siglent didn't use the opportunity to integrate the touch capabilities into a more modern U/I concept. The touch capabilities rather appear as a half-hearted attempt to "pimp" an already existing, non-touch U/I.

If you need some connectivity (individual -per channel- modulation/trigger inputs, sync outputs and a really capable counter with full statistics and graphical trend display), the DG4000 wins hands down. For its two channels, there are a total of four additional BNC interfaces to "talk" to other equipment while while the SDG?000X series of instruments offers only a single one. The integrated frequency counter is a joke, even basic information doesn't get displayed while in the generator screens, probably due to the limited screen size. IMO, it would have been much more reasonable to do without the counter completely and use its BNC connector to add a second AUX interface for the generator channels. Anyway, I don't know anyone who looks at the frequency counter functionality when he's in the market for a mid-range AWG.

As far as I could find out, the sample width  (14 vs. 16 bits) doesn't really affect signal fidelity of the built-in/standard waveforms. Compared to the SDG6000X, the DG4000 produces an ever-so-slightly cleaner signal (sine, harmonics-wise). But the differences are marginal. What may actually make a difference is that the output level that the generators are able to supply at higher frequencies drops on the SDG2000X less quickly than on the DG4000. On the other hand, the DG4000 can output a maximum of 200MHz while the SDG2000X series maxes out at 120MHz. Depending on the application, this means the DG4000 can be used as a makeshift RF signal generator for repairing/testing VHF aircraft radios or for applications in the 2m HAM band while the SDG2000X cannot.

There are other thoughts that may affect the decision which instrument is better suited, i.e. simply form a physical size point of view (small frontal area but higher depth or big frontal area but shallow). If there is already a Siglent oscilloscope present that supports the Bode plot function, to accompany the AWG, the the choice of a Siglent AWG is virtually obvious. On the other hand, Rigol's AWGs can directly copy a waveform form a Rigol scope to their arbitrary buffer to duplicate it (whoever needs that).

I rarely use "my own" arbitraries so there's little I can tell about the memory limit on the two instruments -- I never had a problem regarding this with any of the mentioned instruments.

In the end, the choice is broken down to the individual requirements and jobs that the AWG is required for. IMO, as a general lab instrument with good connectivity and easier/faster (to use) U/I, the DG4000 is the better choice, while in certain applications, steeper/more accurate pulse reproduction, higher output levels of moderately high frequencies, or as a source in combination with a Siglent oscilloscope, the SDG2000X is the instrument of choice.

To make the confusion complete: If a signal bendwidth of 100MHz is enough, and the connectivity to a Siglent Scope to enable Bode plot functions isn't required (and the ridiculous outer appearance doesn't put you off... ), I'ld also have a look at Rigol's DG811 and "open it up" to DG992 level. That's currently the best bang-for-buck AWG available, provided the specs meet the requirements.

If you would like me to do specific tests on the DG4000, please let me know...

Thank ! your feedback helped me a lot !
I have a Rigol MSO5074 oscilloscope upgraded to Rigol MSO5354
I think I need to buy Rigol DG4000

Tell me, how are things with Hantek HDG6202B ?