Author Topic: Siglent SDG6000X series 200-500 MHz AWG's  (Read 91550 times)

0 Members and 1 Guest are viewing this topic.

Offline gperoni

  • Contributor
  • Posts: 38
  • Country: it
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #75 on: August 14, 2018, 09:29:55 am »
Ok so, first off, a HUGE thank you to janekivi, as only thanks to his file I got root access to the device.

@tv84: I tried reading the SSA thread and pretty much all the methods, the only one that seems to work for me (upgrading from 200 to 500 MHz) is removing (moving to a different name) the /usr/bin/siglent/NSP_system_info.xml file. Touching the NSP_trends_config_info file results in me losing the PRBS/IQ entry in the Waveform menu. I tried playing around with the ./config directory too, as some files there are the same or similar, but I couldn't get the IQ option to work (the 500 MHz hack works fine though!).

Can you write a line on what you did?

Thank you :)
 

Offline tv84

  • Super Contributor
  • ***
  • Posts: 3211
  • Country: pt
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #76 on: August 15, 2018, 02:41:51 pm »
Can you write a line on what you did?

There are several alternatives... The most accessible for everyone: search in the memdump.
 

Online tautechTopic starter

  • Super Contributor
  • ***
  • Posts: 28058
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #77 on: August 16, 2018, 07:52:30 pm »
well, screen is about twice as big..
in the lecroy we demoed the touchscreen was so bad that you had to press too hard to use it and when you didn't move your finger the cursor would move frenetically. with a stylus, not so bad but at that point i'd rather use a mouse thank you very much :(
@ JPortici
I've not long had the new SVA1015X SA/VNA unit, it has a 10" touch screen as well as the normal UI so to address any reservations you might have on the latest Siglent touch screens I can assure you this new SVA screen is a joy to drive. There's also some shortcut menu icons and they're very nice to use.
This SVA is the first big touch display we've seen from Siglent other than the WS3000 that LeCroy market and Siglent engineers look to have done a nice job implementing it.
SVA also offers mouse and keyboard control......I've only had a little play with it and only with a wireless KB/mouse combo which of course is not ideal for SA usage.
Hope that addresses any nagging doubts you had.  :)
« Last Edit: August 17, 2018, 01:08:50 am by tautech »
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #78 on: September 13, 2018, 06:14:38 pm »
Today I received an SDG6022X and I've got to admit I'm already disappointed after the first ten minutes of testing. I believed (I know, that's generally a bad thing to do when purchasing Chinese test gear...) that I could use it to replace my Rigol DG4102 but as it seems, it's got EXACTLY the same shortcomings as the latter! You may find my rants about the Rigol generator here: https://www.eevblog.com/forum/testgear/annoying-bug-in-rigol-dg4000-series-function-generator/

I found the SDG6000x to generate runt pulses if "Waveform: Pulse" is selected and the pulse width is adjusted with the encoder. That is a job that's not too special to tolerate this behavior, shown in the attached screenshots.

The user interface is partially okay and partially still a nightmare, especially considering the availability of the touchscreen. Selecting built-in arbitrary waveforms is virtually the same catastrophe as with the Rigol. Siglent, don't copy other's crappy solutions, rather have your engineers use their own brains! A list of some 200 ARB waveforms without any way to quickly scroll through them with at least a small indication of their shape is mostly useless. At least in my opinion.

The incremental encoder on this machine also skips a step now and then although it's by far not as bad as on the DG4000. Anyway, to reliably do incremental adjustments of pulse widths or frequencies or whatyoulike without looking at the screen, just by sensing the detents of the encoder, it simply isn't good enough.

I didn't do much more testing so far, rise time and general signal quality looks decent at the first glance. But considering the price of the instrument, I've got to admit that I expected more than that, especially usability-wise. Or am I asking too much?

Probably I'll have a second, more thorough look at the instrument during the next days but if my first impressions prevail, this thing will get returned...
« Last Edit: September 14, 2018, 08:40:25 am by TurboTom »
 

Online tautechTopic starter

  • Super Contributor
  • ***
  • Posts: 28058
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #79 on: September 13, 2018, 07:24:30 pm »
Tom, can I ask if this unit has the latest firmware installed, Version: 6.01.01.29R10 ?
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #80 on: September 13, 2018, 08:44:07 pm »
Yes it came from the distributor with the mentioned firmware installed.
 

Online tautechTopic starter

  • Super Contributor
  • ***
  • Posts: 28058
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #81 on: September 13, 2018, 08:53:38 pm »
Yes it came from the distributor with the mentioned firmware installed.
Thanks for confirmation. I'll pass your findings to the factory.
Probably I'll have a second, more thorough look at the instrument during the next days but if my first impressions prevail, this thing will get returned...
Yes do that and study what it can do and a liberated one can.  ;)
If you want to venture that path these units are good value.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #82 on: September 14, 2018, 09:28:17 am »
Did some more testing already and my impression of the instrument doesn't get better. The signal appears to get interrupted/distorted upon any timing-related change of the settings via the incremental encoder, regardless of the waveform. I attached three screenshots, the fist changing the frequency of a 5kHz sine by one least significant digit, the second the same thing for a 50% duty cycle square (now also a runt pulse is generated), and finally I change the duty cycle by 1% of the same square wave, causing the signal to pause for more than 2ms. If you use the generator to test a switch mode power stage, it easily goes "poof"  :-BROKE if this happens with the drive signal.

Edit: Just added a screenshot of what a frequency change looks like on the Rigol DG4000 (same for ramp; sqare & pulse similar or even a little worse than on the SDG6000X).

Edit 2: I figured out that if the -> Utility -> Phase Mode setting is changed from the default "Phase Locked" to "Independent", the Sine (and probably also Ramp) signal permit change of frequencies without the mentioned discontinuities. So in this configuration, the SDG6000X performs just as the Rigol DG4000. Square and pulse still broken, unfortunately.

The user interface will also easily ruin your day since it doesn't store the digit of the parameter that was adjusted, i.e. if you do fine adjustments of the frequency (say you're changing the 100Hz digit of a 100kHz output signal) via the encoder and then do an amplitude correction, to return to the frequency adjustment afterwards, you will be adjusting the 100kHz digit. This kind of behavior is very unproductive since especially in case of a signal generator, you're usually not looking at the generator's screen but at the screen of an oscilloscope or other test gear to observe the results of changes of the input parameters to your DUT. It must be possible to reliably change parameters incrementally without having to look at the generator all the time only to check that it's performing correctly.

The SDG6000X may have many bells and whistles but still it has to operate as a plain and simple "old" versatile fequency generator in my opinion. If it doesn't work okay in all these modes, I consider it not up to the task and not worth the money. Even the Rigol DG4000 performs better in many of these scenarios, and the best generator I own concerning this, is my cheapest one (though it's -- been -- plagued with other problems), the Hantek HDG2002B.

So far I think Siglent should focus on the details that make an instrument comfortable and enjoyable to use in real lab scenarios, otherwise (and here I remember one statement from some other thread about LCR meters, refering to some TFT-equipped poo ...) the instrument is not much more than some touchscreen-featured poo.

Even though all these issues can probably be adressed by firmware / FPGA code updates, I doubt that Siglent will take care of them all. I haven't got any experience with the SDG2000X generator but I assume that the SDG6000X's code bases heavily on the aforementioned one so I'm almost certain that these shortcomings can be found there as well (maybe someone who owns an SDG2000X could test it?). If this is the case and Siglent didn't adress these problems yet, I doubt they will do so on the SDG6000X either. I will test the generator probably for a week or two and then return it. It's not worth the money if you're not buying it specifically for its I/Q functionality (costs more money, I don't want to talk about the "liberation of options" here and now), the baseline model is just not adequate as an allround lab generator that's confortable to use. In it's current condition, I'ld rate it two stars out of five for its signal quality and fast slopes. It's too expensive for its usability value.
« Last Edit: September 14, 2018, 12:45:35 pm by TurboTom »
 
The following users thanked this post: tautech

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 922
  • Country: ee
    • lab!fyi
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #83 on: September 14, 2018, 12:41:08 pm »
I haven't got any experience with the SDG2000X generator but I assume that the SDG6000X's code bases heavily on the aforementioned one so I'm almost certain that these shortcomings can be found there as well (maybe someone who owns an SDG2000X could test it?).

No switching issues with SDG2000X. Still very pleased with my 4 channel 120MHz phase locked combo for <1000€ ;)

« Last Edit: September 14, 2018, 12:43:16 pm by MrW0lf »
 
The following users thanked this post: TurboTom

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #84 on: September 14, 2018, 12:53:52 pm »
Thanks for testing this @MrWolf -

so there may be hope that Siglent will sort this problem on the SDG6000X eventually...
If this will happen in acceptable time for me to risk to keep the instrument is another question, though  :-\
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 922
  • Country: ee
    • lab!fyi
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #85 on: September 14, 2018, 01:28:48 pm »
Edit 2: I figured out that if the -> Utility -> Phase Mode setting is changed from the default "Phase Locked" to "Independent", the Sine (and probably also Ramp) signal permit change of frequencies without the mentioned discontinuities. So in this configuration, the SDG6000X performs just as the Rigol DG4000. Square and pulse still broken, unfortunately.

Oh no! :scared:
Phase Locked it was, but switched CH Copy Coupling=OFF (before both channels were locked in all but phase deviation). 5kHz square on one channel, set second channel to 20MHz. Now look :'(
Glitches on both channels but <50us, not ~2ms like in your case.



However when I set to Independent, duty transition is again seamless. Now if think about it may be indeed somewhat problematic to do seamless switch when both channels phase locked, and run different frequencies. But it might work a little better... Think would be helpful if mr tautech pokes HQ a bit on this...
« Last Edit: September 14, 2018, 01:35:15 pm by MrW0lf »
 
The following users thanked this post: TurboTom

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #86 on: September 14, 2018, 02:33:04 pm »
In my tests, I was't looking at different channels. It's just a single one that pauses for a certain duration (probably while the FPGA is reconfigured) and then resumes to output the signal with the new parameters. No phase lock, no coupling, just a single channel. And in my opinion it must be possible to get a seamless signal changes in square and pulse mode (if minimum slope times are configured so no slope scaling is required).

I did some more tests regarding time base accuracy and this looks quite good. I find on "my" specimen the time base to be approx. 0.6ppm slow, i.e. if I enter a frequency of 10.0000059MHz to be generated, I get a slow drift on the scope agains my 10MHz Rb source as shown on the screenshot. It's funny that I just must have recorded an adjustment step of the TCXO that's used as the reference inside the SDG6000X.

DC amplitude accuracy looks good, deviating next to nothing below 3V and a maximum of 0.5% at 10V. AC level accuracy still needs to be tested.

Edit: I just checked level accuracy of the Sine wave from 100kHz to 200MHz (instrument isn't "liberated") with my venerable HP 436A + 8482A and found it (converted to Veff) to be better than -0 / +0.8% absolute on both channels over the full range (-0 / +0.07dB). So it's well within spec.
« Last Edit: September 15, 2018, 03:08:29 pm by TurboTom »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 922
  • Country: ee
    • lab!fyi
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #87 on: September 14, 2018, 03:46:00 pm »
Can you please check how 5kHz 50=>51% duty transition looks on DG4000 with second channel phase locked and set to max square freq (or 20MHz like in my test)? This could prove a good "torture test" for gen. Solving this situation elegantly would require certain effort.
 

Online tautechTopic starter

  • Super Contributor
  • ***
  • Posts: 28058
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #88 on: September 14, 2018, 08:45:12 pm »
Think would be helpful if mr tautech pokes HQ a bit on this...
Done yesterday but it's the weekend here and at HQ now.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #89 on: September 14, 2018, 08:49:51 pm »
@MrWolf

Okay, here we are. Frequencies were 5kHz and 40MHz (CH1 / CH2), phase coupling between the two with CH1 as "master" as shown in the first screenshot. The DG4000 pauses a long time, more than 6ms and also produces some glitch in between as shown in the last screenshot. It performs clearly worse than the SDG6000X. But in my opinion the difference is marginal as long as there's a pause at all. This renders the generator useless for many critical applications.

One clear advantage of the DG4000 is its frequency counter which is way more advanced than those included with any other AWG, featuring full statistics and graphics display. But that's not really what I get an AWG for even though it may be a nice "goodie" if everything else was working flawlessly...
 
The following users thanked this post: MrW0lf

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #90 on: September 15, 2018, 10:13:00 pm »
I guess I have missed the opportunity to download the file "SDG6000X_eevblog_29R10.zip" linked to by @janekivi about a month ago. Now the wikisend link has expired and I would like to gain root access to my SDG6000X. Any chance someone could upload it again or provide it attached to a PN? You would do me a great favor and I'll be eternally thankful  ;).

Cheers,
Thomas
« Last Edit: September 16, 2018, 07:29:54 am by TurboTom »
 

Offline tv84

  • Super Contributor
  • ***
  • Posts: 3211
  • Country: pt
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #91 on: September 16, 2018, 11:39:10 am »
I guess I have missed the opportunity to download the file "SDG6000X_eevblog_29R10.zip" linked to by @janekivi about a month ago. Now the wikisend link has expired and I would like to gain root access to my SDG6000X. Any chance someone could upload it again or provide it attached to a PN? You would do me a great favor and I'll be eternally thankful  ;).

Does that mean you are convinced by the SDG6000X?   ::) 
 
The following users thanked this post: TurboTom

Offline janekivi

  • Frequent Contributor
  • **
  • Posts: 368
  • Country: ee
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #92 on: September 16, 2018, 11:49:09 am »
 
The following users thanked this post: egonotto, TurboTom, 2N3055, kerouanton, willie.from.texas, vt100

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6407
  • Country: hr
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #93 on: September 16, 2018, 07:14:01 pm »
I did a quick test on a DG1032Z.

1. Sinewave is glitch free
2. Square and pulse wave is glitch free up until 1MHz, for both frequency and duty cycle. Over 1MHz they start glitching.
3. Ramp is glitching all the time
4. AWG seems to be glitch free

Glitches are represented with abrupt cutoff of generated waveform at whatever random moment, creating really short runt pulse, followed by 3-4 ms zero level without signal generated..

Didn't try chanel phase coupling..
 
The following users thanked this post: TurboTom

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #94 on: September 16, 2018, 08:11:48 pm »
I guess I have missed the opportunity to download the file "SDG6000X_eevblog_29R10.zip" linked to by @janekivi about a month ago. Now the wikisend link has expired and I would like to gain root access to my SDG6000X. Any chance someone could upload it again or provide it attached to a PN? You would do me a great favor and I'll be eternally thankful  ;).

Does that mean you are convinced by the SDG6000X?   ::)

No, actually I'm looking for a way to convince myself to keep it, bad case of TEA, you know what I mean  ;)

Thanks a lot to all those who helped me with getting root access to the device by PM or on the forum.

As it seems, Siglent intended to also provide bandwidth upgrades but they may have shelved that idea. So far, I'm still ambiguous regarding the performance of the SDG6000X. Signal quality, accuracy and options are really good but the shortcomings that I mentioned are showstoppers for me and the applications that I'ld like to use the instrument for. Fortunately, I've still got a few weeks time to make a decision.

Cheers,
Thomas
 

Offline willie.from.texas

  • Newbie
  • Posts: 2
  • Country: us
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #95 on: September 17, 2018, 11:46:17 pm »
I upgraded both my SDG6022X function generator and my SDS1104X-E four channel oscilloscope. Here is an indication of my performance improvement from both updates.  The chart below was generated using a 1.5Vpp sine wave driving a 50 ohm load through a 12" coax as measured by the oscilloscope...
« Last Edit: September 17, 2018, 11:49:08 pm by willie.from.texas »
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #96 on: September 18, 2018, 06:50:19 am »
I've got some new findings with the SDG6022X that appear to indicate that there's actually a hardware difference between this and the higher rated models of the SDG6000X series that preclude hacks to higher frequencies. As yet, I'm not 100% sure since the behaviour may also point into the direction of a hardware problem of the particular unit that I got. Anyway, currently my recommendation to all those who are playing with the thought to purchase such an instrument with the perspective of a bandwidth upgrade, don't take it for granted that it will work.

I've got to review some screenshots and edit some recordings before I can post a more detailed report of my findings. This may happen later today or tomorrow...
 

Online markus_jlrb

  • Regular Contributor
  • *
  • Posts: 137
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #97 on: September 18, 2018, 02:12:21 pm »
Dears,

FYI

just to show you my experience with the SDG6022X device.
Was playing with FSK sigs.
The attached pics, show a 2MHz/10MHz carrier switched with
1MHz frequency.
To get a proper 0V change in frequency some adjustments
with phase angle parameter were necessary due to coax
delay to the dso.
Extern triggering with switching frequency was done on a
SDS2000X.

So for my purpose the dev is quite suitable.

Markus
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1380
  • Country: de
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #98 on: September 18, 2018, 09:13:08 pm »
As I mentioned before, I did some more experiments with the SDG6022X and made some peculiar findings. As it seems, the instrument (at least the one that I have got) is either speed graded or defective. In its original configuration, I initially didn't find any unforseen behavior except for the short pauses when changing settings that I mentioned several posts above.

Now that I was able to "liberate" the machine with the help of @janekivi's firmware file  :-+, I did some more testing and initially everything appeared to work normally. After some testing, I noticed more and more strange reaction to parameter changes which finally led to the following two video clips (I apologize for the unsteady camera, I recorded the clip with my cell phone and had to operate the AWG / 'scope with the left hand... but I guess it hilights my points):

http://www.turbinemuseum.de/files/SDG6000X/SDG6000X_lib1.mp4

http://www.turbinemuseum.de/files/SDG6000X/SDG6000X_lib2.mp4

Please keep in mind that these two clips had been taken with the "liberated" machine, even though the software configuration should have matched a stock SDG6052X. I even rolled back the firmware (to the stock one without known root access) without any observable change. I also got the impression that the effect worsened over time. It is definitely not related to operating temperatures of the instrument since I just powered it up before the first take after it had been disconnected from mains over the night.

After that I removed all my patches completely so the instrument is in its original configuration as I initially received it. The following tests didn't show any of the curiosities that I observed before, it reacts flawlessly to keyboard inputs (well, at least not worse than before... ::) ). What I noticed when I observed steep square wave or pulse edges is some minor "undulation" or whatever one may call it, shown in the following clip (sorry, german language since I thought I may use it for my future correspondence with the distributor):

http://www.turbinemuseum.de/files/SDG6000X/SDG6000X_org1.mp4

I'ld be glad if those of you who own or have access to one of these AWGs could try to replicate these tests and report their findings. Proper termination and wiring is required during all of these tests.

One thing that really surprises me is that Siglent must be calibrating all the units up to 500MHz since the amplitude flatness over the full range is superb. With an HP RF power meter, I find the 0dBm level to fall between 0.05 and 0.15 dBm from 100kHz to 500MHz, a reading taken every 10MHz. Interestingly, with an offset of 2V (the HP 8482A power sensor is A/C coupled at the input with a capacitor of approx. 125nF), the level flatness goes pretty much south (relatively speaking) with excursions of 0.9dBm around 330MHz.

This obvious full-range calibration makes me wonder if the particular unit that I got may actually be faulty. But I'll have some difficulty explaining why it's faulty while it performs okay  (-ish) in its specified frequency range. I guess I'm just gonna return it and don't start any arguments at all. Just don't like it...  ;) And maybe go for another one after (if?) Siglent addressed the pauses during parameter changes.

Cheers,
Thomas




 
The following users thanked this post: thm_w

Online tautechTopic starter

  • Super Contributor
  • ***
  • Posts: 28058
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG6000X series 200-500 MHz AWG's
« Reply #99 on: September 18, 2018, 10:22:14 pm »

This obvious full-range calibration makes me wonder if the particular unit that I got may actually be faulty. But I'll have some difficulty explaining why it's faulty while it performs okay  (-ish) in its specified frequency range. I guess I'm just gonna return it and don't start any arguments at all. Just don't like it...  ;) And maybe go for another one after (if?) Siglent addressed the pauses during parameter changes.
Tom can I ask why you would attempt to evaluate a 200 MHz (stock) or 500 MHz (liberated) AWG with a 70 MHz DSO ?  :-//
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf