The Siglent SDG2042X Thread

[ebastler]

I keep wanting to get this machine.... then keep thinking won't use it enough.
Anyone give some clues or encouragement?!   

You can hack it for free, to give you features which you will use even less! 

Does that count as "encouragement"?
The argument worked for me... 

[k8943]

That summarises some of the temptation and most of the fear quite well.

It would be as expensive as anything else on the bench but will I start finding uses in proportion?

[BillB]

It would be as expensive as anything else on the bench but will I start finding uses in proportion?

Like any tool, an AWG has its uses and comes in very handy when you need it.  However, it's unlikely that you'll use it even a fraction of the time of your DMM or scope.

[k8943]

Yes those things - and in the most literal sense the decent Stereo Microscope - are literally the eyes with which one can see.... Was sort of hoping the the AWG could be a part of that in ways that might surprise one that doesn't have one.

[tautech]

Yes those things - and in the most literal sense the decent Stereo Microscope - are literally the eyes with which one can see.... Was sort of hoping the the AWG could be a part of that in ways that might surprise one that doesn't have one.

Does the cheaper SDG1032X interest you then ?
No touch screen, lower sampling rate and less mem depth but otherwise can do everything 2042X can do.
It also can be 'improved' to 1062X model. 

[k8943]

16bit DAC and better navigation seem "must haves" and the price is not off putting. Just try to follow the discipline of being sure am going to use something before actually getting it. We live in a materialist world but do not have to agree with it. In this case am aware that, never having had an AWG have limited awareness of the possibilities so it's a hard one to evaluate.

[ebastler]

16bit DAC and better navigation seem "must haves" and the price is not off putting.

As much as I like the SDG2000 series (I did buy one!), there are some not-so-obvious limitations:

The stated maximum frequency (40 to 120 MHz) applies to sine waves only. Square wave output is limited to 25 MHz, and with 9 ns rise/fall times does not look very square at that frequency. The SDG1000 series actually has better square wave output, with 4ns rise time and 60 MHz maximum frequency. (But it's worse in pulse mode.) Also, the 1.2 GSa/s sample rate does not apply in Arb mode. It's worthwhile to study the datasheets carefully!

The 16 bit DAC and improved harmonic distortion specs vs. a 14 bit device are "real" and apply in all operating modes, it seems. But I have not yet encountered an application where I need that low distortion. For audio measurements, I use a sound card anyway, with 24 bit DAC and ADC.

[McBryce]

I keep wanting to get this machine.... then keep thinking won't use it enough.

Anyone give some clues or encouragement?!   

Even if you don't do projects that would required an AWG, they can still be quite useful for testing all sorts of equipment and they are also a great tool to get yourself accustomed to a new scope and it's features.

(was that a circular argument endorsing a new scope too??)

McBryce.

[k8943]

There is a bit of circular argument with scopes: my scope maxes at 100Mhz....

I guess one can get quite a bit done by using reference designs as building blocks that in turn eliminate a lot of the fundamental questions.

Interesting point about the specs on the 1000 series... Now thinking about that as well!

[Plasmateur]

Hello. Could anyone point me to someone on these forums who might now how to hack into this on a deeper level.

I don't want to unlock this device. I just want to program it so it can be remotely controlled to trigger both channels at the same time as well as sending out an AUX trigger.

I'm wondering if this is possible.

[pigrew]

Hello. Could anyone point me to someone on these forums who might now how to hack into this on a deeper level.

I don't want to unlock this device. I just want to program it so it can be remotely controlled to trigger both channels at the same time as well as sending out an AUX trigger.

I'm wondering if this is possible.

It's perfectly possible, but would likely require revision of the FPGA logic, and maybe sacrificing one unit in order to reverse-engineer the PCB. It would take quite a lot of time without having the original source files as most/all of the software would need to be rewritten (ASAIK disassembling a FPGA binary is generally not done). I don't have the available time or desire to do this, but could give you some pointers on how to proceed. I may even be able to XRay the analog PCB for you.

[ebastler]

I just want to program it so it can be remotely controlled to trigger both channels at the same time as well as sending out an AUX trigger.

I'm wondering if this is possible.

I'm not sure which aspect you can't achieve via standard setup of the SDG2042. Could you provide a bit more detail?

What would you want to use as the trigger input? There is a single AUX connector, which can either be the trigger input or an output. Or did you mean to trigger via the MTRIG SCPI command, which is equivalent to a manual trigger? If it is the latter, that should work without special trickery, I think (but have not tried).

Which leaves the "trigger both channels at the same time" part. I just had a quick look at my SDG2042X, and it seems possible to e.g. set up both channels in burst mode with a manual trigger, at the same time. Could you elaborate please? What do you want each channel to do, and could not convince the SDG to provide?

Finally, the trigger output you mentioned. Maybe that is where you run into a limitation? You should be able to get a sync output on the AUX jack, but with limitations (chapter 2.13.7 of the user manual). Again, could you please explain which signals you need on the main outputs, and on the AUX output?

[Plasmateur]

It's perfectly possible, but would likely require revision of the FPGA logic, and maybe sacrificing one unit in order to reverse-engineer the PCB. It would take quite a lot of time without having the original source files as most/all of the software would need to be rewritten (ASAIK disassembling a FPGA binary is generally not done). I don't have the available time or desire to do this, but could give you some pointers on how to proceed. I may even be able to XRay the analog PCB for you.

Xray? Seriously? Wow. Well, to be honest I had no idea there would be that big of a learning curve. Is it possible to telenet into this thing and....I don't know if I'm using the right terminology here... image it? I thought this thing ran off of linux. I thought maybe it would be possible to set up a virtual machine on my computer and take a look at the code to see if I could understand it. It appear there is alot more to this than I thought. I don't want to take up too much of your time, but I am a little curious about learning how I might start to attempt this.

I'm not sure which aspect you can't achieve via standard setup of the SDG2042. Could you provide a bit more detail?

What would you want to use as the trigger input? There is a single AUX connector, which can either be the trigger input or an output. Or did you mean to trigger via the MTRIG SCPI command, which is equivalent to a manual trigger? If it is the latter, that should work without special trickery, I think (but have not tried).

Which leaves the "trigger both channels at the same time" part. I just had a quick look at my SDG2042X, and it seems possible to e.g. set up both channels in burst mode with a manual trigger, at the same time. Could you elaborate please? What do you want each channel to do, and could not convince the SDG to provide?

Finally, the trigger output you mentioned. Maybe that is where you run into a limitation? You should be able to get a sync output on the AUX jack, but with limitations (chapter 2.13.7 of the user manual). Again, could you please explain which signals you need on the main outputs, and on the AUX output?

The trigger input is being send via software.

I send it a command using SCPI code via pyvisa. The problem is, you can't trigger both ch.1 and ch.2 simultaneously unless channel coupling is enabled. Well, that's a problem because all you can do with that is vary the phase of each channel. The frequency, mode, delay, amplitude, cycles - all that stays the same. I would like to independently set each one of these, and then send a software trigger that would trigger both ch.1 and ch.2 simultaneously, Aux out would also send its trigger signal as well. - This is shown in the attached picture, but I had to do it using CH Copy Coupling. Only the phase is different.

Now, if you simply trigger the SDG2042X via external AUX, then you can trigger both ch.1 and ch.2 with all the independent features. The problem is, I don't have an expensive two output pulse delay generator with a 10MHz clock snyc at my house, which is what I would need in order to make time of flight measurements for an experiment I would like to try....as cheap as possible.

Edit: Try to trigger both channels at the same time with a manual trigger without using CH Copy Coupling. Not possible. Or attempt to program it. Wish it could unless the new firmware allows this. It didn't make any mention of it in the notes.

[jtruc34]

The 1032X also can be 'improved' to 1062X model. 

Since I own an SDG1032X, I'm curious to know how I could do that. There may be the answer in this thread, but it is very hard to find it through more than a thousand replies.

[ebastler]

Edit: Try to trigger both channels at the same time with a manual trigger without using CH Copy Coupling. Not possible. Or attempt to program it.

I just gave this a quick try: Both channels in burst mode with manual trigger, no channel coupling. Selected different amplitudes, frequencies, cycle counts on the two channels. I can use the "trigger" softkey in either channel 1 or 2, and it will trigger a burst in both channels. I can also enable the "Trig Out" option for either channel 1 or 2, and it will provide a rising or falling edge trigger on the AUX output.

My hardware version is 02-02-00-35-00, and firmware 2.01.01.23R8.

[Plasmateur]

Edit: Try to trigger both channels at the same time with a manual trigger without using CH Copy Coupling. Not possible. Or attempt to program it.

I just gave this a quick try: Both channels in burst mode with manual trigger, no channel coupling. Selected different amplitudes, frequencies, cycle counts on the two channels. I can use the "trigger" softkey in either channel 1 or 2, and it will trigger a burst in both channels. I can also enable the "Trig Out" option for either channel 1 or 2, and it will provide a rising or falling edge trigger on the AUX output.

My hardware version is 02-02-00-35-00, and firmware 2.01.01.23R8.

I can confirm. I can also do this. I went into Utilities, System, Set to default. Attempted again and was able to produce the same results.

The problem still remains. Why can't this be done through a SCPI command? The manual -> http://www.siglentamerica.com/USA_website_2014/Documents/Program_Material/SDG_ProgrammingGuide_PG_E03B.pdf

states only CH1 or CH2 can be sent a trigger. But not both at the same time.

[Plasmateur]

Edit: Try to trigger both channels at the same time with a manual trigger without using CH Copy Coupling. Not possible. Or attempt to program it.

I just gave this a quick try: Both channels in burst mode with manual trigger, no channel coupling. Selected different amplitudes, frequencies, cycle counts on the two channels. I can use the "trigger" softkey in either channel 1 or 2, and it will trigger a burst in both channels. I can also enable the "Trig Out" option for either channel 1 or 2, and it will provide a rising or falling edge trigger on the AUX output.

My hardware version is 02-02-00-35-00, and firmware 2.01.01.23R8.

EDIT: I can do it with SCPI command.

Whenever I had attempted to do this in the past. I must have had something wrong in the code. Thanks for pushing me to double check. This is great!

Now I just need to be able to do burst and sweep at the same time without having to use a high pass filter.

[tautech]

The 1032X also can be 'improved' to 1062X model. 

Since I own an SDG1032X, I'm curious to know how I could do that. There may be the answer in this thread, but it is very hard to find it through more than a thousand replies.

SDG1000X models have their own thread and the info you seek is there:
https://www.eevblog.com/forum/testgear/siglent-sdg1000x-waveform-generators/

[ebastler]

Now I just need to be able to do burst and sweep at the same time without having to use a high pass filter.

You have piqued my curiosity there, and with your earlier mention of time-of-flight measurements. Would you mind describing what experiment you have in mind? Thanks!

[Plasmateur]

Now I just need to be able to do burst and sweep at the same time without having to use a high pass filter.

You have piqued my curiosity there, and with your earlier mention of time-of-flight measurements. Would you mind describing what experiment you have in mind? Thanks!

Sure.

Let's say you have some gas-like medium with a really high index of refraction with respect to an electromagnetic wave with a particular frequency propagating though this medium. Well, then it's fairly easy to map out the propagating electromagnetic field topology, and play it as a movie.

Why? 30MHz has a freespace wavelength of around 10meters. But in a medium with a high index of refraction, that wavelength can shrink down to centimeters as the speed of propagation is slowed down to some percentage of the speed of light.

Sweet. So how do we go about mapping out this time varying electric/magnetic field within the gas-like medium?

Have one antenna as your exciter which is connected to a SDG2042X and another one that is connected to an oscilloscope and mounted to a 2D or 3D movable stage. The SDG2042X will send a sinwave with some amount of cycles to the exciting antenna.

Everytime you trigger your SDG2042X via a software trigger, the AUX out will trigger the oscilloscope, while the other channel on the scope will capture the electric/magnetic field seen by the receiving antenna. As you begin to move the receiving antenna away from the exciter, you will notice an increasing distance between the AUX trigger and the beginning of the sinwave pulse seen on your scope.

You can then use all the data collected, which is a function of position and time, to reconstructed the electromagnetic field topology, and watch it evolve in time.

If you have a few horn antennas, this is pretty easy to do in free space with a high enough frequency. But the equipment will cost you a fortune.

[ebastler]

Let's say you have some gas-like medium with a really high index of refraction with respect to an electromagnetic wave with a particular frequency propagating though this medium. Well, then it's fairly easy to map out the propagating electromagnetic field topology, and play it as a movie.

Why? 30MHz has a freespace wavelength of around 10meters. But in a medium with a high index of refraction, that wavelength can shrink down to centimeters as the speed of propagation is slowed down to some percentage of the speed of light.

Ah, thank you! And I assume the gas-ike medium, true to your user name and prior project, is a plasma? 

I have no experience with that branch of physics at all. What dielectric constants, and hence wavelengths, do you obtain? If you get the wavelength down to centimeters, doesn't that still imply a rather sizeable plasma (for a homebuilt system?) if you want to obtain a meaningful spatial map of the propagation?

Back to the SDG2042X, what capability is missing which you need for your measurements? Are you looking for an envelope to modulate your burst amplitude, for a soft start and decay? (Maybe using a modulated continuous sine wave would work better?) Or you could store a modulated wavetrain, with the desired number of cycles, in AWG memory and play it just once?

[panman]

Looking to capture frequency counter statistics via SCPI in either graphical form (screen capture) or numerical form (individual frequencies). Per the programming manual, FCNT? provides some information including the current frequency readings. It does not provide statistical info like the min/max frequencies and the standard deviation. It also appears as there is no SCPI command that can be used to capture the screen display which could also be used to snapshot the data.

Anyone had any luck capturing this data via SCPI? Failing that there is always the cell phone camera

Thanks, Ron

[Plasmateur]

Ah, thank you! And I assume the gas-ike medium, true to your user name and prior project, is a plasma? 

I have no experience with that branch of physics at all. What dielectric constants, and hence wavelengths, do you obtain? If you get the wavelength down to centimeters, doesn't that still imply a rather sizeable plasma (for a homebuilt system?) if you want to obtain a meaningful spatial map of the propagation?

Back to the SDG2042X, what capability is missing which you need for your measurements? Are you looking for an envelope to modulate your burst amplitude, for a soft start and decay? (Maybe using a modulated continuous sine wave would work better?) Or you could store a modulated wavetrain, with the desired number of cycles, in AWG memory and play it just once?

Yep. It would be a plasma. And some wavelengths (depending on the type of plasma wave) can be on to order of micrometers, so you need a pretty high resolution positioning system, and precision diagnostics to map those out. My homebrew system is still missing some components I can't find for cheap - turbo molecular pump. But I have about a cubic centimeter to work with in the lab if a static magnetic field isn't applied.

I've already done the experiments in the lab a few years ago (using a SDG2042X triggered off a DG535), but I'd like to be able to repeat some of them at home with some modification to them. As of now, I can pretty much do everything now that I figured out what was wrong with my code and knowing I can just add a high pass filter to the SDG2042X when I want to do a "burst sweep".

Sure, the sweep can be triggered on the SDG2042X, but before the trigger, a sinusoid is still output from the device. I can't have that before the experiments start. As you said, I could store a modulate wave train, but that would take some time to program each one. Also, when I attempted to generate a two frequency sinusoid with the Easywave, the intermod products, or the sidebands that I observed on a spectrum analyzer appeared to much much higher than when just combining two channels.

On other short coming is that the sweep time cannot be lower than 1ms. Again, this can be remedied with a high pass filter, but you also don't get the full peak at the start of the sweep wavetrain. No biggie considering this device is just so much better and so much cheaper than the 33250A (keysight/agilent) signal generator for almost a tenth of the cost.

.....As for my DS1054Z oscilloscope, I really wish there was a trigger channel that could be used so my sampling rate wouldn't halve when making this measurement.

[k8943]

.....As for my DS1054Z oscilloscope, I really wish there was a trigger channel that could be used so my sampling rate wouldn't halve when making this measurement.

Might it be possible to send the trigger signal to the same channel of the scope that is used for the measurement? If the pulse duration from AUX out is too long then maybe a pulse waveform from the second channel of the AWG?

[voltsandjolts]

Also, when I attempted to generate a two frequency sinusoid with the Easywave, the intermod products, or the sidebands that I observed on a spectrum analyzer appeared to much much higher than when just combining two channels.

Just in case Easywave was causing issues there, you could double check by creating the two-tone wave in another way.
I wrote a simple script for GNU Octave to create arb waves in Siglent CSV format. The included example is a two tone sine.

https://www.eevblog.com/forum/testgear/siglent-sdg1000-arbitrary-csv-file-format/msg1973726/#msg1973726