Author Topic: The Siglent SDG2042X Thread  (Read 269110 times)

alank2 and 2 Guests are viewing this topic.

Online tautech

  • Super Contributor
  • ***
  • Posts: 15688
  • Country: nz
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent SDG2042X Thread
« Reply #450 on: June 25, 2016, 07:53:14 pm »
@ Tautech,

Yes, I did consider making an external add-on to provide analog isolation. But it would probably be more expensive than adding isolation between CPU and GEN boards, and it would also degrade signal quality. 40 MHz is quite a tall order for such a thing... It seems to me as counterproductive to buy a generator for its good signal quality, and then degrade it with an add-on.
I quite understand and after reading your analysis of inserting the isolation within the generator I only suggested it as a alternative option. The ISFE weighs so little that it just hangs from the BNC's of a DSO and while there is some wobble used with care might be better than a standalone isolator that need be cable connected.


Quote
......I'll just have to swap the entire CPU board for a new one. Any idea where I could get one?
Your supplier should be able to source you one from the factory as a spare part.
Member Lightages blew his SDG1000 AWG main PCB and contacted Siglent and sourced a replacement PCB no problems.
Ensure any contact with Siglent reveals you're an EEVblog member....maybe use the links to contact them in the Siglent Support thread.  ;)

Good luck and keep us in the loop.
Avid Rabid Hobbyist
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15688
  • Country: nz
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent SDG2042X Thread
« Reply #451 on: June 25, 2016, 08:00:36 pm »
Tequipment also has a similar thread if you are looking at ordering from them.
TE is no longer stocking Siglent, Saelig is the preferred online seller for Siglent now.

https://www.eevblog.com/forum/testgear/equipment-discounts-from-saelig/

All of the 3 official Siglent websites have the list of authorised sellers and their contact details
http://www.siglentamerica.com/how_to_buy
Avid Rabid Hobbyist
 
The following users thanked this post: saelig

Online tautech

  • Super Contributor
  • ***
  • Posts: 15688
  • Country: nz
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent SDG2042X Thread
« Reply #452 on: June 25, 2016, 08:49:13 pm »
@ Tautech,

Yes, I did consider making an external add-on to provide analog isolation. But it would probably be more expensive than adding isolation between CPU and GEN boards, and it would also degrade signal quality. 40 MHz is quite a tall order for such a thing... It seems to me as counterproductive to buy a generator for its good signal quality, and then degrade it with an add-on.
Further thoughts
The SDG5000 series AWG's have isolated outputs in case you weren't aware and Dave did a teardown of these, they have up to 160 MHz Sine capability (SDG5162).
There may be some clues as to how Siglent accomplished the isolation in the vid.
Avid Rabid Hobbyist
 

Offline CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 459
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #453 on: June 25, 2016, 09:12:05 pm »
TE is no longer stocking Siglent, Saelig is the preferred online seller for Siglent now.

Wow, I know I know that, but clearly wasn't thinking (  :=\ <- me when posting). Thanks for the correction, I will add a note to my previous message.
 

Offline Timpert

  • Contributor
  • Posts: 43
  • Country: nl
  • Country: nl
Re: The Siglent SDG2042X Thread
« Reply #454 on: June 26, 2016, 06:46:35 am »
@ Tautech
Quote
The SDG5000 series AWG's have isolated outputs
The scheme that I propose is the same as used in the SDG5000 and in the SDM3055. The SDM3055 uses the same CPU board as the SDG2000X. In both devices, the CPU is grounded, while the signal processor boards float. The isolators are at the edge of the signal boards, directly connected to the interface from the CPU. It is the most convenient spot to place the isolation barrier, and the two board construction of the SDG2000X makes it possible to place an isolation board in that device too. If it had the single board construction of the SDG1000, it wouldn't be doable.
 
The following users thanked this post: tautech

Offline UNI-T_goodANDbad_FACTS

  • Newbie
  • Posts: 1
Re: The Siglent SDG2042X Thread
« Reply #455 on: June 29, 2016, 09:19:05 pm »
Hi guys,

I followed the Guide for dummies from this thread and it works smooth! More good news: at least two ships are
on the way to Hamburg with hack-able generators on board :D

I made video of the whole hack, was very easy. Very happy with my 120MHz now!
« Last Edit: June 30, 2016, 06:08:31 am by UNI-T_goodANDbad_FACTS »
 
The following users thanked this post: Saronni

Offline cio74

  • Regular Contributor
  • *
  • Posts: 166
  • Country: gb
  • Country: gb
Re: The Siglent SDG2042X Thread
« Reply #456 on: June 30, 2016, 10:13:51 pm »
Do you need 120MHz?
 

Online alank2

  • Super Contributor
  • ***
  • Posts: 1992
Re: The Siglent SDG2042X Thread
« Reply #457 on: June 30, 2016, 11:31:39 pm »
You don't even need to risk hacking it to get 120 MHz.  Just make an arbitrary wave that is 10 sinewaves and set it to 12 MHz.
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1404
  • Country: no
  • Country: no
Re: The Siglent SDG2042X Thread
« Reply #458 on: June 30, 2016, 11:37:51 pm »
Is there any video tutorial about this?
This has to be hard-coded for each specific frequency right?
 

Online alank2

  • Super Contributor
  • ***
  • Posts: 1992
Re: The Siglent SDG2042X Thread
« Reply #459 on: July 01, 2016, 12:12:47 am »
Is there any video tutorial about this?

The file is enclosed.  Just load it into the arb mode (the CSV file extracted from the ZIP)

This has to be hard-coded for each specific frequency right?

No.  The file itself just has a number of points.  You set the frequency on the unit to 1/10th of the frequency you desire.
 

Offline CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 459
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #460 on: July 01, 2016, 12:15:22 am »
You don't even need to risk hacking it to get 120 MHz.  Just make an arbitrary wave that is 10 sinewaves and set it to 12 MHz.

I've heard several people say that but I haven't seen anyone show results from actually doing it? How does the performance of an arbitrary waveform generated that way compare to the 120MHz builtin sine wave? I would think you would have a lot more jitter on the arb version.

Edit: Just asking because I've never seen anyone actually try it. Its certainly possible someone else has (and it was all good) and I just haven't seen it.
« Last Edit: July 01, 2016, 12:16:59 am by CustomEngineerer »
 

Online alank2

  • Super Contributor
  • ***
  • Posts: 1992
Re: The Siglent SDG2042X Thread
« Reply #461 on: July 01, 2016, 12:18:50 am »
In addition to the 2.5k points version, here is the 160k points version.

I've never tested them other than looking at them on a scope.  Perhaps someone else can do some testing and see how they compare.  You can go higher than 120 MHz, but the amplitude decreases sharply.
 
The following users thanked this post: CustomEngineerer, uv3afl

Online tautech

  • Super Contributor
  • ***
  • Posts: 15688
  • Country: nz
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent SDG2042X Thread
« Reply #462 on: July 01, 2016, 12:43:44 am »
rf-loop has described obtaining the max frequency from the SDG1000 series also by using this same Arb method.

IIRC it's early in the SDG1000 and 800 thread.
Avid Rabid Hobbyist
 
The following users thanked this post: CustomEngineerer

Offline CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 459
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #463 on: July 01, 2016, 07:25:23 am »
In addition to the 2.5k points version, here is the 160k points version.

I've never tested them other than looking at them on a scope.  Perhaps someone else can do some testing and see how they compare.  You can go higher than 120 MHz, but the amplitude decreases sharply.

Thanks. I will try playing around with it when i get some time. I've never had much luck getting decent waveforms with EasyWave software, though this is probably due to not enough time learning how to use it correctly (or even thinking the problem through like I should have). The first time someone had suggested doing similar I tried a couple of times, but just ended up with crap results. I seem to remember thinking to be able to get to 120MHz I would need to use as few of points as possible (this is what I get for not taking the time to do the calculations). Seeing how many points are in those arb files though makes sense. DDS has a fixed clock, so as long as the waveform has 6 cycles at 20MHz, it should come out to 120MHz, and the more points (up to a limit) the smoother the output should look. I wanna say I had attempted it with under 100 pts, and had probably been playing around with TrueArb mode before hand and just had the less points thing stuck in my head  :palm:
 

Offline CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 459
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #464 on: July 01, 2016, 07:51:03 am »
I had to give it a try before going to bed. Ran the 2.5K arb version on ch1 at 10MHz (so 120MHz output) and used the builtin sine on ch2 at 120MHz. Put both on the scope at the same time, overlapping one on the other and they line up perfectly. No drifting, both looked clean. I don't have the equipment to do any serious analysis of the waveforms, but using the arb method appears to be a valid replacement for using the builtin sine.

Thanks for the info.

Edit: I found why I was thinking an arb version wouldn't be as clean as the built-in waveform. From an Agilent doc, Measurement Tips: Comparing Function Generator Performance: Direct Digital Synthesis Versus Point-by-Point Technology
http://cp.literature.agilent.com/litweb/pdf/5990-7460EN.pdf

On the third page there is a "Measurement Tip" that states
Quote
If you are using a DDS arbitrary waveform generator such as the Agilent 33220A and you want to create a square wave or pulse function, for the best performance, use the built-in waveform instead of creating the waveform yourself with the arb function. The built-in square and pulse waveform generation techniques are different from the standard DDS technique. Agilent has carefully chosen these techniques to eliminate distortion due to aliasing at higher frequencies.

Of course thats talking about Agilent generators and square/pulse waveforms, so does not apply to a sine wave generated on a Siglent generator  :palm:
« Last Edit: July 01, 2016, 08:02:55 am by CustomEngineerer »
 

Offline bson

  • Supporter
  • ****
  • Posts: 1545
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #465 on: July 01, 2016, 09:06:53 am »
Here are the two csv files, viewed on the Signal Hound running off a Thunderbolt for reference.  The SDG2122X is using is internal clock source.

2.5k



160k



No real difference to speak of.  (I centered the second one on the peak, but forgot to do so on the first.)

Output is 250mV 200mV pp, 50ohm... ARB mode doesn't appear to allow specifying the output level in dBm.  :-//
« Last Edit: July 01, 2016, 09:10:01 am by bson »
 
The following users thanked this post: julian1

Offline bson

  • Supporter
  • ****
  • Posts: 1545
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #466 on: July 01, 2016, 09:16:44 am »
Here's the internal one, using the sine waveform.  Looks a little cleaner.  Again, I forgot to center it on the peak... getting late.  :palm:


 

Offline cio74

  • Regular Contributor
  • *
  • Posts: 166
  • Country: gb
  • Country: gb
Re: The Siglent SDG2042X Thread
« Reply #467 on: July 01, 2016, 02:19:00 pm »
Apologies for the off topic, which Signal Hound SA do you have?
 

Offline bson

  • Supporter
  • ****
  • Posts: 1545
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #468 on: July 01, 2016, 06:03:54 pm »
Apologies for the off topic, which Signal Hound SA do you have?
SA44B
 

Offline cio74

  • Regular Contributor
  • *
  • Posts: 166
  • Country: gb
  • Country: gb
Re: The Siglent SDG2042X Thread
« Reply #469 on: July 01, 2016, 06:43:48 pm »
Apologies for the off topic, which Signal Hound SA do you have?
SA44B

I am looking at the combo with the TG44A, quite the price of the Siglent SA but it looks to have more features.
 

Offline bson

  • Supporter
  • ****
  • Posts: 1545
  • Country: us
  • Country: us
Re: The Siglent SDG2042X Thread
« Reply #470 on: July 01, 2016, 11:15:00 pm »
I am looking at the combo with the TG44A, quite the price of the Siglent SA but it looks to have more features.
I got the TG44 to go with it, but frankly it's exceedingly limited.  I wouldn't bother and instead put the money towards a used VNA.

Some limitations:
* Can't do log sweeps
* Spike can't display magnitude in V
* It's possible to rig it to measure the reflection coeff, but it can't for instance measure reverse isolation which makes it very difficult say to calculate the input impedance for a specific load
* It's only scalar - so can't be used for impedance matching (can't determine its conjugate impedance)

Just get a 2-port VNA that can measure S params both ways and provides phase information...

Edit: oh, and I forgot to mention the TG or Spike doesn't seem to have an amplitude control for the TG; changing the amplitude requires external pads.  Hence it obviously won't do an amplitude sweep... which in turn means it can't calculate the IP3 or such.  It's very, very, very super basic to the point of not being worth the bother.

The SA44B is great to look at signals though, although its FFT span is only 250kHz.  (The BB60 steps up to I think 27MHz in this regard, so you can sit and watch and record frequency hopping in the time domain!)
« Last Edit: July 01, 2016, 11:23:27 pm by bson »
 

Offline McBryce

  • Super Contributor
  • ***
  • Posts: 1281
  • Country: de
  • Country: de
Re: The Siglent SDG2042X Thread
« Reply #471 on: July 05, 2016, 07:20:08 pm »
Hi all,
     I got an SDG2042X today. Upgraded it to 120MHz and then upgraded the Firmware. There's now a small lock symbol with a "P" to the left of the network symbol. Can anyone tell me what this means. I don't think it was there before I upgraded the firmware.

McBryce.

Edit: Forget the question. It shows that the phase lock is on :) Btw. I can also confirm that the "liberated clock" remains after a firmware upgrade.
« Last Edit: July 05, 2016, 07:36:48 pm by McBryce »
 

Offline Timpert

  • Contributor
  • Posts: 43
  • Country: nl
  • Country: nl
Re: The Siglent SDG2042X Thread
« Reply #472 on: July 05, 2016, 08:08:14 pm »
To see if I could do it, I tried to replace the CPU on my broken CPU board. At least now I know that this was a bit beyond my soldering skills. But anyway, since I had voided my warranty because I probed around inside my device, I thought I'd probe around a bit more. More specifically, I probed the SYSBOOT[15:0] pins when I had the CPU off the board. These determine the boot order amongst other things. Normally, the boot order is NAND, NANDI2C, MMC0 (which is the on board SD card), UART0 (which is the UART board on the short edge of the board. With jumper J16 installed, this changes into UART0, XIP (not present, so it is skipped), MMC0, NAND. So with J16 installed, you could probably boot one of the many available Linux images that run on the Sitara processor from the SD card...

Still trying to get a replacement CPU board though. It appears to be difficult.
 
The following users thanked this post: CustomEngineerer

Online tautech

  • Super Contributor
  • ***
  • Posts: 15688
  • Country: nz
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent SDG2042X Thread
« Reply #473 on: July 05, 2016, 08:45:34 pm »
........

Still trying to get a replacement CPU board though. It appears to be difficult.
How so?
Your supplier should be able to source one.  :-//

Let me know if I can help.
Avid Rabid Hobbyist
 

Offline Timpert

  • Contributor
  • Posts: 43
  • Country: nl
  • Country: nl
Re: The Siglent SDG2042X Thread
« Reply #474 on: July 05, 2016, 09:11:32 pm »
Well, my original supplier (Elektor) simply isn't answering my request, so I contacted a Dutch Siglent dealer, and they are now trying to get a quote for a new CPU board from Siglent. But apparently, a request like mine happens very rarely, and so getting Siglent to fish a replacement board out of their logistics chain and put a price on it is going to take some time. But they're working on it. The only thing I can do now is be patient. And I am as good at that as I am good at BGA rework...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf