Rigol New Function Gen DG822Pro Teardown

[PELL]

I bought a DG822Pro and take it apart

I am not an expert in test equipment. that's why I only show you some teardown photos instead of "reviewing it", in my opinion, good enough for casual use.

The FPGA is Artix7

[DaneLaw]

I bought a DG822Pro and take it apart
I am not an expert in test equipment. that's why I only show you some teardown photos instead of "reviewing it", in my opinion, good enough for casual use.

Dont you OP already have one thread on the DG800 series.  https://www.eevblog.com/forum/testgear/rigol-just-launch-its-new-function-generator-dg800pro-900pro/msg5148759/#msg5148759
Why not post it there, unless the number of threads is some merits in itself and every aspect, needs to carry its own thread.

Post a screen dump of your DG822PRO stats (infopage "About") ..would like to see the stats on what present models are running, if you just purchased one and it's your portfolio of different tear down pictures above, as I had one since mid-december 2023 and IMO they are quite a mixed bag, not least when trying to compare it to the last-gen DG800 which even seems superior on quite a few factors.

[PELL]

Thanks for your advice

I would think posting these photos in a separate thread would be easier to find and read

[DaneLaw]

Thanks for your advice

I would think posting these photos in a separate thread would be easier to find and read

Usually, people just search for relevant things in the main thread, if a given thread is getting too big, which is far from the case on your other thread, but newermind..
Can you take a picture of your "about page".?

[TurboTom]

PELL, thanks for taking and sharing these photos. 

Basically, here we've got the usual layout of an AWG with the SOC module, FPGA with attached smaple memory, high-speed DAC (here 2.5GSa/s), PLL section (labeled "unknown"), switched attenuators (relays close to the outputs) and some high speed output amplifiers / drivers.

I find two details quite interesting here: Rigol provided two different reconstruction filters for each channel that are assiged to the (individually switchable) four outputs of the Texas Instruments DAC38J82. This may be related to the two product lines (DG800 Pro and DG900 Pro) with their considerably different sample frequencies to be able to get along with a single hardware, but it may also provide for different filter envelopes for certain frequency ranges or waveform shapes (I rather assume it's the first... ). The other peculiar detail is that Rigol spun their own output driver hybrid. It appears to be a metal core PCB with either bonded or soldered amplifiers and passives that have afterwards been potted in epoxy. Since the AWG doesn't provide any higher output amplitude than "conventional" approaches with TI's THS3491 (or its siblings), and the output drivers of an AWG are at least somewhat exposed to "mishaps", I consider the utilization of a hybrid module here rather disadvantageous -- a DIY repair would hardly be possible and I doubt Rigol would provide these hybrids to customers as spares.

Shame that Rigol didn't attach the DAC with all eight high-speed serial lanes to the FPGA (maybe the FPGA doesn't provide more connectivity, I didn't check for that just consulted the DS: the Artix in the 15x15mm² enclosure is only available with a maximum of 4 GTP transceivers, so in order to use all 8 lanes, a higher FPGA type would have been req'd); otherwise it should have been possible to run the hardware at the DAC's full 2.5MSa/s.

Otherwise, I'ld bet that the 800 Pro can be made believe it's a 900 Pro -- which it actually probably is anyway, except for some calibration points, just like the "non-Pro" versions were. 

[PELL]

Thanks for your advice

I would think posting these photos in a separate thread would be easier to find and read

Can you take a picture of your "about page".?

Oh sorry, I was so tired when reply this and didn't see it. So here is:

[PELL]

PLL section (labeled "unknown")

Thank you, I was unable to find it at the beginning. The PLL chip model is LMK04826

[ebastler]

There seems to be a lot of distortion on that sine wave, both the yellow and blue one. 

Sheesh, all that screen resolution overkill, and what does the graphics designer do with it? Hard-to-read numbers and some made-up wiggly lines.

[TurboTom]

There seems to be a lot of distortion on that sine wave, both the yellow and blue one. 

Sheesh, all that screen resolution overkill, and what does the graphics designer do with it? Hard-to-read numbers and some made-up wiggly lines.

... +1 ...  I wholeheartedly agree! I really don't get it why it's necessary to "invent" a horizontally scrolling ancient-airport-tableau-style font for an AWG 
I already strongly disliked Rigol's approach to have the hardware frequency counter displayed in a sixteen-segment style font on their DHO series scopes.

They really need to get some product designers that are worth their payment (maybe they are -- and this rather tells something about their salary  ).

Whatsoever, since the fonts are accessible in the android file system, it may be possible to replace the "fancy" ones with something better readable...

[DaneLaw]

The big touch screen is a pleasure to work with on a signal-gen and rig or stack-up waveforms, no doubt about that but not a lot of polishing went into the UI on this new Rigol PRO-line of signal gens.
Its quite rudimentary, not least when having the last gen in mind, which perhaps was too graphical "polished" for a lab product with some of the color-tone grading in the UI & page-flipping, but on videos, it did look good for a product in that pricebracket.
These 7" touchscreens (1024x600) do make adjustments/setup and digits easy to read on the big screen in regards to the more cassette-style SG's that often had low resolution alongside an extensive signal-gen UI that makes the adjustment somewhat cumbersome.
- at 250 to 300 euros it gonna be interesting to see how liberateable this 16-bit DH800PRO line actually is, as if you can whack it a good part of the Pro-series model ladder, it could be an attractive 16-bit signal gen with decent memory & fast sample rate at this lower pricelevel.

[DaneLaw]

Thanks for your advice

I would think posting these photos in a separate thread would be easier to find and read

Can you take a picture of your "about page".?

Oh sorry, I was so tired when reply this and didn't see it. So here is:

Thx..
Its identical, wondered if the update from 12-1 2024 was implemented from scrach but the values are the same as mine.
You havent updated with the 12-1 2024 DG800/900PRO firmware file ..?
 
GEL-fw file is the same between DG800PRO & DG900PRO..
File https://download.rigol.com/cn/Firmware/Waveform%20Generator/DG800%26DG900%20Pro/DG800Pro_DG900Pro(ARM)Update.rar

Can you translate what the release notes says o the above update, as none of my script standards here in the West will play along.
Source https://www.rigol.com/products/DG_detail/DG800Pro

[PELL]

Can you translate what the release notes says o the above update, as none of my script standards here in the West will play along.

Me either 

The txt file is obviously encrypted. If you open it with a binary editor you will see "TSD-Header" which indicates this file is encrypted.

This is not the first time they forgot to decrypt txt file before releasing firmware. e.g. MSO8000 firmware release note.

[jackcheng]

Can you translate what the release notes says o the above update, as none of my script standards here in the West will play along.

Me either 

The txt file is obviously encrypted. If you open it with a binary editor you will see "TSD-Header" which indicates this file is encrypted.

This is not the first time they forgot to decrypt txt file before releasing firmware. e.g. MSO8000 firmware release note.

I ask rigol and they now fix the release fw and version file. They say the origin file is encrypted.