Author Topic: Hacking the Rigol DHO800/900 Scope  (Read 1595913 times)

stj, mb.raghavendra@gmail.com and 15 Guests are viewing this topic.

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1575 on: February 22, 2024, 08:43:22 pm »
No, in reality the numbers on the icons are almost unreadable. For example, it is impossible to see the voltage scale on the channel, horizontal scale parameters, etc. without looking closely. True, I have a matte film pasted on it, which slightly blurs the image, but I think even without it it will be difficult to read small numbers.
Hmm, but I was wondering how such a picture would be displayed on an external high-resolution display...
You see it better on a physically bigger screen.

1920x1080 7" is a lot of pixels very very close together, high physical DPI. 1920x1080 on a 36" monitor is same pixel count, but physical DPI is way less.
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1603
  • Country: ua
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1576 on: February 22, 2024, 08:48:11 pm »
There's a bunch of Android commands that show resolution and, probably, physical display info.

For example:

Code: [Select]
rk3399_rigol:/ $ dumpsys|grep PhysicalDisplayInfo
      PhysicalDisplayInfo{1024 x 600, 60.000004 fps, density 1.425, 213.0 x 213.0 dpi, secure true, appVsyncOffset 1000000, bufferDeadline 16666666}

Code: [Select]
rk3399_rigol:/ $ wm size                                                                                                                     
Physical size: 1024x600

Code: [Select]
1|rk3399_rigol:/ $ dumpsys window displays                                                                                                   
WINDOW MANAGER DISPLAY CONTENTS (dumpsys window displays)
  Display: mDisplayId=0
    init=1024x600 228dpi cur=1024x600 app=1024x600 rng=600x600-1024x1024
    deferred=false layoutNeeded=false

Code: [Select]
rk3399_rigol:/ $ cat /sys/devices/platform/display-subsystem/drm/card0/card0-DSI-1/mode
1024x600p97
rk3399_rigol:/ $ cat /sys/devices/platform/display-subsystem/drm/card0/card0-DSI-1/modes                                                     
1024x600p97

I'm pretty sure it's a physically 1024x600 screen.
 
The following users thanked this post: AndyBig, AceyTech, Randy222

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1577 on: February 22, 2024, 09:03:22 pm »
I'm pretty sure it's a physically 1024x600 screen.
I'm absolutely sure of this.
 
The following users thanked this post: AceyTech, Randy222

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1578 on: February 22, 2024, 09:10:33 pm »
I'm pretty sure it's a physically 1024x600 screen.
I'm absolutely sure of this.
Indeed. I did the phys measure too (camera and a measurable line).

So, we know phys native dpi is 170.6665px/inch
Is the actual source in this native resolution? Ideally we want 1:1 so there's no adding of artifacts or removal of data.
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 6676
  • Country: de
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1579 on: February 22, 2024, 09:13:28 pm »
Hallelujah!  :-+
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1580 on: February 22, 2024, 09:23:39 pm »
So, I played around with different settings for the resolution of the external display (4k) when the resolution of 1440p was specified in the application properties. No improvements are noticeable, it seems that with any settings there is simply scaling from the native physical resolution of the display - 1024x600.
The thickness of the grid lines and the trace always remains the same.
 

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1581 on: February 22, 2024, 09:27:10 pm »
So, this seems odd, but I guess I will come to understand why later.
My reinstall of the Sparrow.apk from Rigol GEL installed as non-system app. And then I had no signal aquisition.
Hmmmm.
Just for giggles, I ran the calibrate routine. Now it aquires signal and shows on the screen.

Also, if I halt rigol launcher, scope still runs, but the controls dont work.
 

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1582 on: February 22, 2024, 09:28:32 pm »
So, I played around with different settings for the resolution of the external display (4k) when the resolution of 1440p was specified in the application properties. No improvements are noticeable, it seems that with any settings there is simply scaling from the native physical resolution of the display - 1024x600.
The thickness of the grid lines and the trace always remains the same.
What setting are you using?
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1583 on: February 22, 2024, 09:30:42 pm »
My reinstall of the Sparrow.apk from Rigol GEL installed as non-system app.
Does it save screenshots from the Storage menu or by clicking the Quick button?
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1584 on: February 22, 2024, 09:33:35 pm »
So, I played around with different settings for the resolution of the external display (4k) when the resolution of 1440p was specified in the application properties. No improvements are noticeable, it seems that with any settings there is simply scaling from the native physical resolution of the display - 1024x600.
The thickness of the grid lines and the trace always remains the same.
What setting are you using?
In the application manifest:
Code: [Select]
         <meta-data android:name="design_width_in_dp" android:value="1024"/>
         <meta-data android:name="design_height_in_dp" android:value="600"/>
I tried a variety of HDMI resolution settings, both from the oscilloscope menu and from the Android settings. From 1280x1024 to 3840x2160.
 
The following users thanked this post: AceyTech

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1585 on: February 22, 2024, 09:57:00 pm »
My reinstall of the Sparrow.apk from Rigol GEL installed as non-system app.
Does it save screenshots from the Storage menu or by clicking the Quick button?
PNG files save a-ok to /data/UserData/

But noted, I reinstalled the Rigol signed APK, not my edited version. For some reason it does not show up in the system apps list (adb shell cmd package list packages -s). Webcontrol and Launcher still do, I did no mess with those yet.
« Last Edit: February 22, 2024, 10:05:17 pm by Randy222 »
 

Offline DaneLaw

  • Frequent Contributor
  • **
  • Posts: 478
  • Country: dk
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1586 on: February 22, 2024, 09:58:52 pm »
So, I played around with different settings for the resolution of the external display (4k) when the resolution of 1440p was specified in the application properties. No improvements are noticeable, it seems that with any settings there is simply scaling from the native physical resolution of the display - 1024x600.
The thickness of the grid lines and the trace always remains the same.
Makes sense, and the safe way to do it.
It was also how it came across on Daves's video, and the waveform looked very rough on the bigger screen.  https://youtu.be/r_BYYgCqScE?si=NH9vGg2ALA7p8_02&t=47
 

Offline Randy222

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: ca
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1587 on: February 22, 2024, 10:12:00 pm »
There's a bunch of Android commands that show resolution and, probably, physical display info.

For example:

Code: [Select]
rk3399_rigol:/ $ dumpsys|grep PhysicalDisplayInfo
      PhysicalDisplayInfo{1024 x 600, 60.000004 fps, density 1.425, 213.0 x 213.0 dpi, secure true, appVsyncOffset 1000000, bufferDeadline 16666666}

Code: [Select]
rk3399_rigol:/ $ wm size                                                                                                                     
Physical size: 1024x600

Code: [Select]
1|rk3399_rigol:/ $ dumpsys window displays                                                                                                   
WINDOW MANAGER DISPLAY CONTENTS (dumpsys window displays)
  Display: mDisplayId=0
    init=1024x600 228dpi cur=1024x600 app=1024x600 rng=600x600-1024x1024
    deferred=false layoutNeeded=false

Code: [Select]
rk3399_rigol:/ $ cat /sys/devices/platform/display-subsystem/drm/card0/card0-DSI-1/mode
1024x600p97
rk3399_rigol:/ $ cat /sys/devices/platform/display-subsystem/drm/card0/card0-DSI-1/modes                                                     
1024x600p97

I'm pretty sure it's a physically 1024x600 screen.

Change the setting in build.prop file, then re-run your dumpsys. Can't always rely on software. ;)

My measure does seem to indicate a physical 1024x600 screen.
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1588 on: February 22, 2024, 10:36:42 pm »
Another good news is that the application works quite well not only after changing some properties in the .xml files, but also after changing the algorithms themselves in the source codes. True, only in the source code of the Java analogue of assembler - smali. I tried changing some small things just to check that the application would work after compiling back. After a quick search, I found an area in the source code that was easy to change - I simply reduced the radius of curvature of the elements in the measurement panel. Everything worked :)
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1589 on: February 22, 2024, 10:37:39 pm »
But noted, I reinstalled the Rigol signed APK, not my edited version.
Ah, well then of course it will save :)
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1603
  • Country: ua
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1590 on: February 22, 2024, 10:54:12 pm »
After a quick search, I found an area in the source code that was easy to change - I simply reduced the radius of curvature of the elements in the measurement panel. Everything worked :)
Maybe one day someone will find how to also remove those icons wasting the precious vertical space.
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1591 on: February 22, 2024, 11:26:41 pm »
Maybe one day someone will find how to also remove those icons wasting the precious vertical space.
That's what I wanted to find, but I couldn't find it quickly :)
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1592 on: February 23, 2024, 01:31:23 am »
Well, it almost worked :) You just need to correct the drawing of the border around the dimension :)
 

Offline jsobell

  • Contributor
  • Posts: 24
  • Country: au
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1593 on: February 23, 2024, 04:01:09 am »
My scope arrived yesterday, and after running for an hour the fan stopped and the back of the scope reached over 85°C.
When I contacted the supplier they said that they would have to get it returned, sent off for testing, etc.etc.
So I opened the scope and found the the fan was faulty, and it has 'dead spots' when it stops and won't start spinning without a flick.
I checked through my junk, found an old 40mm fan (with casing), and chopped off the case to leave only the three support pieces. These fitted when the bolts went, so I just screwed then under the edges of the bolts, and it works perfectly.
The interesting thing is that this fan is also ball-bearing, unlike the stock one. It is nice and quiet, but keeps the CPU at a pretty consistent 57°C (CPU_AMBIENT of 56°) in a 20°C room.
The importer is sending me a replacement for the stock fan, but this one is so much quieter than the stock one (when it worked) that I'm not going to swap it over.
Perhaps taking enclosed ball-bearing fans and chopping them is a better way to go? 40mm is not ideal, but as it works effectively and is so much quieter...
 
The following users thanked this post: AceyTech

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1594 on: February 23, 2024, 04:02:15 am »
Measurements without space-consuming icons in the headers :)
 
The following users thanked this post: thm_w, ebastler, AceyTech, calabr

Offline AceyTech

  • Regular Contributor
  • *
  • Posts: 194
  • Country: us
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1595 on: February 23, 2024, 04:42:29 am »
Measurements without space-consuming icons in the headers :)

That is super hot.!  Wow @AndyBig, Nicely done.   Now, just roll that into a Rigol UI tweak tool and we'll be all set!  ;)  Seriously tho', great job.  So much better looking. 

BTW: It's probably a dirty word to use here, but I honestly think the crappy FNIRSI 1013D 7" tablet scope that I returned displayed it's measurements better than my DHO.
 

Offline AceyTech

  • Regular Contributor
  • *
  • Posts: 194
  • Country: us
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1596 on: February 23, 2024, 04:53:27 am »
My scope arrived yesterday, and after running for an hour the fan stopped and the back of the scope reached over 85°C.
When I contacted the supplier they said that they would have to get it returned, sent off for testing, etc.etc.
So I opened the scope and found the the fan was faulty, and it has 'dead spots' when it stops and won't start spinning without a flick.

Cool little hack.  Nicely done!  Proof that 'not everything new is always better'. 
Side note:  If you care, you might try peeling up the label on the defective fan and look for cold solder joints on the SOT part(s).  There's often a hall effect sensor & FETs under there.
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1597 on: February 23, 2024, 05:03:14 am »
BTW: It's probably a dirty word to use here, but I honestly think the crappy FNIRSI 1013D 7" tablet scope that I returned displayed it's measurements better than my DHO.
It seems to me that only the lazy have not yet scolded the way Rigol shows measurements :)) And quite deservedly, I think.

In a few minutes I will show you an even more condensed version of the measurements :)
« Last Edit: February 23, 2024, 05:05:51 am by AndyBig »
 

Offline AndyBig

  • Frequent Contributor
  • **
  • Posts: 394
  • Country: ru
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1598 on: February 23, 2024, 05:18:35 am »
It seems to me that there is no point in compressing it any more :) It’s already much better than it was in the original :) For comparison, I’ve attached a comparative picture of how it was and how it turned out :)
 
The following users thanked this post: Ivan7enych, jsobell

Offline AceyTech

  • Regular Contributor
  • *
  • Posts: 194
  • Country: us
Re: Hacking the Rigol DHO800/900 Scope
« Reply #1599 on: February 23, 2024, 05:29:20 am »
BTW: It's probably a dirty word to use here, but I honestly think the crappy FNIRSI 1013D 7" tablet scope that I returned displayed it's measurements better than my DHO.
It seems to me that only the lazy have not yet scolded the way Rigol shows measurements :)) And quite deservedly, I think.

In a few minutes I will show you an even more condensed version of the measurements :)

Sorry, I probably should have kept that comment to myself and just celebrated your accomplishments of the day., but I couldn't control it.

So is it extremely difficult to do these UI mods, and have you discovered any other defects like the screenshots not functioning?

It seems to me that there is no point in compressing it any more :) It’s already much better than it was in the original :) For comparison, I’ve attached a comparative picture of how it was and how it turned out :)

IF the label and value was all on one line, it could be smaller.  But would we see enough digits to still be useable?
« Last Edit: February 23, 2024, 05:32:19 am by AceyTech »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf