New 2ch pocket DSO+SG - Zeeweii DSO2512G

[Aldo22]

V1.3.0C has the same "bug". I have the formula that calculates the "Mean" value from the raw FPGA data so I will check if it is a bug or a feature 

Thanks.
It is interesting that the line on the graticule seems to be more accurate than the number (Mean).

PS I came across this because I thought I had made a mistake with my screenshot analyzer tool, which I am currently fiddling with (last screenshot). 

[timschuerewegen]

It is interesting that the line on the graticule seems to be more accurate than the number (Mean).

This is the formula that calculates the "Mean" value from the raw FPGA data.

Mean = (ChanGain * ((((ChanVolDiv / 25) * DataSum) / DataCount) - (ChanVolDiv * ChanPosZeroVolt / 25))) / 970

ChanGain is default 1000 and can be changed in steps of 2 by entering "Calib:AC gain" mode
ChanVolDiv is the vertical sensitivity
DataSum is the sum of all FPGA signed 8-bit values
DataCount is the number of FPGA values
ChanPosZeroVolt is the zero voltage position (can be adjusted by moving waveform up/down)
25, 2000 and 970 are fixed

Example:

(1000 * ((((50000 / 25) * 6583) / 1200) - (50000 * -56 / 25))) / 970 = 126774 µV

[timschuerewegen]

This is the formula that calculates the "Mean" value from the raw FPGA data.

That formula basically converts "screen y value" back to µV.

Test with +2V :
Mean = (1000 * ((((1000000 / 25) * 58269) / 1200) - (1000000 * 0 / 25))) / 970 = 2002371 µV

Test with -2V (swapped 2V and GND) :
Mean = (1000 * ((((1000000 / 25) * -62102) / 1200) - (1000000 * 0 / 25))) / 970 = -2134088 µV

Now I need to check if the values came directly from the FPGA or had some calibration applied to it.

(The .wav files attached to this post are the standard DSO2512G .wav blobs with extra data appended to it. The 1200 signed 8-bit values start at offset 0x116C)

[ptluis]

what else can I add/change

About NORMAL trigger mode: if I set the scope to NORMAL TRIGGERING when Stop button is pressed, the trigger must change to STOP and KEEP the signal already acquired instead of going into a wait state that make me loose what's on screen.
If it STOP the trigger, the last signal recorded before the stop can be inspected with the cursors or zoom or even saved.

An option to enable/disable 20MHz bandwidth limit would be nice to have also. I know it's a hardware thing and we could build an external one, but maybe it could be done through software somehow.

[Aldo22]

An option to enable/disable 20MHz bandwidth limit would be nice to have also. I know it's a hardware thing and we could build an external one, but maybe it could be done through software somehow.

I think for the problem with the jittery signal with steep square wave sources (rise time) at higher frequencies, a 20MHz LPF is too much.
I have a 30MHz LPF from the "RF Demo Kit" and a homemade ~64MHz LPF with the parts I had on hand.
The signal is much smoother with the 64MHz LPF than without a filter, but you can still see a 12MHz square wave.
With the 3OMHz filter the same square wave is completely ironed out.

[timschuerewegen]

I am considering to buy this oscilloscope to analyze ~1MHz digital/analog LCD signals.

I almost forgot why I bought this oscilloscope

https://youtu.be/lMQ3HzmfUE4

[ptluis]

I am considering to buy this oscilloscope to analyze ~1MHz digital/analog LCD signals.

I almost forgot why I bought this oscilloscope

https://youtu.be/lMQ3HzmfUE4

(Attachment Link)

What about adding our custom waveforms to the function generator, could that be possible?

[timschuerewegen]

What about adding our custom waveforms to the function generator, could that be possible?

Yes, that is possible. The firmware uses math to generate 256 samples (unsigned 8-bit) for 6 of the 7 waveforms. For the other one it uses hardcoded data.

[RAPo]

So you can update the menu structure of the scope?

What about adding our custom waveforms to the function generator, could that be possible?

Yes, that is possible. The firmware uses math to generate 256 samples (unsigned 8-bit) for 6 of the 7 waveforms. For the other one it uses hardcoded data.

[timschuerewegen]

So you can update the menu structure of the scope?

In general, it depends. In case of adding an extra waveform, it should be easy. I need a small bitmap and 256 bytes.

[Cihan]

I'm seeking a cheap oscilloscope and need help. it doesn't matter desktop or handheld for me. I'm still stuck between dso2512g, dso3d12 and dpox180h. I'll be using oscilloscope mostly for repairing audio amplifier and old UPS. I'm open for other model recommendations too. I thought dpox180h could be best option but after reading comments on bugs, capacitor change on dpox180h dso2512g seemed to be better option but still cannot decide.

[RAPo]

I get that, but how to assign an "action" to such a bitmap.
Could you add for instance a button that sets both channels probes to 10x and both ac-coupled?
Can such a button be assigned to a key?

So you can update the menu structure of the scope?

In general, it depends. In case of adding an extra waveform, it should be easy. I need a small bitmap and 256 bytes.

[Aldo22]

I'm seeking a cheap oscilloscope and need help. it doesn't matter desktop or handheld for me. I'm still stuck between dso2512g, dso3d12 and dpox180h. I'll be using oscilloscope mostly for repairing audio amplifier and old UPS. I'm open for other model recommendations too. I thought dpox180h could be best option but after reading comments on bugs, capacitor change on dpox180h dso2512g seemed to be better option but still cannot decide.

What is your budget?
The dso2512g is the most affordable.
The dpox180h has more bandwidth (which you may not need) and the dso3d12 is a multimeter + oscilloscope.

[timschuerewegen]

What about adding our custom waveforms to the function generator, could that be possible?

Something like this? Or do you mean that you want a way to add your own custom waveforms to the signal generator instead? If so, how do you want to add them? Please be very very specific with feature requests so that I fully understand what you mean/want.
https://youtu.be/3B_Uj_5CezU

[timschuerewegen]

I get that, but how to assign an "action" to such a bitmap.
Could you add for instance a button that sets both channels probes to 10x and both ac-coupled?
Can such a button be assigned to a key?

I was talking about extending the signal generator by adding extra hardcoded waveforms. To do that I need a 40x20 1bpp bitmap and 256 bytes of waveform data.

I can also change the existing menu bitmaps and add extra things but I rather use my own text based menu (activated by long press menu key) with actions/options.

For instance, one of the actions could create a ramdisk (16MB), write the contents of the SPI flash to "flash.bin" and start USB mass storage mode, so that you can "download" the contents of your SPI flash chip via USB.

[RAPo]

Hi Tim,
I wasn't asking for a feature but more for a clarification how far you can "adapt" the scope in terms of menu and associated actions.
I see you from your post that adaption of text based menu is possible.
Can you describe your method in more detail?

I was talking about extending the signal generator by adding extra hardcoded waveforms. To do that I need a 40x20 1bpp bitmap and 256 bytes of waveform data.

I can also change the existing menu bitmaps and add extra things but I rather use my own text based menu (activated by long press menu key) with actions/options.

For instance, one of the actions could create a ramdisk (16MB), write the contents of the SPI flash to "flash.bin" and start USB mass storage mode, so that you can "download" the contents of your SPI flash chip via USB.

[Aldo22]

Something like this?
...
https://youtu.be/3B_Uj_5CezU

The heart looks great! Well done! 

Just a little hint: For me, changing the trigger settings via the menu is tedious.
Fortunately, there is a shortcut for all of them. For example
Power + F1 for switching the trigger level mode ( automatic or manual ).
For more, see the cheat sheet.

[timschuerewegen]

Hi Tim,
I wasn't asking for a feature but more for a clarification how far you can "adapt" the scope in terms of menu and associated actions.
I see you from your post that adaption of text based menu is possible.
Can you describe your method in more detail?

I currently use "long press menu key" to activate drawing of some debug information but I can replace it with a simple text based menu with items that you can select. An item can execute an action (when pressing ok) or changing a setting (by pressing left/right). Much easier to implement than the bitmap based menu system. The menu does not need to be fullscreen and can be drawn on top of everything else, like the debug information.

EDIT: With "action" I mean my own C code, not some predefined "thing".

[Aldo22]

@timschuerewegen

I would also be interested to know what is possible in this way.

I'm still tinkering with my screenshot analyzer.
At the moment I am adding features that the DSO2512G does not have.
For example, an automatic rise-time measurement. (Attachment, 9ns).
Math functions might follow.

- Is it possible for you to add such functions to the FW, or are there limits (memory etc.)?
- Is it possible to build a USB-interface to control the DSO2512G from a PC or to retrieve data?

Thanks!

[timschuerewegen]

I would also be interested to know what is possible in this way.

I'm still tinkering with my screenshot analyzer.
At the moment I am adding features that the DSO2512G does not have.
For example, an automatic rise-time measurement. (Attachment, 9ns).
Math functions might follow.

- Is it possible for you to add such functions to the FW, or are there limits (memory etc.)?

Yes, I think it is possible. The firmware only has the data of what is currently being displayed on the screen, nothing to the left and right, that data simply does not exist in the ARM9 RAM. The firmware (running on ARM9) only receives "visible" data from the FPGA. The firmware basically gives all the parameters (time base, vertical sensitivity, trigger source, trigger type, etc.), to the FPGA and it will aquire the signal which the firmware can then retrieve. In single channel mode for time base 500ns or higher there are actually 4 samples/bytes "per pixel", so 1200 samples/byte in total for one LCD screen (300 pixels). It is on that data that the firmware does its mean/rms/etc. calculations. For time base 5ns there are only ~50 samples/bytes so the firmware uses interpolation to generate a 300 pixel wide waveform. That data is available in the bigger DSO2512G .wav attachments that I shared earlier. You can use those to test/try your math.

- Is it possible to build a USB-interface to control the DSO2512G from a PC or to retrieve data?

Yes and no. I can only create a ramdisk (16MB), create files on the ramdisk and start USB mass storage mode (which can only be exited by powering off). That is how you can transfer data to a PC, or even vice versa, like how the firmware update mechanism works, so you could for instance use this to add YOUR own custom waveforms to the firmware by providing a .bin blob that contains X number of 40x20 bitmaps and 256 byte waveforms.

[Aldo22]

Yes, I think it is possible. The firmware only has the data of what is currently being displayed on the screen, nothing to the left and right, that data simply does not exist in the ARM9 RAM. The firmware (running on ARM9) only receives "visible" data from the FPGA. The firmware basically gives all the parameters (time base, vertical sensitivity, trigger source, trigger type, etc.), to the FPGA and it will aquire the signal which the firmware can then retrieve. In single channel mode for time base 500ns or higher there are actually 4 samples/bytes "per pixel", so 1200 samples/byte in total for one LCD screen (300 pixels). It is on that data that the firmware does its mean/rms/etc. calculations. For time base 5ns there are only ~50 samples/bytes so the firmware uses interpolation to generate a 300 pixel wide waveform. That data is available in the bigger DSO2512G .wav attachments that I shared earlier. You can use those to test/try your math.

OK, thanks for the information.

Yes and no. I can only create a ramdisk (16MB), create files on the ramdisk and start USB mass storage mode (which can only be exited by powering off). That is how you can transfer data to a PC, or even vice versa, like how the firmware update mechanism works, so you could for instance use this to add YOUR own custom waveforms to the firmware by providing a .bin blob that contains X number of 40x20 bitmaps and 256 byte waveforms.

That is not exactly what I meant.
I meant a serial protocol, such as e.g. Feeltech AWGs have (even an FY3200S).
- I can control its settings remotely from another device (PC).
- I can upload waveforms
- I can query measured values (frequency measurement, counters etc.)
and all this in normal operation, not in a special mode.

[timschuerewegen]

That is not exactly what I meant.
I meant a serial protocol, such as e.g. Feeltech AWGs have (even an FY3200S).
- I can control its settings remotely from another device (PC).
- I can upload waveforms
- I can query measured values (frequency measurement, counters etc.)
and all this in normal operation, not in a special mode.

It might be possible. The firmware contains (unused) code for a USB CDC (serial) mode, and it appears that it works and has been set up in a simple echo mode, where the firmware will simply send back any characters it receives.

[Aldo22]

It might be possible. The firmware contains (unused) code for a USB CDC (serial) mode, and it appears that it works and has been set up in a simple echo mode, where the firmware will simply send back any characters it receives.

Hehe!
That could be fun. 
I've already toyed with getting the DSO2512G's video into a self written application. It basically works (Attachment).
Maybe one could control the Zeeweii completely with the PC and have the bigger picture on the PC-monitor?

I mean as an option, I think it's good that the DSO2512G is a standalone device.

[timschuerewegen]

Hehe!
That could be fun. 
I've already toyed with getting the DSO2512G's video into a self written application. It basically works (Attachment).
Maybe one could control the Zeeweii completely with the PC and have the bigger picture on the PC-monitor?

I mean as an option, I think it's good that the DSO2512G is a standalone device.

Those small DSO2512G 2KB .wav files contain all the data for the firmware to generate the .bmp files, because it stores .wav blobs on the SPI flash, not .bmp files, so you could transfer that .wav data via USB to a PC in realtime and an application on the PC could use it to render the waveforms, show the measured values, etc.

EDIT: If you want to process the .wav blobs and want to know where/what/how, ask me on Discord.

[timschuerewegen]

so you could transfer that .wav data via USB to a PC in realtime and an application on the PC could use it to render the waveforms

Proof of concept: