FNIRSI-1013D "100MHz" tablet oscilloscope

[mishail71]

I accidentally clicked "UPLOAD CPI BACKUP Flash," and after that the screen went black. I have this programmer. What firmware should I upload? If it's the original firmware, I've already tried it. The result was zero. The device won't turn on at all.
Thank you

[Atlan]

If there is a bootloader, then you can upload the program, connect a programmer or use a clip, turn on the oscilloscope and try to detect the chip. Try loading the content several times and watch if the loaded data does not change. If it does not change, you can overwrite the original fnirasi file (the file that the osi uploaded is not fnirasi but an alternative, so you do not have a backup) use this one. Another thing you can do with the flash is to write fuses against programming. Give a photo of what software you use for programming. The programmer is adjusted to 3.3V because it usually puts 5V there

[mishail71]

Atlant, please tell me if I need to unsolder four wires to flash the firmware...the program I'm using is CH341A_C. And is it possible to flash the SPI off-board? Thanks.

Forgive me, this is all a dark forest!!!
Atlant, could you tell me where I can watch a video showing this solution? The operating and boot (recovery) algorithms are completely unknown to me! Thank you. In the post above, you described the device recovery algorithm. But I don't understand these terms...))

[Atlan]

Desoldering the SPI flash is the safest way to program it.

[mishail71]

Atlant, that's what I did... I just don't get it, do I need to upload your fpga-1013D-1-zb25vq80.bin.txt file to the SPI?
It's probably a coincidence... and the controller itself has failed. The power indicator doesn't work when turned on. The display backlight doesn't light. The computer doesn't recognize the connected device. Oh well... I never used it anyway. Thanks, Atlant, for all the help.

[Atlan]

The backlight is turned on by the FPGA. The uP also checks for the presence of the FPGA at startup, if the fpga is not working, the uP remains in a loop. Yes, open the file in the programmer, detect the memory and upload the data. I already asked you for a photo of the software. If you want to help.

More than once, I managed to block the oscilloscope with a bad fpga program.
Yes, it looks dead, but don't worry. Just need the right FPGA content, it will come to life.

[Atlan]

v1.00n fix switch sensitivity in main screen without update base time. Add orange center line for single mode. Because of the resolution of tapping above and below the X axis, it has different functions.

[Atlan]

Atlant, that's what I did... I just don't get it, do I need to upload your fpga-1013D-1-zb25vq80.bin.txt file to the SPI?
It's probably a coincidence... and the controller itself has failed. The power indicator doesn't work when turned on. The display backlight doesn't light. The computer doesn't recognize the connected device. Oh well... I never used it anyway. Thanks, Atlant, for all the help.

PLEASE NOTE: modify the programmer to 3.3V or use a programmer with a switching voltage, or install a 1117 3V3 on a breadboard. See instructions on google.


Here is a video , a demonstration of how to read the contents of the flash, I tried it on the flash where the firmware is inside, but you use what is for the FPGA (basically it doesn't matter) I loaded the contents, ideally it would be a good idea to verify if it loads correctly. Then open the file with the contents for the FPGA and press GO, this will load the correct contents into the flash. Put the memory back into the oscilloscope (pay attention to the orientation of the pins) and it should be turned on.

Possible problems:
1. The FLASH contents must not be locked, there are special bits for this - something can be removed and the memory can be used again - or new memory must be bought.
2. BE CAREFUL about the position of the pins in the programmer and when soldering back, 1 must be on 1.
3. Warning, sometimes the big 3V regulator breaks, so it is advisable to measure as soon as possible whether it is giving 3V as it should.

[mishail71]

Atlant, thank you. I'll try to use your advice... Especially thanks to you, we now have your video with the work algorithm... Good health to you and all the best!!!

[Atlan]

From 6000 to 8000 samples I see some anomalies on CH1 (thereby the frequency calculation is wrong), interestingly it doesn't happen on CH2. That's why I needed some testers. To check it.



Make a video of how to program it, and after watching it I'll tell you if you can solder it to an oscilloscope. When I have time over the weekend I'll make some adjustments to the Updater and remove some things.

[mishail71]

I ordered the complete programmer kit. I'll try to get to work on the device soon...and make a video...Although YouTube is a big problem in Russia right now.))) Thanks again, Atlant.

[Atlan]

Try rutube

[Atlan]

Added buttons to the ACQ menu, for selecting the memory depth of 3000-15000 samples. And buttons for Trace - I recommend using PRO or PEAK, (you can try the difference in display if you put a rectangular signal 10khz duty 1%, 2MSa/500us/div there) the more memory depth the better.

There is still backward compatibility with fnirsi FPGA. But you can't set the sample memory to more than 3000 samples, trigger only 50%, and the generator setting will not work. These things require modified fpga firmware.

[engineer.r152]

Good afternoon! I will support the project with my contribution. I made all the improvements and went through all the steps.
1. I installed Bootloader 0.8, updater 0.2u, scope v1.00n
2. Made a copy of flash spin
3. Flash has been unsoldered
4. Soldered the jumpers between the connectors
5. I made a backup copy of the fnirsi FPGA, a photo of the flash drive and a copy of the firmware.
6. Installed FPGA
Everything worked and installed without errors.
Problems I found after installation
1. The oscilloscopes of channel 1 and 2, which are located when shifted to the left behind the screen, are not stable and do not correspond to the real signal, if the trigger is set to 75%, then a similar effect occurs at the beginning of the signal
2. The sensor in the left half of the screen does not work correctly, when pressed at the bottom, the measuring range switches instead of changing the scan time, this effect is present in several places on the screen
3. The V and T measurement cursors are very difficult to select and shift, the area where they are pressed does not work correctly, and the priority when pressing the display is not the cursors, but the waveforms.
4. If you turn on the oscilloscope when there is a signal at the input, the rays are red.
All the problems are shown in the video.
Is the generator menu not active yet? Changing the settings does not affect the output signal.

https://rutube.ru/video/private/a701d4730ecd78b6fd1e1ea21f9cd548/?p=1aHaMPT_wu71vYz-bQ2eNw

https://rutube.ru/video/private/6437dd32f557c4b24e5e8dc1d788bd58/?p=FLfn8t8VMPPY5jOLYY9_LA

[engineer.r152]

I installed version 1.00n3 and when turned on, there is garbage on the screen.

[Atlan]

You do not have a new fpga firmware uploaded.

Here is the new Updater, it allows you to upload only fpga spi flash, either the original fnirsi (if you download it in the oscilloscope using download or copy it using a PC) and upload peco Atlan fpga. I edited it to avoid the error like one user did here.

1. Upload UPDATER 0.02v to the oscilloscope.
2. Copy the alternative FPGA.bin file to the "Atlan & PEco FPGA firmware" folder
3. Unmount the disk.
4. Press the Upload button in the PECO firmware section: FPGA firmware
5.restart and run DSO firmvare 1.00n3
6. Reset the settings with the orange button to default in the diagnostic menu.

[engineer.r152]

I followed all the steps and it worked! New
noticed flaws:
1. The output waveform SQUR, SINE, TRGL does not switch in the pulse generator menu and the menu items ampl, offset, phaser do not work.
2. If you set 15000 samples, then at the end the signal is displayed as 0
3. The problem with cursors remains, it is very difficult to choose
4. The left part of the screen still works in some areas as if there are hidden V+, V- buttons
 Is a new trigger function of 25%, 50%, 75% works only for 3000 samples, it can be further refined so that it works correctly for 6000, 9000...
I didn't quite understand the Trace functions, can I explain what they mean?

[Atlan]

1. That's okay, the hardware doesn't support it.
2. Memory is up to 163xx samples. It's OK.
3 I don't know, I'll check.
 4 That's how you wanted it.
5. Yes, I know about that, fixed values ​​are entered 25% 750 samples, 75% 2225 samples.

 Trace-the way the samples are drawn on the screen. The screen allows you to display 723 points, if you sample 3000 samples something will be thrown away, the question is what and what will you see then, res. You won't see.

[Atlan]

Turn off CH1 and try only ch2

[engineer.r152]

Unfortunately, it also incorrectly measures the frequency at more than 6000 samples.

[Atlan]

Yes, I know about that. But. The signal is red only on CH1 and on CH2 is it ok?

[rozwell]

A little side suggestion: add very subtle outlined background markings for touchable areas with text/symbols of their functions + config option to turn them on/off, optionally with configurable transparency.
This will stop the touch interface confusion once and for all, maybe allow to find some bugs or simply prove there are none.
I'm pretty sure most people will leave this feature on permanently

[Atlan]

FNIRSI 1013D  alternative firmware v1.00n3v ANG v1.doc

[Atlan]

If it passed verification and the protection bits were not overwritten. That's how it should work. As far as I know, the FPGA uses 25vq80 1MB memory, whereas you have 2MB. How will it work, I'm just suspicious, and I guess it has a slightly wider case than 1MB. Why didn't you use the memory that was original for the FPGA?

Upload a photo of the oscilloscope PCB here.

[mishail71]

No, this is his native SPI. 25VQ16ASIG
Thanks, Atlant... You know, for some reason, the stock CPI didn't work. I installed a 25VQ32. And, surprisingly, everything worked fine with it.
Hooray...