FNIRSI-1013D "100MHz" tablet oscilloscope

[dmitrkov]

I started with re-engineering the schematic, and by the looks of it the 8 pin ic might be an i2c eeprom. Since pin 7 is left floating (as well as pins 1, 2 and 3) it does not appear to be a 24LC04 since they need pin 7 (WP) tied either high or low. On semiconductor devices on the other hand allow for these pins to be floating. There is the N24C64 for instance, but probing the bus will need to be done to see if that is right.

Is going to be quite a bit of work, but as an retired engineer I have time on my hands

Are you talking about this?
https://ibb.co/mG7R6R8
https://ibb.co/R475387

this is 25q80dvnig

[pcprogrammer]

No that is the one that holds the fpga configuration. There is another chip nearer to the fpga just below the 50MHz oscillator. (In the attached picture above the osc) It has two connections to the fpga and is also routed to the 4 pin not mounted connector near the winbond chip you pointed out.

[dmitrkov]

No that is the one that holds the fpga configuration. There is another chip nearer to the fpga just below the 50MHz oscillator. (In the attached picture above the osc) It has two connections to the fpga and is also routed to the 4 pin not mounted connector near the winbond chip you pointed out.

I understand. I didn't see her because of the matrix cable 

[pcprogrammer]

I understand. I didn't see her because of the matrix cable 

Is yours also unmarked? (Same as the fpga and the adc's)

[dmitrkov]

I understand. I didn't see her because of the matrix cable 

Is yours also unmarked? (Same as the fpga and the adc's)

Yes, i have a chip unmarked too 

[eljot]

Hi to everyone !  I am back. I was waiting for another touch screen. The same I,ve bought before. As I described this touch screen was working correct till I tried to be back to my previous memory content.
Unfortunately the new touch screen came broken !!! Poor packing. To buy next ? Never !!! I decided to resolder the GT911 from the (new) broken screen to the one I,ve bought before. Operation was successful !!! And the most important - the touch screen again started to work properly !!!
Conclusion. When programming flash memory with ISP something is sending to GT911 and causing something wrong with it. I am fool about it but I will not continue any experiment....  any more !!!

[eljot]

Summarizing my previous actions. Whoever have problem witch touch screen.
The touch screen type ( https://www.aliexpress.com/item/33042057555.html?spm=a2g0s.9042311.0.0.6af64c4dZopY9k ) works correctly with the  W25Q16 flash memory file (old type oscilloscope file). When programming flash memory with ISP the new touch screen must not to be connect (or do it with the old touch screen). I am not sure if the touch screen works with the original W25Q32 memory... maybe yes but when I tried to flash the memory with my original W25Q32 content (with the new touch screen connected to the device) I faild. I dont want to check it.
I think the simplest way is to connect the new touch screen (the link above) and check if it works. When not it will be necessary to change the flash memory with W25Q16 file but never do it with (new) touch screen connectet do the device !!!

[pcprogrammer]

Hi Eljot, bold move, de-soldering and soldering on a flex pcb. What did you use? A hot air gun to get the job done or a soldering iron?
Still a strange story with the GT911. Just to be sure, you are now using the old firmware? (W25Q16 variant you downloaded from this forum)
How did you do the programming of the flash? With the device on or off? (or even with the battery disconnected?) Which programming software did you use?

Been away on family business, so not much progress on the reverse engineering yet.

[eljot]

Hi 
I've resolder the GT911 with hot air. I mean that I've removed GT 911 with hot air and then I soledered it with iron. I was very nervous but I can manage. I programmed flash memory with ISP. It is necessarily to power the main board. I think it is  possible to power the GT911 with the programmer but it is synonymous to power the whole main board. I programmed flash memory with ISP CH341A program and with the old touch screen (tat one I've spoiled) on board so I did not care if something went wrong with the screen. After reprograming I put the new touch screen and it startet to work properly. I used the W25Q16 from that forum. But I must emphasize that I dont know if the touch screen will not work with its native W25Q32 flash just because when I startet to solve the problem with the touch screen malfunction I startet with reprogramming the flash memory. Than when I tried to came back to native memory content I had problem. Jut because I reprogrammed the memory with ISP and the new touch screen connected to main board. Maybe nothing would happen if I had the new touch screen unconnected from the main board. I did not want to check it because I was afraid of buing next touch screen (the third time) . So I am not sure If the touch screen I"ve bought works with W25Q32 memory or not. For sure it works with W25Q16 file.

[pcprogrammer]

Hi Eljot,
I can imagine you were nervous, because it is small and fragile. From my observation of what I got so far of the schematics the device does not need to be powered on for flashing when power is applied to the flash itself. There is a diode in the VCC path of the flash. This might give problems with parasitic power from the clock and data lines, but that depends on the protection system in the F1C100s chip.

The cables I ordered for connecting the touch screen via a breakout board arrived, so later this week I'm going to monitor the I2C data being send to the touch screen. As said it is now a hobby project and might lead to open source scope software. It is in line with interests I have.

[eljot]

Hi, I will be wating for Your new experiences  Hope for improvements. I am glad that my oscilloscope starts to work !!! I also hope that You will manage with Yours.

[pcprogrammer]

It is quite a puzzle, but the first part is almost there. The FPGA with the ADC's. Not sure about the capacitors (just used 1nF for all of them for now) because there are no markings and measuring in circuit is a bit tricky. Also the ferrites are a guess. Still a lot to do.

[pcprogrammer]

The front end looks a little bit different compared to the old one. I removed the tin cans to be able to further reverse engineer the schematic.

[pcprogrammer]

In the front end there is a 5 pin smd (sot23-5) with markings 8751 2023. In the old one the marking is OAAI which probably is an OPA356 operational amplifier. Anyone an idea if it is the same ic?

[iscle]

As said it is now a hobby project and might lead to open source scope software. It is in line with interests I have.

I tried to boot Linux on the F1C100S but I couldn't get any display (and yes, the backlight was on). Without a UART port it's quite hard to know what was going wrong...

I'd like to know what protocol are they using to talk between the F1C100S and FPGA, just so that we don't have to reimplement everything from scratch at the beginning.

[UniSoft]

In the front end there is a 5 pin smd (sot23-5) with markings 8751 2023. In the old one the marking is OAAI which probably is an OPA356 operational amplifier. Anyone an idea if it is the same ic?

OAAI is OPA356
8751 is RS8751XF

I'm more interested in whether the original libraries/SDK of F1C100s can be found.
I started to restore the original code, but after a while I gave up.
It takes a lot of time to restore the library code. In the absence of complete documentation.

[pcprogrammer]

For sure it will be quite a task to get down to the last bit. If the ones that already did some research in the code are willing to share what they have so far it would help.

I'm a bit new to the disassembly and hacking game, so first have to get used to the tools.

The first step of getting to know the hardware is probably child's play compared to getting to know the software

[pcprogrammer]

Filled in the processor part. Can anyone verify the LCD connection. Tried to find pinouts on the net but failed. The display has a barcode on the back with "HT070QB-27W2012HZ". It looks like a 565 RGB setup with sync and clock.
The connection to the FPGA is made with 11 port pins (PE0~PE10), so maybe 8 data and 3 control lines.
Capacitors are still undetermined and just set to 1nF.

[UniSoft]

Can anyone verify the LCD connection. Tried to find pinouts on the net but failed.

[pcprogrammer]

Power supply also done so leaves the analog input. Was able to determine the values of some capacitors and figured that the ones with the same size most likely have the same value

[pcprogrammer]

Schematics are done. No idea if the analog makes any sense. I have never been very good with analog stuff.  More a digital and programming guy

Next up is taking a look at what is being communicated with the touch screen and the eeprom.

[pcprogrammer]

I'm looking for a 7 inch lcd for a lichee nano to do development for the scope but find it hard to obtain specifications of displays offered on aliexpress.
I asked the seller of this one (https://nl.aliexpress.com/item/32691414835.html) for a datasheet, but they don't have it
Searching on google does not render much info either.
I think it is a 50pin and might be the same as what is in the FNIRSI-1013D. If so I only need to make a converter from 50pin to 40pin.

Anyone any suggestions or a pointer to a cheap 40pin 7 inch 800x480 lcd compatible with the lichee nano. Preferably on aliexpress.

[dmitrkov]

Hi, I received two touchscreens from different sellers:
1.https://aliexpress.ru/item/33003864443.html?spm=a2g0s.12269583.0.0.728a4b88KF5Hr5
2.https://aliexpress.ru/item/33042057555.html?spm=a2g0s.12269583.0.0.219c4940cF1t1D&sku_id=67429988410
Both have the GT9157 chip. The first one works great! But the second one does not work at all. What do you think will work if i resolder the chip from the first to the touchscreen of the oscilloscope? Or is it better to change the entire touchscreen at once?

The first photo is the first touchscreen

Others - photo of the second

Added a comparison photo, on the left is the first touchscreen, on the right is the touchscreen of the oscilloscope

P.S. The right edge of the touchscreen is not working on my oscilloscope

[pcprogrammer]

Hi dmitrkov,
it is most likely that the defect in your original touch screen is not in the chip but in the sensor (the glass plate with the electrodes on it) which will not be solved by exchanging the chips. (I might be wrong of course)

Based on your video, one of the tx lines is down. The resolution of the touch screen used here is not that big. Only 16 TX lines and 10 RX lines, so one scan line makes up a whole section of the screen and in your case the buttons on the right.

The one with the HY007 marking looks the same as what I received. It works with inverted coordinates, which is what I'm trying to fix by looking into the hard and software.

Does yours show the same reversal of the coordinates when you connect it to the scope?

The other one looks like it uses more scan lines which might not work properly on the scope. The connection to the glass plate seems to have more traces connected. You can confirm this by looking at the sides of the panel and count how many sections there are. In my picture the gray bar's behind the double sided (black) tape. There are 16 on my sensor for the horizontal and 10 for the vertical axis.

[dmitrkov]

Does yours show the same reversal of the coordinates when you connect it to the scope?

No. The touchscreen I received is doing fine with the coordinate system. Works correctly.

I read that the chip contains the calibration data of the glass. So I think it makes no sense to re-solder the chip ... I'm waiting for a string from China to separate the glass (https://aliexpress.ru/item/1005001476505787.html?spm=a2g0s.12269583.0.0.901fbdd6O044bZ). I will try to replace the touchscreen entirely