Author Topic: FNIRSI-1013D "100MHz" tablet oscilloscope (Read 407586 times)

wolfy007 · « **Reply #525 on:** February 16, 2021, 12:00:08 pm »

Quote from: hans1961 on February 16, 2021, 11:24:57 am

Has anyone experience with the 1013D measuring voltages above 400 volts? With a 1: 100 probe head?
Is this possible with this oscilloscope? Does it have a default setting of 1:100?

Yes it does have a x100 scale on the channel settings.

Fungus · « **Reply #526 on:** February 16, 2021, 01:23:25 pm »

The advert says:

(but it also says "100M bandwidth 1GS sampling rate" and we know that's not true

)

eljot · « **Reply #527 on:** February 19, 2021, 02:59:56 pm »

Hi, I received the touch screen yesterday. First I've veryfied 6pin tape connection. 1. RST, 2. Vcc, 3. GND, 4. INT, 5. SDA, 6. SCL. It was the same that the original. I,ve found that the screen has its own pull up resistors (SDA, SCL). On the main board there are already pull up resisors. So one of them should be removed. I decided to remove them from the touch screen.
Time to connect everyting together. The scope starts to work. Touch screen works properly !!!

I've noticed that because of the touch screen is a little bit biger than the original it seems the picture fits the whole LCD. It is a little biger so the front glass frame could be now smaler (2..3 mm?). I'm not sure of this efect because I cant compare it to the original.
The touch screen has its own adhensive tape so I've glued it to LCD. Next I've put double sided adhensive tape on the front glass (enough wide to be possible to glue plastic housing and then screen). I put the glass into plastic housing and then put LCD screen).

eljot · « **Reply #528 on:** February 21, 2021, 08:19:35 pm »

HELP!!! Disaster has happened !!! As I wrote above I put the new touch screen. Everything started to work properly. I decided to write back my original W25Q32 content of the memory. After I had done this the new touch screen stopped working. It reacted in some random places but generally it was comletly uncalibrated. I have changed a few times the content of the memory (W25Q16 and W25Q32) but with no effect. What happened?

pcprogrammer · « **Reply #529 on:** February 23, 2021, 11:24:29 am »

Hi Eljot, that is very strange. The fact that it worked at first tells us that it did respond the way it should. Also since you replaced the firmware before with the old touchscreen on it and that did not break the system makes it rather difficult to think of what might have gone wrong.

Does the scope work in the sense that if a signal is attached you see a proper response on the screen?

Not sure about the working of a touchscreen but it might need some calibration. Aliexpress just shipped mine so it will take a couple of weeks before I can test things.

Did you test it throuhgout the different stages of the repair? So after taping it to the lcd and then after taping the pair to the front glass panel?

eljot · « **Reply #530 on:** February 23, 2021, 01:28:14 pm »

Hi, Pcprogrammer

I think that during writening the memory something is sending to GT911 IC. The same I have found on youtube. Someone has written the memory in 1013d and the touch screen stopped work properly.
Conclusion: newer write the memory when the touch screen is connected the main board !!! I think that one thing I can do is to buy another touch screen. The same I've bought before

. I think that the TG 911 IC is programmed by the vendor and something happens when ISP is working in the main board. Maybe next conclusion: not every touch screen with GT911 seems to work properly in the scope. The one I've bought did... until I,ve written the memory.

pcprogrammer · « **Reply #531 on:** February 23, 2021, 01:57:12 pm »

Hi Eljot,
I don't think that is likely, unless the cpu does strange things while you are programming the flash memory and the GT911 has some sort of eprom storage, or the flash is working on the same i2c bus, which I doubt since it will be SPI.

Looking at the video you mentioned shows that for him the X coordinates of the touch are somehow inverted and not really a random effect that you describe.

A problem with your device might be that the touchscreen is no longer responding well behind the glass panel. What I found on the net is that when a touchscreen is placed behind glass or acrylic it should be glued entirely. Take a look at what they say on this page: https://riverdi.com/capacitive-touch-panel-construction-and-working-principles/

eljot · « **Reply #532 on:** February 23, 2021, 04:18:30 pm »

Everything was ok after I had installed the new touch screen. Even when I've installed the front glass without a glue. I am angry with me because I could work with the old version program. I dont know if it is even any difference. But of course I had to go back to my native program. What for ? I dont know !!! I'am fool !!! Touch screen still reacts but has totally lost coordinates. It reacts not a random way, rather with shift coodinates.

pcprogrammer · « **Reply #533 on:** February 23, 2021, 04:23:49 pm »

Very strange indeed. Can you shoot a video of what it does and upload that to youtube. Might give some insight in what is wrong.

Another idea: disconnected the battery and waite some time to clear residual memory content. After that power it up again and see what happens.

eljot · « **Reply #534 on:** February 23, 2021, 04:29:28 pm »

I will try to make a movie but now it is in pieces. I must put together everything

pcprogrammer · « **Reply #535 on:** February 23, 2021, 04:32:14 pm »

Ok. I will check later to see if you wrote a new entry

$:=\$

pcprogrammer · « **Reply #536 on:** February 24, 2021, 11:48:37 am »

Hi Eljot,
you mentioned that you removed the pullup resistors on the touchscreen. So I thought lets check the ones on the main board. These resistors are ~7K5 which is rather high for I2C where on 5V it is normal to use 2K2 and on 3V3 I guess they could be lower. So removing the resistors on the touchscreen was not necessary and might even cause problems considering the weird capacitor across sda and scl on the main board

It is just a thought so take from it what you will

dmitrkov · « **Reply #537 on:** March 03, 2021, 08:25:46 am »

Hi Eljot, which touchscreen is still better to order in your opinion? 7 or 6.2 inches?

pcprogrammer · « **Reply #538 on:** March 08, 2021, 01:20:18 pm »

Hi Dmitrkov, I just received the one I ordered. The size (7 inch) is correct, but unfortunately the coordinate system is wrong. So to get it to work I need to rotate the touch screen. Outside the case it is doable and works, but due to different border sizes it will not fit in the case.

The one I ordered is from this shop: https://nl.aliexpress.com/item/33003864443.html?spm=a2g0s.9042311.0.0.6a9d4c4dko7B3Z
I choose the 7 inch version.

It arrived well packaged and with a useless extension cable (only 4 pins)

Not sure how to solve the incorrectness but maybe the older firmware Eljot used in his tests might solve it. I know from fast scanning the GT911 datasheet there is a setting to invert the X direction but I guess this would involve modifying the existing firmware, which at the moment is one leap to far.

Looked at the chip on the flexboard of the touch screen and it turns out to be a GT9157 and not a GT911. Not sure about the differences yet. The number of used scan lines on the panel is the same as the original one. 10 RX and 16 TX lines.

Here you can see how it is in the normal mounting position: https://youtu.be/cuPgxOR_OlM

DaneLaw · « **Reply #539 on:** March 09, 2021, 12:57:54 pm »

Quote from: Numex106 on November 11, 2020, 08:11:20 pm

I asked on another post and it seems like Apple likes to use these, too??

https://youtu.be/5AwdkGKmZ0I?t=943

Hilarious, the Fnirsi1013D a poster boy in Apple's promo vids about going above and beyond and their highly clinical lab.
but also very clearly shows, that such a tool do have a place, where its relevant and the big bulky lab-grade scopes in the back, are perhaps not ideal, even though in such an environment lab scope ftw...
It looks like each of the 4 Apple workstations has a Fnirsi1013, alongside the Apple 5K display and other lab gear, and then there is one big labscope they can share in the middle on a rolling table..

but can also be a stage, solely for vid purpose, certainly looks like a rigged lap for vid & promo purpose..

Fungus · « **Reply #540 on:** March 09, 2021, 01:20:24 pm »

Quote from: DaneLaw on March 09, 2021, 12:57:54 pm

but can also be a stage, solely for vid purpose.

Does that look like a real workplace to anybody here...?

Where's the coffee cups?

DaneLaw · « **Reply #541 on:** March 09, 2021, 01:30:00 pm »

If you take a look at the vid, it becomes quite clear, that it's rigged.
to many circuits on display with blinking LEDs all over.
though the rolling table with an expensive lab-grade-scope, on wheels, would make sense amongst numerous worktables..
8 workplaces, with a bonafide legit Fnirsi1013D on each of the 4 main table.
c'mon Apple you can give each workstation a full fledge Mac Pro tower and a 5K Display, but they need to share Fnirsi's.

https://youtu.be/5AwdkGKmZ0I?t=547

pcprogrammer · « **Reply #542 on:** March 09, 2021, 08:59:26 pm »

Got a reply from FNIRSI after my request about spare parts. It landed in my spam folder and just only noticed it today (20 days later) Did not matter, because they state "hi, friend
We are very sorry, but we cannot provide you with replacement parts!" In a second email they state "Or you can go to the store where you bought this product and look for customer service!"

Already got large part of my money back so the later is not an option. Leaves the route of hacking the firmware to get the orientation of the new touch screen to line up with the lcd. There is a register in the GT911 (and most likely also in the GT9157 of the new touch screen I bought) where the cpu can set x2x and y2y to invert the coordinates. To investigate what the cpu is doing I'm going to monitor the I2C bus, but have to wait until the needed ffc/fpc cables arrive I ordered to make a breakout setup. Don't like soldering in the device just yet.

As of now it is a true hobby project.

I read way back in this thread that some people tried to hack it but did not got to the part (if there) where they state it is done and they have new software for it. So need to do some more reading.

And yeah the Apple lab looks a bit to clean to be the real deal.

pcprogrammer · « **Reply #543 on:** March 10, 2021, 08:42:42 am »

Has anyone reversed engineered the schematics of this device?
What is the status of the creating open source software for it?

I noticed that there is a second winbond flash on the board which by the looks of it is connected to the FPGA so most likely the configuration for it is stored in there and not in the main flash connected to the F1C100s.

Am I correct in stating that the proper way of programming these flashes is by not powering up the device and connect a programmer with its own power to the 6 pin headers found near the flash chips? (Or by using a special clip on the flash chip) There is a diode in the power connection of the flash, so this way the rest of the device is not powered.

Near the FPGA there is an other 8 pin chip (just below the oscillator) with the markings removed. Has anyone any idea on what this chip is or does?

KPL · « **Reply #544 on:** March 12, 2021, 08:57:17 am »

Hi,
I'm evaluating this "scope" as a battery-operated tool to use when isolation from ground might be needed, like driver circuits in SMPS.
Is the limited sensitivity actually a problem in this kind of use? Or is it "good enough" to use for this? With 10x probe that sensitivity should show driving pulses well enough, but will small artefacts like ringing and overshoots be visible well?

I see Hantek 2D72/2D72 as another option for this, but bigger screen of 1013D seems like a nice feature. I do not think touch screen or buttons would make a big difference to me.

Are there any realistic options to cheaply build an active probe that could eliminate this sensitivity problem, for lower frequencies at least?

I haven't had to deal with scopes since my old analog one died several years ago, and I can not decide which of the current DSO's to buy. I have never handled a DSO. Cheap handheld scope might be the one to use while I'm thinking about spending bigger money.

tunk · « **Reply #545 on:** March 12, 2021, 10:44:24 am »

Quote

I see Hantek 2D72/2D72 as another option for this ...

You could also take a look at the Owon HDS200 series with a 3.5" screen.

KPL · « **Reply #546 on:** March 12, 2021, 11:51:14 am »

Quote from: tunk on March 12, 2021, 10:44:24 am

Quote
I see Hantek 2D72/2D72 as another option for this ...
You could also take a look at the Owon HDS200 series with a 3.5" screen.

Yes, I have noticed those as well, but looks like those are just the same as mentioned Hantek's?
I still have not seen a proper review, is there any?

Then, I am kind of not interested in those multimeter-related functions, so would like to pay for more scope-related ones instead.
Sure, one can't have enough multimeters...

Fungus · « **Reply #547 on:** March 12, 2021, 11:57:58 am »

Quote from: KPL on March 12, 2021, 08:57:17 am

Are there any realistic options to cheaply build an active probe that could eliminate this sensitivity problem, for lower frequencies at least?

Yes, a simple op-amp will do it.

I don't know your exact needs but start here: https://www.aliexpress.com/wholesale?catId=502&SearchText=signal%20amplifier

pcprogrammer · « **Reply #548 on:** March 12, 2021, 01:23:03 pm »

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

wavoigt · « **Reply #549 on:** March 12, 2021, 01:42:19 pm »

Load, View, Analyze FNIRSI 1013D .wav files in Excel with VBA:
https://github.com/wavoigt/FNIRSI-1013D-WAV-Viewer-in-Excel-VBA

Credits @btidey


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: FNIRSI-1013D "100MHz" tablet oscilloscope (Read 407586 times)

Share me