Power & Source of Big Ideas

Check the schematic to find out which GPIO pin it is connected to.
Do you have the correct permissions set in the .rc file?
Have you tried just connecting to HDMI without the LCD connected and see if you get anything?
What does the debug output show? That is the best way to know what has happened.
The baud rate is very high at 1,500,000 bps. Putty works.
eldridgek1975 wrote:
I have the SIMCOM SIM7600E-PCIE module working with the T4 under Android.


Which Android version? I have it working with 5 and 7 on a T3 but have the drivers for 8 but not yet tested it on the T4.
Hey there. Just moved overseas to Perth, Australia and in quarantine so I was not logged in on the laptop.

Thanks for checking up though, it is much appreciated.

I get out of quarantine on the 25th Feb and once I get to the new home, I'll be back on the main PC where all my code is.
The best solution for RF testing is with a spectrum analyser and come RF probes. They are cheap now compared to a few years ago.
Which OS are you using it with?

If Windows, what does device manager show?

If Linux, what does dmesg show?
That has an 8 bit SSD1963 so you'd need a driver and enough GPIO to use it. From what I can see, the Neo Air is a headless design only. There is no interface for video that would be plug-in. You might be able to get a small form factor SPI interfaced LCD to work but you'll need to find drivers etc. ...
Good day. If you follow the instructions from the integration manual, it is fairly simple to add to the OS build. You'll need a little experience of the OS build to know what files to change. Use the debug UART to monitor the radio debug (logcat -b radio) from the command line. This will indicate if...
The best way to do this is to build your own Android OS and set the SELINUX command line in the build.

You also need to set the tty ports in the init.rc to be accessible from the user space. Setting them to 0666 works for me. I also do this with the I2C and SPI etc.
Double click the file and then top right, choose download.
Be aware that changes to the I2C bus assignments might upset the system although most drivers do a probe to try and locate which bus they are present on.
Have you measured the voltage from the 5V supply?

That IC you refer to if labeled U110 is a regulator, MP2143DJ, and on pin 5 there should be 3.3V and pin 2 is the 5V input. This regulator is enabled by the PMIC.
It should be possible to swap them in software at the kernel level as they are allocated based on register settings. I don't even think you have access to the FriendlyCore Xenial source to do this. If the OS uses a device tree, you can swap them very easily by simply making changes to the tree. This...
Both I2C0 and I2C1 are available on the GPIO header so you should be able to make the software work but changing pins.

http://wiki.friendlyarm.com/wiki/index.php/NanoPi_Fire3
You have a serial cable to connect to the debug port? That is your best diagnostic tool for headless systems as you can see what is happening during the boot cycle, long before even HDMI output is enabled. Connect to the: Serial Debug Port: 2.54mm pitch 4-pin-header, 3V level, 1500000bps You'll need...
I did a design to use the T3 (same applied to T4) in a mobile application and used an ATTINY414 to handle the ignition detection and shutdown. The ATTiny was powered by the same 5V that powers the T3. Scenario 1. Ignition ON : The ATTiny reads the ignition status and if it detects that the input is ...
Hello again. The T4 GPIO Is easy to use from Android but you do need to ensure that you have user access to it and to do this, you need to add some settings to the .rc file to set the ownership and chmod for access. To do the power up and down via another GPIO Pin is not so easy but the power button...
The RIL is easy if you follow the instructions from SIMCOM. I think Android 8 is the current highest support. https://techship.com/downloads/simcom-3g-and-lte-4g-modules-android-ril-drivers-and-guide/ You'll need to register to download it. Good luck. Let me know how you get on with it. It needs to ...
Any GPIO pin should work. You just have to find the driver for whatever OS you plan to use.
The Ethernet pair is available via one of the GPIO headers but you will need an external magnetic and RJ45 connector to use it.
There is no Ethernet hardware on the NEO PI Air. The reference in the Wiki is likely from a copy and paste.

Wifi only.
I use a SIM7600E-PCIE module to provide a GSM connection and it also comes with a GPS as part of the module. Simcom offer various Android RIL for their modules and a GPS drive that hooks into the OS and instructions on how to build this into the OS. I've used this with 5.1 and 8.1 as they have not r...
To debug you connect to the header DBG UART on the edge of the board. With this you can see the debug output and also be able to type into this various linux commands as if you were using the terminal. Note that the baud rate is 1,500,000 bps so you will need the use Putty to read this. Any 3.3V TTL...
Connect a debug cable to the T4 and see what is showing up in the output. If you don't have one, get one, they are cheap and it will help to diagnose what is wrong. The other option is to create a bootable SD image and hold down the BOOT button so that is boots from the SD card. If this doesn't work...
What else is in that directory?

This is T4 Android 8.1 from Gitlab?

This is the same directory and math is also a directory within this. Does this match up with your directory structure?

https://gitlab.com/friendlyelec/rk3399- ... gle/common
Under my 8.1 source, math is a directory.

What is in the text file?
What is here?

opendir failed: external/guava/guava-gwt/test/com/google/common/math: Not a directory
Yeah, remove the code you added and rename that old file and then do a clean and then rebuild. That will get you back to a working system. By the way, why do you need this SU version? I have found that the T4 build already has SU so you can do root via the debug UART or through ADB SHELL commands. I...
The LED wasn't blinking before the problem started, and I have since changed the OS, SD card, and power, so the common denominator is the firmware, I believe. Is there something in the firmware that might have triggered this? Looking at the schematic, the LED is driven by a GPIO pin. If this is ON ...
The debug serial port is your best place to see what's happening during the bootup.
The M4 I think is the same baud rate as the T4 which is 1,500,000 so use something like Putty to connect to the debug port.

I got caught out with this same thing too.
There are no other options. All of the other headers are outputs only. I actually like the 12V option as it avoids issues with voltage drops with a 5V supply, which I've seen a number of times. With the 12V input connection, it works from 9V to 14V without any issues. In fact, the datasheet shows 5....
Do you have access to a hot air re-work station?

It might be nothing more than a bad connection on one of the BGA balls.and a reflow of the memory chips might work. No guarantee but worth a try if you have access to one.
Any cheap USB to TTL 3.3V interface will work for this. Baud rate is 1500000 so need to use putty for this.

It will be the best investment you can make if playing with embedded Linux boards. You'll be able to track down the issue once you see the debug output.
Can you show us a picture of how this is wired up to the Smart6818?
Do you have a serial debug cable to see what is happening during the boot?

At least then you can find out where in the boot sequence failure is happening.
I added the capital versions and now it compiles without errors.

I've not installed it on my board as I need 8.1 for the time being.

Let me know how it goes for you and if it works on the T4 etc.
Some progress has been made.

Rename the Android.mk file in system/extras/su so that it won't be part of the build.

Now I get an error about the PLATFORM.pk8 file missing. I have not yet made copies of these in upper case.
Yes, there is another PRODUCT_COPY_FILES that creates SU into the xbin folder so we need to track that one down and somehow comment it out.

I've been searching for something to override the copy but can't find anything.
I bought this one and it was detected and I can save to it and read from it. I can't figure out yet how to boot from it but as storage, but it was detected with FriendlyArm Desktop.

NVME M.2 128GB MIDASFORCE LIGHTNING MAX SSD NVMe M.2 PCIe Gen3 x4
I am still trying to track this down. I can't find the product copy file for the existing su into system/xbin.
OK, following your instructions and I get the same failure message so not a setup issue.

Now to find out why.

I did notice that there is already an SU file in system/xbin so wondering if that has anything to do with it?
Thanks for your kind wishes. I am getting better every day and back working from home. I had a look and have not yet attempted to try and build this into my system. I will get around to this during the rest of the week. I wanted to check if the binary that is on the SU git page is compatible with th...
Hi there, I too use the T3 Plus board but for commercial projects. I run Android 5.1 on the current designs but I also have 7.1 in development. It's not quite ready to use yet as issues with getting SELINUX to be in permissive mode on startup. The boards are fitted to a motherboard via Samtec header...
The best place to do this is in u-boot as it will be the first thing to run. You can do this in the kernel but that runs a little later. To guarantee a GPIO is low or high at power up, the only real solution is a pullup or down resistor.
You can use a wakelock to turn on the screen but to turn it off will require system app access using the system keys to build your app. Loads of stuff here about what you are trying to do but the system app is the only real way. https://stackoverflow.com/questions/9561320/android-how-to-turn-screen-...
Which OS?