It could be EMC related. Spark ignition are the primary source of problems.
Try filtering every signal connected to your board. Diodes, capacitors, etc...
From the protocol and your code, it seems you send a read status request (0x30) to both devices.
The first device send its status at the same time you send the request to the second device. So there is collision on the bus and the second device...
What append if you send only to the second actuator. Comment the first Serial3.write.
And what about 120ohms termination resistors, twisted pair cable,... ????
No need of Delay or flush. Use Serial1.transmitterEnable(pin) so the driver will take care of switching on and off the DE pin of the RS485 driver.
More details here https://www.pjrc.com/teensy/td_uart.html
On STM32, when noise is critical on an analog input, all the other pins in the same port should not be used for digital signals.
Could be similar on the NXP.
Is the DSP activity using all the processing power of the first T4.1 ????
What is the signal to be treated ????
What kind of display do you plan to use ???
Unless the displays is large and complex and/or the signal is fast, 600MHz, should be...
The LCD is powered by the 5V. So its 4 data lines will get 3.3 volts signals in write mode, but put 5V signals in read mode
edit: correct, the R/W line is on GND, so LCD data lines are inputs.
What is the serial protocol on the DTV ? Is it binary ? Text ? command-response, either binary or text ???
It will probably accept commands ending by a carriage return. And the respond with datas.
In you example, you don't send any carriage return.
Back in 2014 I started a thread about my Burning Man project and claimed that I'd post pictures of the project, but never ever did, and now the thread is closed for reasons.
But anyway, Teensy and OctoWS2811 have been serving us well for 10...
In fig 1, the the LED and resistors are wrongly connected to the photo-transistor. In this schematic, the led could not emmit light.
But you probably soldered your prototype the right way.
The circuit in figure 2 is correct. But you might reconsider choosing PC817. It may be on the slow side for 31250 baud. Normally H11L1 is used. This PC817 part is much slower.
May be a combination of "non-audio" DAC (on I2C or SPI) for the amplitude signal, and digital output. All connected to an op-amp.
Or a DAC to drive a variable gain amplifier.
Why using 5V on collector of Q1 ? Use 3.3V and you would be sure your Teensy will not fry.
What the use of VB ?? Only for measurement ? Or it is the inverted VO ?? If you need it as a digital output, you need to insert a resistor on the Q2 base.
Here's a quick attempt to answer all the questions.
From a purely data processing point of view, just one Teensy 4.1 ought to be capable of controlling 9000 LEDs.
But due to the electrical challenges of so much power, already mentioned quite...
Ignition systems for gas engines are the best EMI generators. Both RF and conducted.
AndyA's advice is spot on. Put anything away of the spark plug cable. Filter the power supply and I/O's with transil diodes, ceramic capacitors, and so on.
Thanks for the answer.
Probably I don't need 600MHz to spy data exchange between two MC6803 running at 4MHz.
So, no need to adapt anything when changing main clock.
As I have not access to my PC to check by myself, so I ask here first.
I need to receive and send data to a very old piece of hardware through serial. It seems to be 62500 baud from the reverse engineering of MC6803 assembly code.
Will the 4.x...
You can actually select which clock speed that should be used for running the Teensy 4.x
and at least in theory minimize the power usage (have not done any tests myself).
The clock can be as low as 24MHz.
Small quantities batches means higher unit price. It is not simply the components price. When 3.2 was in production in 1000's unit, or even more, batches, I am sure the uC was not bought at 7$.
At 100 or 200 units batch, we will see comments...
For the railway industry, I use a "breathing" device. It is like an M12 bolt, with integrated Goretex membrane. It let the ROSE enclosure breathe and expell humidity due to temperature changes.
And the pcb is also coated.
megax,
in your bug report, it seems you set the clock at 60MHz.
Read this post https://forum.pjrc.com/index.php?threads/flexcan_t4-flexcan-for-teensy-4.56035/page-48#post-339909
As usual when having problems on CAN:
twisted pair cable ?
120 ohms resistors at both ends ?
3.3V transceiver ?
good power supply decoupling on the transceiver ?