USO just stops working?

I started having problems with the USO and keeping it sending data.

I have tried two different USB-Serial adapters, both with the same results. I start sending data, Profiler is running, I see data transmit LED on the adapter flash - no problems.

As soon as I apply 12V to the circuit, profiler stops like I pressed the stop button, sometimes it sends data for a bit, then stop. If I try to Start again, Profiler will show it is running with the big red Stop button, but there is no data being sent. I have to unplug it from the USB, and only this time I get the error: Error writing on serial port COM4, even when I have the profiler stopped. It seems like it thinks data is still being transmitted, but there is nothing coming out.

I am monitoring the adapter in Device Manager also and it only disconnects when I unplug it. It seems like a hardware issue, but connection is not lost to the adapter and I know they work properly.


Any advice or methods to try would be appreciated.

I just noticed when the transmission stops, the USO engine output drops from 100 (10ms delay) to 28-30. This is while the profiler is stopped and start button is green. If I start it again, USO engine output remains the same, but I notice the Synaptrix output raise from 0 to 100.

Now I have tried with 2 adapters, 2 circuits (proto board and soldered), and no change. The transmission cuts off usually when I connect the 12V supply, but sometimes works for a bit.

If its FTDI based adapter, try enabling the Serial Printer option...

Also be sure you have common grounds tied toggether!

I hope that helps

Thanos

Thanks for the tip, but it doesn't matter if I click the option or not. Of course grounds are connected..

I suspect some sort of power supply issue, but I don't know what else I could do. The FTDI modules run on their own isolated power supply from the USB port, the rest of the circuit is connected to 12V and a regulated 5V from it. The 12V supply doesn't even go on the microcontroller's board, only the regulated 5V. Everything is decoupled properly and all grounds common.

For some reason communication is lost and sometimes I notice in the Device Manager that the COM port disappears. This only happens when I apply 12V, or put load on the circuit.

This sounds somewhat like a windows delay and resulting hardware timeout. To prove the theory, you could set the FTDI timeout setting to the highest possible value and see if it helps. If it does, then we can start down the path of improving Windows / Profiler performance.

In a few rare cases, I saw similar behavior on extremely low memory single processor machines.

It did help with the FT232R based module a little bit, but the problem still exists. I tried it up to 5000ms time out. If 5 seconds is not helping, I don't think 10 seconds will be any better.

I will try it from a 12V battery instead of a supply. Maybe it will help..

I was right. With a 12V battery I get no disconnections, everything works reliably.