Boot problem with RLP code

Hi all,

i’m pretty new to .net gadgeteer and i have build a robot for my master thesis. Because my code was way to slow and i had some c code experiences i tried RLP for a part of my program. When i start my robot with visual studio and my usb cable in my pc everything goes okey. But when i unplug my robot from my laptop en starts hem from his external battery hey won’t go through the RLP code. I did some debugging and it looks like he wont load the elf file. Could it be true that the elf file is loaded from my pc everytime i starts the robot from visual studio? More important Is it possible to place the elf file on the motherboard itself?

I have a fez spider with a camera, oled display, wifi and a motor controller and a USB dp client, all working good but if the RLP code doesn’t run he will not start doing anything.

grtz Carlo

You’re elf is deployed to the device’s internal EEPROM as a resource whenever you program it, so that won’t be your problem.

You mention that you’re running a robot that’s fine on the PC but not on battery. Are you sure you have enough power on the battery?

You might use the OLED as a debug screen to show you the progress of your code in battery mode to see what’s happening. Perhaps you’re losing too much power as you attempt to control the robot.

Also, there should be different batteries for the motor and the processor board.

2 Likes

@ Skewworks:
I will follow your way of thinking that the elf file is stored in the EEPROM but what is strange is that when I load the program in the robot from usb also with the power already on the program starts good and it stays running when I unplug the USB from the USB dp client, the robot drives like it need to be. So I don’t think it is a power problem, correct me if I’m wrong. When I power of the robot and power on it again it just don’t get over the elf file loading I guess. I did already some debugging with testing to print a circle on the screen before loading the elf file normally after drawing the circle the elf file needs to be loaded and than the program starts with saying the program is started. But he just print the circle and that’s it. I will do some more debugging when I got some more time. Maybe tomorrow.

@ Mike:
Thats true but I’m in a debugging phase so it need to be simple and not have to much overhead. Do you think that will solve the problem? I will test it with an extra power source like a DC adapter or something.

thanks for your fast responses

@ carlo.hutsebaut - a motor creates lots of noise on the power line and often will cause problems with CPU.

I updated my fez spider to the latest firmware and right now the problem is gone, I cant explain why but I will suggest others that have the same problem to update to the latest firmware.