I’m poking around with the Spider Kit for some time now. It’s a lot of fun so far. I would like to try some “serious” stuff now and that leads to problems.
The first problem I was struggeling with seems to be a problem with power. Every time I enabled the WiFi Module and did a scan or join operation the display collapsed and the Spider rebooted. I was able to work around this by handling the WiFi as the first Module to be initialised and set up. If the GPS Module was enabled before the WiFi everything crashes. However, this is working now by keeping the GPS disabled until the entire networking stuff is up and running.
The next problem is harder to be sorted out. To be honest: I’m stucked here and currently out of ideas.
After the network setup I’m enabling the GPS Module and setting up the events. System stays stable, no more unexpected reboots. So far so good. To test the GPS a little bit I’m simply dumping out received position data. Running this simple project in debug mode (starting the program from within Visual Studio) gives me a running system, stable for hours.
Running the same thing without attached debugging functionality from within VS it starts running but after quite a short while I get “BUFFER OVFLW” on the display. It is still running but becomes completely unresponsive after a while.
I read about this Seed GPS Module almost every post now (to be honest, every post I could find that was related to my problem). I could see that there were many guys out there having problems with this module. But I never found a reply with a solution to this weird behavior.
I’m running the project with .NETMF 4.3, GPS driver is the latest from GHI.
Did someone found a solution? I’m off ides by now and a little bit frustrated here :wall:
@ andre.m - I’m using the USB EDP Module that shipped with the Spider Kit. Tried USB only as well as an external power source (currently the external source delivers 6V and max. 500 mA).
It makes no real difference in general. Further testing today showed me that it is totally random when the system collapses. The only pattern I can see is that it collapses mostly when enabling a device (currently I have only the display, Wifi RS21 and Seeed GPS connected). Sometimes it collapses/reboots when performing a network scan or a join operation. Sometime when the GPS becomes enabled sometimes when the WiFi becomes enabled.
Today (after several spontanous reboots) I was able to run the whole system stable for at least one hour with everything (display TE35, WiFiRS21 and Seed GPS) enabled and normal timer triggered operation (it simply dumps out the last valid position and its last valid timespan data along with the current IP address.
I also did some changes withing the GPS driver (discard the buffer whenever data is received and tried different bitrates of the serial socket, now at 9600) to try to work around these annoying buffer overflows. I’m optimistic that it helps but it needs more testing here.
Overall the spontanous rebooting Spider drives me crazy. It needed somewhat around twenty reboots today to get a running system :wall:
BTW: The Power Estimation function of VS shows me 3.3V at 390mA to max. 400 mA. There’s a line for 5V but without any data for mA?!
I’d suggest you get a 7v minimum external power supply that is capable of at least 1A (1000mA). WiFi uses a lot of power. Screens use a lot of power. GPS is sensitive to power. Spider not that much, but when you add all those hungry/sensitive items, if you don’t have sufficient reserves in those times of high need you’re likely to cause temporary voltage collapse and you’ll get resets and other strange behaviours (like you’re seeing).
7V is important so that you can regulate down to 5v without causing any issues - the voltage drop of the conversion process could be causing your 6v supply to give you something lower than 5v. Even though the EDP does say 6v, I’d always err on the side of 7v or 7.5v.
Thanks for Your input on this. My external power source is capable to deliver 7.5V also but it’s limited to a maximum of 500mA. I guess You’re right and I need another power source to get a stable system here. I couldn’t find any related specifications for this.
For develpment purposes I will look for another external power source.
When the system is fully blown (somewhat in the future) with WiFi, GPS, compass, motor drivers, camera and display… how would You handle the power demand for this? In “Lab” and espacially autonomous out in the “wild”?