First Impressions

[Mitch]

I received the board a few days ago and am very impressed with Tom's attention to the details. It's a quality product and the size of the board is much smaller than I expected. I particulary like the thin profile.

I powered it up yesterday for the first time using the R13 firmware and software. The configuration tool has a few bugs (maybe visual studio bugs!) but it does work. I haven't done anything with the bootloader yet.

Using the configuration utility and the rtos_pilot firmware, I calibrated the sensors and had the cube displayed and stable within a few minutes.

The GPS interface in this version of the configuration utility appears to be a work in progress. Couldn't get the baud rates set - it looks like the "operational baud rate" is hard coded to 115200 in the rtos firmware. I hard coded the startup and operating baud rates in the firmware to 115200, and configured my gps the same. Works great with the gps_test example code but not in the rtos_pilot firmware.

I also couldn't get ppm input to function from the configuration utility with the rtos_pilot code. Again, it works extremely well standalone using the ppm_in_test code.

[Tom]

Hi Mitch,

Thanks for this short review.

Yes, I only did some quick tests with the GPS. It works for my unit (Locosys from sparkfun), but I need to do some more tests first. The initial_baudrate should work, but changing the GPS to work on another baudrate doesn't work flawlessly with different kinds of GPS units.

I wonder why the ppm isn't working. You can't see the "progress bars" in the configuration tool moving?

Enable line 57 & 58 in http://code.google.com/p/gluonpilot/sou ... /control.c for servo control using your transmitter.

I need to fix the case where the transmitter doesn't have 8 channels. (see http://code.google.com/p/gluonpilot/sou ... n/ppm_in.c line 17). Do you get a happy smiley in ppm_in_test?

Tom

[Mitch]

I missed the second line with the ";" the first time around. I have 8 channels on the ppm stream. All eight are measured with smily faces with ppm_in_test.

1512 1504 1921 1512 2072 1516 1508 1515
1512 1500 1923 1513 2068 1516 1510 1516
1513 1504 1920 1513 2072 1516 1508 1515

It wasn't finding valid frames for some reason and locking here

Gluonpilot v0.1
Config: 220 bytes
Loading configurationdone
Control task initializing...Waiting for PPM frame...

Interesting find!!! I was using a Futaba 9C transmitter which transmits 8 channels in PPM mode. I checked the PPM trace on my oscilloscope, everything looks OK and the ppm_in_test works fine.

Switched to a Futaba 7C transmitter in ppm mode and the rtos_pilot firmware is working fine with the config tool!!!

edited NUM_CHANNELS = 9 in ppm_in.c, now the 9C is also functioning!!!

Only channels 1-6 are being sent with telemetry.

I'll have to investigate with the scope to see what the difference is.

[Tom]

So I will try to change the fixed number of channels.

It would be great if you could debug it a little. I only have 2 multiplex transmitters, and they both work fine (and both 8 channels). I'll PM you when I have some new PPM code!

[Mitch]

If you could reliably find the channel count from the ppm data. Alternately could make the number of channels a configuration option. I'll sort out the differences with the radios I have and let you know. I'm in the gradual process of switching everything to 2.4ghz, and will switch this as soon as I find a shared spectrum receiver that I can tap with all of the channels in one place on a ppm stream.

I spent some time playing with the gps. Here are a few comments to maybe help in the first revision.

I'm using a Locosys LS20031. I hard coded the uart2 baud rate to 115200 and configured the gps module for 115200 baud, 5 hz, GGA and RMC sentences both at 1/position.

The gps_test example code performed flawlessly. Under rtos_pilot GGA sentences were received (sat count functioning) but the RMC sentences were not. I put a few fprints in gps.c to double check this.

I reduced the GGA sentence rate to 1/5 positions and both GGA and RMC functioned. Possibly don't even need the GGA sentence data with the baro altitude available. I wanted to keep the RMC sentence rate as high as possible.

Found a minor discrepancy in the firmware communication_csv.c, in the GPS part, sending sensor_data.gps.heading_rad*100, and received as Heading_deg in configuration tool code. Also, not sending GPS altitude.

I added the last three lines below to the GPS tab of ConfigurationFrame.cs. May as well display the data if we have it!

_tb_gps_numsat.Text = gb.NumberOfSatellites.ToString();
_tb_gps_latitude.Text = gb.Latitude.ToString();
_tb_gps_longitude.Text = gb.Longitude.ToString();
//_tb_gps_height.Text = gb.Height_m.ToString();
_tb_gps_speed.Text = gb.Speed_ms.ToString();
_tb_gps_heading.Text = gb.Heading_deg.ToString();

[Tom]

Hi Mitch,

I uploaded new code to the SVN: The GPS should work better now. I also included an "auto detect" of the number of PPM channels. Could you test it? (It will wait for 2 seconds for PPM pulses, after that it will timeout)

Thanks!

[Mitch]

I loaded the r19 firmware and software exactly as downloaded via svn. Built and installed the rtos pilot firmware. The gps is functioning perfectly. PPM is not functioning with either of my transmitters.

ppm_in_test is functioning with both futaba 9c and 7c transmitters. I'll pm more trouble shooting info.