GPS App Power

Today Samad and I are planning to do power measurements for the GPS application board (DE3) when running.  But first I have to get the app up and running again - it's been a while.

I am going to temporarily detach the DE3 power supply from my earlier jury-rigged setup that tapped out the +5V supply.  After I verify that the DE3 app is still running, we'll patch in some external supplies to the connector.

I'll also plug the Wi-Fi board back into the USB supply for now.

Hooked up the OCXO via the 6" SMA cable, with a jury-rigged connection between the SMA connector on the OCXO side and the test point for the clock output.  We need +3.3V supply for the OCXO; can we get this from the DE3 board?  Pin 29 on each GPIO header is the +3.3V supply, opposite pin 30 GND.  We can rig wires from there to the OCXO board.

At some point, I should maybe break out a ribbon cable from GPIO1 to make a proper connector.

Inserted serial cable 3 between my wire-wrapped DE9 connector on GPIO1, and the plug-wired connector on the Wi-Fi board.

Plugged in the power connector and the USB-blaster cable to the DE3 board.

Before adding the GPS module into the mix, let's verify Wi-Fi communications.

Starting up the server.  Powering on the Wi-Fi board.  TikiTerm windows pop up.

Now, let's power on the DE3 board.  I think it will just start up with the demo?  No, nothing...

Let's start UwTerminal first, so we can see any I/O on Wi-Fi stdout...

Weird, the Wi-Fi board restarted by itself for some reason...

With UwTerminal running (at 57,600 baud), this time when powering up the board the demo started up - but the WiFi board reset itself again!  Weird...

Let's load the GPS app design.  First, have to locate the latest version... 

The version in C:\f\DE3\S3\SB+SOPC\GPS_FPGA_app\Quartus_II_Project\DE3_GPSapp was last touched circa 5/17-5/24.  Whereas, the one on Dropbox (C:\Users\Mike\Documents\My Dropbox\FAMU\COSMICi\GPS_FPGA_app\Quartus_II_Project\DE3_GPSapp) was last touched 1/19.  So the one in \f is newer.

Let's look back at my notes from 5/17-5/24...  Nothing really helpful there.  Probably whatever I touched then was not essential.

The other day, I copied the GPSapp on \f to C:\LOCAL\GPS_FPGA_app, to make it easier to navigate to.  Later I'll put it in C:\SHARED to share it with the students.

Opening the project, C:\LOCAL\GPS_FPGA_app\Quartus_II_Project\DE3_GPSapp\DE3_GPSapp.qpf, I'm finding there are two revisions, Rev_A and Rev_B.  I don't know what the difference was.  We're on Rev_A currently.  Rev_B was created May 16th.  Let's look back at the blog...  Still can't figure it out.

Let's try Rev_B first, since that is the newest revision.  Wait, no programming file??  Recompiling...  OK, now we have a Rev_B.sof file.  Selecting USB-Blaster.

Programming... No stdout yet, but I see the "Hi" LED message.  OK, haven't hooked up the GPS...

Let's open the firmware, it was in the legacy Nios II IDE...  Running code in debugger...  Great, now the IDE seems to have hung...  Before even starting the Nios terminal...  Killing it & restarting.

The hanging problem might be related to moving the project folder.  Reopened original workspace (C:\f\DE3\S3\SB+SOPC\GPS_FPGA_app\Nios2IDE_Workspace) and ran from there and it works, sort of - still not seeing any output on the UART bridge.  However, after some fiddling, I did get the diagnostic output re: PPS and NMEA data from the GPS to appear on stdout.  It is flaky though - the DE3 board keeps resetting itself, and the Wi-Fi board resets itself whenever the DE3 board powers up.  At one point, I couldn't even get the .sof file to load, but fiddling with the wire-wrap spaghetti on GPIO1 fixed that problem, so there must have been a sporadic short somewhere in that mess o' wires.

After Samad got here, we also noticed some actual disconnected wires in the GPIO1 wire-wrap rat's nest, including a broken Tx/Rx wire, which would account for the lack of Wi-Fi data.  We decided to replace the wire-wrap garbage with a proper (40-pin) ribbon cable so it is less prone to breakage.  I brought in a cable assembly earlier, left over from some old IDE drive.  We removed the wire-wrap from GPIO1, cut the ribbon cable, and identified, separated, stripped, and color-coded the appropriate wires of the cable, and identified which pins on the back of a male DE9 connector they need to be soldered to.

Samad is coming back at 2 pm tomorrow and we will actually do the soldering, then connect the cable to the Wi-Fi board, and hopefully see data output from the GPS app again.