2010
05.18

So I’m building a GPS disciplined 10Mhz oscillator to drive all of the test equipment in the office. Doing so will give me some killer accuracy for my frequency generators, radios, spectrum analyzers, and the like. There are a few dozen different designs out there, mostly centering around an HP OXCO oscillator.

And of course, like everything else the controllers all run on +5, and maybe +12. The oscillator runs on +12. The oven on the other hand needs +24v.

And every power supply I have in the inventory is a switcher with +5vdc at a bunch of amps, +/-12vdc at a few amps, and -5 if you’re lucky. +24? Who needs that? Grr.

But about 3am last night, my long thought forgotten switching power supply theory resurfaced so I wandered into the lab and confirmed it.

Many switching supplies generate the plus and minus voltages independently. If you look at the circuits closely you will find the ‘grounds’ tied together someplace close to the output, if there is a common ground at all. Other supplies provide a separate ‘return’ or ground line for each output. If you find a wire tying the returns together pull it. Check the voltage between the returns. If it is floating, you may be able simply tie the -12 line to the +12 and get +24VDC at the output formerly known as ‘-12vdc return’

Your mileage of course, may vary, you may blow up a supply, or you may have a new output at the cost of one of the old ones.

2010
05.18

I’ve been looking for GPS modules for several projects, a GPS disciplined time source, APRS for the street and racecars.

Ebay and a number of other sources turn up lots of GPS modules from various manufacturers. Most of the projects I’m working on seem to play well with the Rockwell Jupiter modules which are supposedly easy to find and are definitely well documented. Unfortunately, many of the vendors of these Rockwell GPS modules don’t or can’t differentiate between the Jupiter modules and the MicroTracker modules. If you buy one for these from Weirdstuff, for example, the docs don’t match the module you’ll get, which is a TU00-D200-401.

The TU00-D200-401 lacks the 10khz output of the Jupiter modules, has a 0.10″ spaced header vs the 2mm spacing, and has a very different pin out.

While the Jupiter is well documented at a number of locations, most notably, http://eshop1.chem.buffalo.edu/600M-GPS.html this site has a photo of a TU00-D200-401 at the top but is actually about the TU30-D140.

If you’d like to use the TU00-D200, the pin out is:

Pin 1: Antenna power input. Normally 3-5 volts. Maximum 12v, 100mA.
Pin 2: GPS unit power input. 5.0 volts DC. Draws approximately 175mA.
Pin 4: NMEA/Binary mode select. +5v for NMEA. Ov or floating for binary mode.
Pin 5: Battery backup for RTC only (internal clock). 2.5-5 volts.
Pin 6: Serial Data Out. Output of GPS messages. 0/5v level RS-232 signal.
Pin 7: Time Mark Out. IHz (1PPS) timing output.
Pin 8: CTS/DTR Serial Data control. Ov or floating for normal. 5v stops serial output.
Pin 9: Serial Data In. Commands or data sent into GPS here. 0/5v level RS-232.
Pin 10: Battery backup for Sram and RTC. 2.5-5 volts.
Pin 13: Master Reset. +5v to turn on GPS. Ov or floating to turn off and reset.
Pin 18: Serial Data In from DGPS correction receiver. 0/5v level RS-232
Pin 20: Baud Rate select. Ov for 4800 baud. 5v or floating for 9600 baud.

Pins 3, 11, 14, 17, 19 all Grounded

Pins 12, 15, 16 No Connection

So, to hook one up to a TinkTrak3, for example, connect as follows:

TU00-D200:
Connect pins 1,2,4,13 together -> TinyTrak3+ pin 4
Connect pins 3, 19 together -> TinyTrak3+ pin 5
Pin 6-> TinyTrak3+ pin 2
Pin 9 -> TinyTrak3+ pin 3

Connect pins 20,17,11,14 together

Plug it in, turn it on, wait 15 min and you should have GPS lock.

If I get time I’ll put together a pcb to connect one to the other without having to resort to an ugly cable like mine.

Note: Beware. When using this board with some GPS monitoring tools (VisualGPS)  it appears to be spitting out binary. Disconnect and reconnect the board (in software) and it should sync up in a few sec.