uBlox and Copernicus GPS issues

[Admin]

A continuation from this thread:
http://www.societyofrobots.com/robotforum/index.php?topic=9190.msg93760#msg93760

Lets start with the Copernicus . . . my connections are:
R2, BOOT, XRST, VCC, XSTBY: 3.3V
GND: ground
TX-B: serial data output (by default this is NMEA at 4800 baud)

It's just giving me the GPS coordinates of Sparkfun non-stop for an hour, and won't lock on to a single sat. I'm using their recommended ceramic antenna, I'm in a tall building, it's a sunny day, near a window. Other GPS units can get a lock.

Am I missing a connection?


-------------
As for the uBlox (with DIYdrones adapter) . . . there was a temporary freak power failure on my mcu setup, and even though the GPS should have been protected, maybe something weird happened that caused it to fry . . .

Anyway, it no longer outputs any serial data (at any baud rate), but voltages check out fine. Is it fried? Any other ideas on how to debug it?
(I'll go post over at DIYdrones in a bit)

[madsci1016]

The ublox adapter connector is extremely flaky. It took several tries in gluing the boards together to get a reliable connection.

That's my only suggestion.

Freak power failure? Or high voltage spike?

[Admin]

Freak power failure? Or high voltage spike?

Good question. I let it run overnight as an experiment for a project I'm working on. The project (mcu, GPS, plus other stuff) was connected to my laptop via USB, and got power from USB via a 3.3V regulator. Other stuff only connected by I/O pins and ground got power from a 2600mAh 3.7V lithium battery (several mA of current required).

In the morning I woke up to see my laptop turned off (no idea how that happened, it's never happened before). The lithium battery (which has a protection circuit built in) also measures 0V and no longer works, although I haven't tried force charging it yet. And the GPS now doesn't work. Nothing else was damaged to my knowledge.

How'd that affect the GPS I have no idea.

The ublox adapter connector is extremely flaky. It took several tries in gluing the boards together to get a reliable connection.

Nah, I had it firmly secured with hotglue. And to verify, I tried wiggling it around but still no luck. I checked the battery cell on it, and got 3.09V (which is fine, assuming it is a 3V lithium cell).

[Admin]

The uBlox is magically working again. yay!

[Admin]

I let the Copernicus sit there for ~4-5 hours, and it finally got a sat lock of 4 sats.

Well, I guess that means the problems have been 'solved' lol . . .

[madsci1016]

The lithium battery (which has a protection circuit built in) also measures 0V and no longer works, although I haven't tried force charging it yet

BTW, Proper Lithium protection circuits will disconnect the load when the voltage drops below a threshold (dies) and won't give a voltage reading till it's recharged. IE, it doesn't need a 'forced charge', it needs a regular one. It's acting normal.

[Admin]

Cool . . . as soon as I hooked up my [homemade] charger to it, the voltage immediately jumped to 3.3V. Charging now . . .

[madsci1016]

Cool . . . as soon as I hooked up my [homemade] charger to it, the voltage immediately jumped to 3.3V. Charging now . . .

Yup, just as it should.

Protection boards pretty much make Li-Ion batteries idiot proof. Can't over charge, over discharge, over current, etc.

Also, if you ever (hopefully by accident) try to draw to much out of a protected battery, it will 'trip' and disconnect the load. It won't reset till it sees an open circuit at the load terminals, meaning you have to turn your robot off and maybe disconnect the batteries to get the battery to work again. I've seen people (unaware of this fact) driven crazy trying to figure out why their robot keeps losing power and won't turn back on till they cycle the battery connection.

I've also had a few protected batteries (fully charged) not reset from an over load till I plugged them into a charger for a few seconds.