GPS Clock with Strange Power Problem

For other supported Arduino products from Adafruit: Shields, accessories, etc.

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

GPS Clock with Strange Power Problem

Post by rcomito »

I've built a GPS clock using these great products:
Arduino Micro http://www.adafruit.com/products/1086
Adafruit Ultimate GPS module http://www.adafruit.com/products/746
Two 4 x 7 segment displays with I2C backpacks:
1.2" http://www.adafruit.com/products/1270
.56" http://www.adafruit.com/products/865

The large display shows hours and minutes. The smaller display shows seconds and
the number of satellites in view with the decimal point between used to indicate if
I have a fix or not.

The clock works great as long as it's powered by the USB port on my computer. If I plug
it into the wall. I lose the satellite count. The fix indicator remains lit.

I tried powering it with a 5v 1a USB power supply http://www.adafruit.com/products/501
and the satellite count dropped to 0. Next I tried powering it with a 12v 2a wall wort to
VIN and ground and experienced the same thing.

I figured, with the VIN as power, I could use the serial monitor to do some debugging, but
as soon as I plugged back in to my USB port, the satellite count appeared.

I'm confused.

Here's the display powered by the USB port:
Clock00.jpg
Clock00.jpg (116.11 KiB) Viewed 579 times
Here's the display powered externally:
Clock01.jpg
Clock01.jpg (109.83 KiB) Viewed 579 times
I've been chasing this off and on for a week.
Has anyone experienced this?

Thanks for your help.

Rick Comito

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Whoops. I posted this in "Other Arduino products" by mistake.

I believe this belongs in "Other Adafruit products".

Moderators, feel free to move it, if you like.

My apologies.

Rick Comito

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

UPDATE:

So I've been thinking about this a lot. Plugged into the computer's USB port everything works as
expected. On external power (5v to USB or 12v to VIN), I see my fix indicator but no satellite
count.

The boolean GPS.fix is set while parsing the $GPRMC sentences. The GPS.satellites is set from
the $GPGGA sentences. Maybe I'm not getting GGA while on external power.

To prove this I set a debug flag that gives me the sentence counts % 100 on the small display;
RMC on the first two segments and GGA on the second two segments.

As expected, powered by the USB port, both numbers update furiously. Plugged in to external
power, the GGA count stays at 00. No GGA, no satellite count. I let it run like this for an hour
and the GGA count never incremented.

I've been powering the Ultimate GPS module with 5v. Changing that to 3.3v didn't change
anything. I changed it back.

No solution yet. Just things that make you go "Hmmmm".

Rick Comito

User avatar
adafruit_support_mike
 
Posts: 67454
Joined: Thu Feb 11, 2010 2:51 pm

Re: GPS Clock with Strange Power Problem

Post by adafruit_support_mike »

The consensus opinion among the tech staff here is "WTF"?

We don't doubt your observations at all, but we have absolutely no idea why a GPS module could even notice the difference in power sources. The only variables that should change are:

- supply voltage
- supply current
- rail voltage under load
- transient behavior of the voltage rail (ripple, noise, etc).

Try feeding it from a battery pack and see what happens. At this point, we aren't ruling out "breaks into song and dance".

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Thanks for the reply Mike. I share your opinion.

I like the idea of a trial with a battery pack. I'll give that a try and let you know.

This build is still bread boarded. I think I'll solder it up on a proto board (after the
battery pack trial) and see if that makes a difference.

I also just ordered another set of components for this build. This clock is going
in my living room. I'm building one for my shop. It will be interesting to see if
a new GPS module acts the same.

Rick Comito

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Hi Mike -

I tried powering the clock with a 9v cell on VIN as you suggested. No luck.

Once again, as soon as I plug into the PC USB port, lots of GGA sentences
and the satellite count displays.

Hmmmmm.

My next step is to solder this up, but I don't expect anything to change.

I'll keep you posted.

Rick Comito

User avatar
adafruit_support_mike
 
Posts: 67454
Joined: Thu Feb 11, 2010 2:51 pm

Re: GPS Clock with Strange Power Problem

Post by adafruit_support_mike »

Your mention of a breadboard caught my attention.. electronics guru Bob Pease used to call them "white blocks of pain" because they're notorious for adding mystery components to a circuit. I once measured a 1.5v voltage drop from one end of a jumper to the other because of unexpected resistance in the clips.

First off, give everything a good "wiggle the wires" test and see if that does any good. Then hit the non-USB power path with a multimeter and see if anything funky is happening there. Use the full spread of dial settings.. resistance and capacitance along the path when the power is disconnected, voltage and current when power is connected. Situations like this are why I have multiple meters.. sometimes it's handy to measure more than one thing at the same time.

If you have an oscilloscope, bring that into play.. a scope is a significant investment, but boy-howdy can it make problems easier to see.

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Hi Mike -

I'm soldering up a proto board. I haven't soldered in the Arduino
Micro or the GPS module yet. I'm waiting for my order from Adafruit
so I can try the new GPS in the breadboard.

I like the idea for measurements. I'll definitely do that. I'll try it with
all of my power options: PC usb; wall wort to usb; wall wort to VIN;
and battery to VIN.

Thanks for the reply.

Rick Comito

danieldd
 
Posts: 18
Joined: Sun Mar 23, 2014 7:19 am

Re: GPS Clock with Strange Power Problem

Post by danieldd »

Hey Rick,

I had many problems with the power supply of the Ultimate GPS breakout too. Finally I put a low pass filter (10 Ohms resistor, 47nF cap) into the power supply for the GPS module and installed the module´s backup battery (this makes it obviously more robust). And now it´s working, fast and reliable.
Maybe you should try a low pass filter as well...

Daniel

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Thanks Daniel - That's a great idea. I'm on it.

Rick Comito

User avatar
adafruit_support_mike
 
Posts: 67454
Joined: Thu Feb 11, 2010 2:51 pm

Re: GPS Clock with Strange Power Problem

Post by adafruit_support_mike »

Filtering the supply power *is* a good idea. Thank you Daniel.

FWIW, you might try a 'bootstrapped' filter:
bootstrap.jpg
bootstrap.jpg (19.78 KiB) Viewed 473 times
The capacitor absorbs most of the spikes that come from Vin, so the voltage across the resistor on the right stays more or less constant. A resistor with a nearly constant voltage across it makes a simple current source.

danieldd
 
Posts: 18
Joined: Sun Mar 23, 2014 7:19 am

Re: GPS Clock with Strange Power Problem

Post by danieldd »

you should try with different resistors. The 10R was the only one I could use, the next higher value I had was 100R which was to high (the GPS did not get enough voltage with this resistor). Also a smaller cap could be good to filter high frequency noise. It´s maybe a good idea to play around with several values.

Daniel

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Thank you gentlemen.

I will definitely work with these suggestions.


Rick Comito

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

Another update:

o I tried both filter types, individually and together with all three of my external sources. No change.

o I neglected to complete my Adafruit order, so the second GPS module was still in my cart. The order should be here today.

o I let the clock run on external power for 24 hours with my debugging routine that displays the GGA % 100 count in the right most two digits. After all the time, the count was at 4. This tells me that the GGA sentences aren't dropped entirely. But what happened to the rest of them.

o I'm still on the breadboard. I hesitate to solder in the Micro and GPS while trying to debug. I've been using breadboards for more than 30 years without having one introduce a bug. But I'm definitely open to the possibility.

Stay tuned.

Rick Comito
Last edited by rcomito on Sat Apr 12, 2014 8:46 am, edited 1 time in total.

User avatar
rcomito
 
Posts: 35
Joined: Tue Jan 28, 2014 9:03 am

Re: GPS Clock with Strange Power Problem

Post by rcomito »

The hardware for the second clock is in.

The new GPS module acts just like the original. (Drat!)

I'm going to solder up the new clock and see if miracles really do happen.
Wish me luck.

I'm stupefied (credit: Al Capp. not Disturbed)

Rick Comito

Locked
Please be positive and constructive with your questions and comments.

Return to “Other Arduino products from Adafruit”