Clock stopped after a few days. Crystal?

For RTC breakouts, etc., use the Other Products from Adafruit forum

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
catspoke
 
Posts: 2
Joined: Wed Feb 06, 2013 10:50 pm

Clock stopped after a few days. Crystal?

Post by catspoke »

Ice cube clock worked great for 5 days. Then it got stuck at 5:47:23. Unplugged, plugged it back in, can do the menus, but in never shows the clock anymore, so it appears to be a crystal issue, but crystals rarely just fail like that????? Should I order another crystal or what?

catspoke
 
Posts: 2
Joined: Wed Feb 06, 2013 10:50 pm

Re: Clock stopped after a few days. Crystal?

Post by catspoke »

figured it was c8 and c9 rotated 90 degrees. Evidently it can work for a while with these wrong.

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

same problem here, if you don't mind could you please send me a picture of ur board with the crystal and capacitor

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: Clock stopped after a few days. Crystal?

Post by adafruit_support_mike »

For reference, it's very unusual for a crystal to fail. They drift a bit as they age, and you can get them to oscillate at wierd overtone frequencies if you give them too much power, but having one just stop working is about as rare as having a resistor drop to a short circuit or open circuit.

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

i might think it's the capacitor, the label on it were 20J instead of 20

User avatar
adafruit_support_bill
 
Posts: 88086
Joined: Sat Feb 07, 2009 10:11 am

Re: Clock stopped after a few days. Crystal?

Post by adafruit_support_bill »

@BANNED - Post photos of the front and back of your board.

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

Back
Back
Optimized-DSC_0011.JPG (448.37 KiB) Viewed 1164 times
Front
Front
Optimized-DSC_0009.JPG (379.83 KiB) Viewed 1164 times

User avatar
adafruit_support_bill
 
Posts: 88086
Joined: Sat Feb 07, 2009 10:11 am

Re: Clock stopped after a few days. Crystal?

Post by adafruit_support_bill »

@BANNED - Looks like you have the right capacitors (the 'J' suffix is just a 5% tolerance code). And it looks like they are installed correctly. Did it work for a while then stop as in the original post? Or was it not working from the start?

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

Hi Bill, it was working fine for couple of minutes, then the time hang but all the other button functions, when try to set time, it will just get stuck on the date after i finish

User avatar
adafruit_support_bill
 
Posts: 88086
Joined: Sat Feb 07, 2009 10:11 am

Re: Clock stopped after a few days. Crystal?

Post by adafruit_support_bill »

Can you post a closer shot of the solder-side of the board - focusing on the general area of the oscillator circuit?

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

back
back
Optimized-DSC_0015(2).JPG (638.88 KiB) Viewed 1120 times
Optimized-DSC_0016.JPG
Optimized-DSC_0016.JPG (553.22 KiB) Viewed 1120 times

User avatar
jarchie
 
Posts: 615
Joined: Sun Jun 24, 2012 2:16 pm

Re: Clock stopped after a few days. Crystal?

Post by jarchie »

For what it's worth, I've built four Ice Tube Clock kits (three were gifts) and had issues with the quartz oscillator twice.

In the first case, the oscillator did not work at all, and the time would not display. The problem seemed to be burnt-on flux, since everything worked fine after thoroughly cleaning the board. I suspect the burnt flux was causing excessive stray capacitance. Another thread contains some good advice for cleaning boards.

In the second case, the oscillator quit working after about a month of use. Cleaning the board did not help, but the 20 pF caps included with the kit seemed too big to me. So I replaced them with 10 pF caps and everything worked fine. Here's why I think switching out the caps worked, but I'm hoping Adafruit support will check my reasoning.

Although the ATMEGA168V datasheet does not specify pin capacitance for Xtal1/Tosc1 and Xtal2/Tosc2, the ATMEGA168A datasheet does. Since the recommended capacitor values for the low power and full swing oscillators are identical between the two datasheets, I suspect the pin capacitance would also be identical. The crystal is designed for a 12.5 pF load. Assuming a stray capacitance of 2 pF and the ATMEGA168A pin capacitances of 8 pF and 18 pF, the load with two 20 pF caps is a bit high:

(8 + 20) * (18 + 20) / (8 + 20 + 18 + 20) + 2 = 18.1 pF

(I'm using formulas from Quartz Crystal Desing Parameters from ECS International.)

But the load with 10 pF caps is closer to the ideal 12.5 pF:

(8 + 10) * (18 + 10) / (8 + 10 + 18 + 10) + 2 = 13.0 pF

Also, given the motional, shunt, and load capacitance of the crystal in the Ice Tube Clock kit (0.0035 pF, 1.60 pF, and 12.5), the clock will run slower if the load capacitance of the circuit is indeed 18.1 pF:

0.0035 * (18.1 - 12.5) * 10^6 / 2 / (1.60 + 18.1) / (1.60 + 12.5) = 35 ppm

This means that the Ice Tube Clock should run slow by 3 seconds/day:

35 ppm / 10^6 * (24 hours / day) * (60 minutes / hour) * (60 seconds / minute) = 3 seconds / day

Interestingly, a popular feature for alternative firmwares is manual drift correction which adjusts timekeeping by units of 24/256 seconds per day. Only two users reported their drift correction values, but both reported -35 which corresponds to the clock running slow by

35 * 24 / 256 = 3 seconds / day

And three seconds per day is exactly what I've observed in my Ice Tube Clocks with 20 pF caps. Therefore, I suspect that the 20 pF caps introduce excessive load capacitance which causes the clock to run slow and also causes the kit to be more sensitive to any source of additional stray capacitance.

User avatar
adafruit_support_bill
 
Posts: 88086
Joined: Sat Feb 07, 2009 10:11 am

Re: Clock stopped after a few days. Crystal?

Post by adafruit_support_bill »

The solder joints look dull, which can be a sign of a cold joint. But that might just be due to the lighting. It looks like you are using an on-camera flash. Light from a window or even outside usually works better for circuit board photos.

Lead-free solder can also have that appearance. Are you using lead-free?

If you are using standard lead/tin solder and your joints do not look smooth and shiny, re-touch them with a clean hot iron until they do. See our soldering guide for tips:
http://learn.adafruit.com/adafruit-guid ... ring/tools

User avatar
maverickhornet
 
Posts: 25
Joined: Mon Mar 11, 2013 11:24 am

Re: Clock stopped after a few days. Crystal?

Post by maverickhornet »

@jarchie thanks for the tip, i never thought it was the flux as this has never happen to me before will clean it and I will tell you the result

@bill must be the flash as u said, but no worries what i will do is, clean the board and re-solder the joint will update you when I complete it

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

Return to “Clock Kits (discontinued)”