Dallas DS1307 Real time clock stop working

[NikosT]

Hi everybody

I am using Dallas DS1307 Real time clock and I have a problem with the oscillator.

There are about 2000 devices with DS1307 that are working for about 5 years.
Some of them have a problem with the oscillator of DS1307; the oscillator suddenly stops.
I have read the configuration byte and I have seen that the oscillator is disabled.

This is not happen in regular times. Some devices have this problem after months while
others after years and most of them never.

The devices are taximeters which means that are permanently powered by car's battery so
DS1307 rarely switches to backup battery.

Also there is software trap on DS1307 writing; the routine which writes in DS1307 does not allow
the firmware to disable the oscillator, so this problem cannot be a software bug.

Had anyone experienced the same problem?

Thanks
Nikos

 

[Imperyal]

Hey there... I have the same issue, I have a board with a DS1307 that I have replaced two times now (only lasted two months)! I don't know what's happening. I have cheep Chinese DS1307 devices... I ordered a new one today, 10x more expensive. But I'm not shure if that's the problem because I have seen lots of people whith the problem, but no solution yet. They simply stop working. In my setup the DS1307 is powered only when in use, both the main and the battery power. When in use, the time is set, and the device start. Any help would be nice. Thank you!

 

[sreepss]

1.Ground the 32.7Khz crystal's body.
2.Use decoupling capacitors.
3.Ensure that the Oscillator Enable bit od DS1307 was properly configured

 

[Easyrider83]

Increase load capacitors value.

 

[NikosT]

I couldn't find the cause of the problem, so I found another solution.
In my application, firmware reads RTC every 10 seconds.
if oscillator is disabled, firmware enables it.
After that, no RTC oscillator has been disabled.

 

[Enlightenment]

Did anyone have to replace the crystal / capacitor(s) / battery to fix the problem?

Is it possible that crystals are breaking because of shock / vibration issues? Being bounced around too much? Is the crystal firmly attached to PCB? Is the metal case grounded?

Are there battery issues? Cold solder joint for socket? Battery is not tight in socket. Battery not holding charge? Cheap or crappy batteries?

Are there capacitor issues? Cheap or crappy caps attached to the crystal?

Before replacing anything, make sure you heat up all pins in crystal part of circuit and battery part of circuit to ensure there aren't any bad solder joints because of shock / vibration issues.

Are the users exposing them to TOO HOT or TOO COLD conditions, thus causing temperature related problems?

---------- Post added at 23:37 ---------- Previous post was at 23:31 ----------

So are you having problems ONLY with cheap china clones, but not true MAXIM parts?

 

[Imperyal]

Hello! Thank you for all the replys.. I think my new top quality ds1307 part is coming tomorrow... But I notice one problem on my board, the crystal metal can is not grounded. And I have no caps on the crystal, I thought that the ds1307 have them built in!? Thanks!

More info: in my application I placed a ceramic cap in the power supply of the chip. And I'm putting 3.3v on the battery pin only when the ds1307 is in use... Don't know if that is good practice. I think my crystal is not damaged because it was working all the time with deferent ds1307s. The bord is operating in -2 to 28 degrees Celsius range.

 

[NikosT]

But I notice one problem on my board, the crystal metal can is not grounded. And I have no caps on the crystal, I thought that the ds1307 have them built in!?

In ds1307 datasheet there is no reference about connecting capacitors on the crystal, so I suppose that somewhere else is the problem but I do not know where is.

 

[alexxx]

Also the used crystal should have specified load capacitance=12.5pF.
https://www.sparkfun.com/datasheets/Components/DS1307.pdf

 

[Enlightenment]

#1 - Do not reference datasheets from 3rd party site. Always go to manufacturers web page for datasheet to ensure you have the very latest. The datasheet at SparkFun is not the latest. Use **broken link removed**

#2 Per above datasheet, if you don't have a battery on pin3, VBAT, then it MUST be GROUNDED. See page 6, "If a backup supply is not required, VBAT must be grounded".

#3 Per above datasheet, I don't think that you need external capacitors. See page 6, "The internal oscillator circuitry is designed for operation with a crystal having a specified load capacitance (CL) of 12.5pF". See Figure 2 on page 7, it doesn't show any capacitors attached to the crystal.

#4 VCC must be 4.5 to 5.5 volt.

#5 My comments in previous post was generic thoughts about crystals and batteries instead of targeting this specific chip.

 

[Imperyal]

Once again, thank you all for the help given me.
The problem turns out to be very easy to solve... I removed the board from the box, and I touched the little crystal and it imiadtly came off... The vibration damaged the legs on it. Soo I replaced it and soldered the case to the ground plane, now it can not vibrate. And working now, with the same ds1307 part that was already in the circuit. Vary stupid thing.
Let's see if the ds1307 (same cheap Chinese one) lasts!

I have another problem on one other project that I will put here, on a new tread, see if you guys can help me out.

Thanks everyone! (again)

 

[Enlightenment]

Where do you buy the cheap chinease DS1307 parts?

 

[Imperyal]

I have the parts (ds1307s only) from cwithk on eBay.