RTC battery issue - Ground discontinuity and battery sudden discharge issue

[Vijetha H.N]

Hi,

We use a Panasonic Lithium coin cell (3V, part #BR2330) for RTC chip DS1307, in the design of a handheld device and its a mass production. Now in some of the boards 2 different issue are found. Please drop your comments/suggestions on the issue.

1. The RTC battery is connected to an 8 layer PCB as through hole. After assembly, the continuity between the PCB Ground point and the Ground point for RTC battery is lost. Now i removed RTC battery, still there is no continuity between the GND point of PCB and ther GND hole for RTC battery.
If I apply pressure into the hole of GND for RTC battery; i.e if i tightly mount or insert any object into that hole for RTC battery GND, the continuity is regained. If the inserted object is removed; i.e if the pressure is released again no continuity. The GND layer comes as 3rd layer. I feel dude to excessive temperature, the copper wall of the GND hole for RTC got deformed internally and hence got disconnected from the 3rd GND layer. Can there be any other possibilities?

There is another different but related issue.

2. A board was kept under ESD protection in a box for 1 month. After 1 month the battery got drained to 0.2-0.5V.
Actually RTC battery in addition to supplying to RTC chip; is also used as back up to GPS module (LEA6H) to minimize time to first fix. RTC chip take s a current of around 500nA and the GPS module takes around 0.024mA (only when GPS application is not active). Battery capacity is 255mAh. So atleast it must supply for minimum 1 year. What can be the reason for the sudden draining of battery then?
After replacing the battery its working fine.

Plz give your comments on these.

Thank You and regards,
Vijethahn

- - - Updated - - -

Let me know if any of the points mentioned above is not clear to the reader.

 

[zia]

plz show me circuit + detail

 

[crutschow]

Multilayer boards can have problems with moisture absorption from humidity which can cause delamination of the via connections when the boards are soldered due to steam pressure. To avoid this at the company I worked, the boards were baked out in a oven at temperature (don't know the temperature or duration) and then placed in a dry nitrogen purge box until they were soldered.

Measure the current drain from the battery circuit to see if it is within requirements.

 

[betwixt]

1. It looks like you are right, is it also possible that other battery hardware is causing distortion of the area around it. I'm thinking that if the battery is held in a metal 'cage' holder, is the thermal mass/capcacity of it and it's solder joints significantly different to the area the battery lies against on the PCB itself. As a final resort, is the break definitely in the inner layer? If the through hole is still connected to the inner layer, can you flood it with conductive paint or solder over it to give a new contact area.

2. Two possibiliies come to mind - firstly, is the box actually conductive and if so, is there a discharge path through contact with it? Secondly, does the GPS unit remain active while searching for a satellite lock in which case the box may be screening it so it never sees a signal and keeps searching instead of 'powers down'.

Brian.

 

[Vijetha H.N]

Dear Brian,
Thanks for the valuable suggestions.
1.I flooded the hole through solder, it was not working still. I did X-Ray scanning of that particular area, observed that in the n.working board the inner copper wall was not uniform completely. In working board it was uniform. I attached the capture of the hole in n.working board. One more thing, the assembly vendors informed that, the RTC battery is connected through hand soldering (other components through Machine soldering) and the temperature the hole can withstand without any damage is 340 deg C. Now m doubting this, we can not expect accuracy with the manual soldering, the temperature might have exceeded the threshold or distributed randomly.

2. GPS was inactive; the RTC battery supplies only as a backup for the stored data.
It was kept inside a non conductive box, covered in an ESD cover.
I need to check and compare the weights of the drained battery and working battery. Because i read that, if the battery is under stress the electrolyte evaporation can happen which can drain off the battery. If the electrolyte solution is evaporated, then there should be weight loss right? Please suggest / comment on this.

Thank You,
Regards,
Vijetha H.N

 

[FvM]

1.I flooded the hole through solder, it was not working still. I did X-Ray scanning of that particular area, observed that in the n.working board the inner copper wall was not uniform completely. In working board it was uniform. I attached the capture of the hole in n.working board. One more thing, the assembly vendors informed that, the RTC battery is connected through hand soldering (other components through Machine soldering) and the temperature the hole can withstand without any damage is 340 deg C. Now m doubting this, we can not expect accuracy with the manual soldering, the temperature might have exceeded the threshold or distributed randomly.

Apart from broken power plane connection due to PCB delamination, you should also consider copper dissolution by lead-free solder alloy. It's particularly likely to happen with larger throughplated holes and plating thickness at the lower end of the process window. Of course elevated solder temperatures and duration are also promoting it.

Besides strict processing instructions and PCB quality control, special solder alloys with reduced copper dissolution can reduce the risk. Most major solder alloy manufacturers are addressing the problem.

If you're not sure about the chances in this regard, you should supplement large throughplated holes with vias and copper at the outer layers as a redundant connection.