1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Changing CR2032 clock battery in S3 OLED

Discussion in 'TiVo Series3 HDTV DVRs' started by sfhub, Sep 27, 2013.

  1. sfhub

    sfhub Active Member

    1,269
    0
    Jan 6, 2007
    I had my S3 OLED open after bad cap and hard drive repair so I figured I'd change out the CR2032 clock battery at the same time.

    This is just info for future reference.

    The unit has been running since around Mar 2007, so that makes around 6.5 years.

    Given the length of operations, I figured the battery was probably pretty close to being done or maybe was already done. When I rebooted, certain values in system information like "Guide cache to" and VCM connection info would reset to December 31.

    It turns out those values reset to December 31 regardless of the clock battery. The only job of the clock battery appears to be to, surprise, keep the clock going when power goes out. So the only reason to replace it (other than being proactive) is if your "Today's date" resets to "Tue Dec 31, 2002 4:00 pm" when you pull the plug.

    Even if it does reset, you don't really need to replace the battery as upon first connect to mothership, it will correct the clock, but it might result in more background indexing than necessary. Also you might miss season passes for up to 24 hours until the first connect to mothership resets the clock to current date/time. You can alleviate that by forcing a connect to mothership after pulling the plug but may not be able to do that if you have a power outage and don't have access to the unit. If you have your TiVo connected to a UPS, there is even less reason to replace it.

    Anyway, if you decide to replace the battery, be CAREFUL when you remove it. The "positive" retainer "arm"/"clip" is not designed to be lifted up. If you try to do so, there is a good chance you will BREAK the arm. Instead, insert a small jeweler's flathead screwdriver underneath the clock battery and twist around 30-45 degrees to get the battery slightly lifted, then push/slide the battery towards the side of the receptacle OPPOSITE of where the retainer arm connects. The battery will just pop out.

    Though it isn't absolutely necessary, I found it easier to get clearance to remove the battery after removing the hard drive cage.

    To insert the new battery, just tilt it slightly to get under the retainer arm, and push/slide it back in.

    The original battery was Panasonic CR2032 from Indonesia. It tested 2.98 V after 6.5 years. My "Today's date" had never reset after pulling the plug, and the CR2032 is listed as supplying 3V, so it probably had a couple of years left.

    The replacement battery tested at 3.25V.

    Upon first boot, your clock will be reset to Tue Dec 31, 2002 4:00 pm. The first connect to mothership takes longer than normal. A bunch of time fields in the system information will reset to Dec 31. Everything is working fine after replacement and subsequent connect to mothership.

    It is interesting to note that TiVo doesn't set its time at startup. It only appears to set the time when it connects to the mothership. If I had not forced a connect to mothership, the time would probably be wrong for 24 hours and season passes wouldn't record. I imagine this is what someone might experience if their clock battery was dead.

    In conclusion, IMO it isn't really necessary to replace the clock battery, and I wouldn't bother in most cases, certainly not until your "Today's date" starts resetting to the past. I did it because I was curious and had the box open already.

    Hope this at least prevents someone from breaking their clock battery arm just for being curious.
     
  2. wtkflhn

    wtkflhn Member

    190
    3
    May 12, 2006
    Metairie, La
    Thanks for the info.
     
  3. squint

    squint New Member

    846
    0
    Jun 15, 2008
    I've found that testing open circuit voltage of lithium coin cells wasn't indicative of their capacity. I had to test them in parallel with a small load, a 100 ohm resistor, to weed out my depleted lithium coin cells.
     
  4. sfhub

    sfhub Active Member

    1,269
    0
    Jan 6, 2007
    I can believe the coin cell can test lower with a small load than just with a voltmeter, but for purposes of coin battery cell elimination, I only deal with new batteries and batteries I pulled.

    The pulled ones, I'm usually pulling because they are near their end of life anyway in terms of years in operation or the device they support stopped functioning, so I'm not testing to decide whether to keep them or not, just for curiosity. After I made the effort to remove them, I'm not going to bother putting them back in, so a new one goes in to replace automatically.

    The new ones, have a very low rate of failure, so I'm mainly testing to make sure they aren't dead and have a value in the expected range.

    As an aside, every coin cell I've pulled because the device they were supporting stopped functioning has been significantly below 3V. I've never pulled a coin cell that stopped working and found the voltage to be 3V or above.
     
  5. squint

    squint New Member

    846
    0
    Jun 15, 2008
  6. sfhub

    sfhub Active Member

    1,269
    0
    Jan 6, 2007
    As I mentioned I have no problem believing that a coin cell voltage can test lower with load, which is what that thread is saying. They are basically discussing coin cells that test at 3V but are actually dead.

    My point was testing high might not mean the coin cell is usable (which is the same thing the thread is saying) but testing low means it is not.

    Also I point out that by the time I end up pulling the cells, it isn't to test them, but rather to replace them regardless. I only test them on the side out of curiosity. I'll pull them either when they have been in service for many years or if what they are supporting is failing. I only replace with new cells and in my personal experience, the failure rate for new cells is very low.

    I'll test some new cells with a load but my guess is the failure rate on new cells is so low that it isn't even worth the effort to dig out a resistor.

    Regarding the old cell testing 2.98V it is possible it would test lower with load, but it wouldn't matter to me because I wouldn't be putting it back in the unit. Also I know for sure it isn't dead to the point of being unusable because my RTC is still being preserved after pulling the plug.
     

Share This Page