Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

replacing rechargeable 3V coin cell battery

Status
Not open for further replies.

mdanh2002

Member
Hi,

Haven't visited the forum in a while.

I am trying to repair three Powerbooks (140/160/180) which I bought from eBay as a lot for a cheap price. All of them work well after some simple hardware fixes (cable reseat, removing main batteries, removing faulty hard drives / modem / memory modules which were probably causing shorts), just that the CMOS batteries are long dead. The design uses a Panasonic VL2320/UL2320 3V 30mAh rechargeable lithium-ion coin cell battery. When searching eBay I found the following close matches:

+ VL2330 (3V, 50mAh)
+ ML2032 (3V, 65mAh)
+ ML1220 (3V, 15mAh)

Is it possible to replace the original battery with either the VL2330 or the ML2032? I seldom care when I replace CMOS battery, but in this case since it's rechargeable I just want to make sure everything is correct. Don't want the thing to explode and destroy my vintage Powerbook.

Another thing, for these rechargeable batteries, sometimes the specs say "lithium", sometimes "lithium-ion". Does the difference matter, or is it just some difference in the way the specs are written? My research shows that most lithium batteries are not rechargeable, so "lithium" is probably referring to "lithium-ion". But just want to be sure :)
 
Last edited:
According to RS, the VL2320 is a lithium vanadium pentoxide cell. It's a tabbed cell; the ML2032 isn't.
 
Here is the data sheet: **broken link removed**

Electrically, any long-shelf-life, 3V battery with a capacity equal to or greater than the original will be fine. But as Alec implies, the physical aspects need to be resolved. You can always solder a coin battery holder to the mother board, or if space prohibits, use flying leads. That way you can use a leadless coin cell.

Are you sure the original batteries are faulty? Remove the batteries from the motherboard and pump around 5mA into them (9V battery and a 1K8 resistor will do) and see if the batteries hold their charge.

spec
 
Here is the data sheet: **broken link removed**

Electrically, any long-shelf-life, 3V battery with a capacity equal to or greater than the original will be fine. But as Alec implies, the physical aspects need to be resolved. You can always solder a coin battery holder to the mother board, or if space prohibits, use flying leads. That way you can use a leadless coin cell.

Are you sure the original batteries are faulty? Remove the batteries from the motherboard and pump around 5mA into them (9V battery and a 1K8 resistor will do) and see if the batteries hold their charge.

spec

Yes, I think the batteries are long dead. The symptom is that the real-time clock for some reasons pauses (not resets, just pauses) when the PowerBook is not powered on despite leaving it long enough (overnight) for the battery to be fully charged. For example, if I start my computer at 7am, the clock will run correctly until the computer is powered off. However, let's say if I power it off at 9am and turn it on again in the afternoon, the clock will start from 9am, exactly when the computer was shut down. This is weird as on most PCs I have seen, a dead CMOS battery will probably cause the time to reset. Maybe the Powerbook stores the system settings, including the time, on PRAM and restores it on the next boot if the RTC fails to work. Does PRAM need power to retain its data? I could not find any reliable information on this.

Btw, soldering a new battery holder on this board is a pain (the battery is beneath the LCD panel in a very confined space). I am probably going to cut the old battery away, buy the the VL2330 with tabs, and solder it back onto the old tabs.
 
Last edited:
Are you sure that the RTC chip uses Phase Change Memory? Never heard of that before. PCM, is non-volatile and does not need a supply line to hold its contents.

Normally RTC chips, in addition to the RTC function, have a small amount of static RAM that is held up by the battery (nearly always, 3V, CR2032).

The RTC chip static RAM normally stores various system configuration settings.

An error with these settings can cause absolute havoc and prevent the computer from booting, as I experienced recently on a laptop. Strangely, a keyboard fault can do the same thing. I'm sure the people who design computers have a sense of humor.:eek:

The only cure for the former is to remove the RTC battery, wait about 10 minutes, and then reconnect the RTC battery.

The cure for the second is to fix the keyboard, or fit a new one.

When the host is off, the RTC chip runs off 3V from the RTC battery, when it consumes a minute amount of current. When the host is powered on the RTC power line changes to 5V typically, which is sourced from the host.

spec

PS: interesting blog :cool:
 
Last edited:
Are you sure that the RTC chip uses Phase Change Memory? Never heard of that before. Normally RTC chips, in addition to the RTC function, have a small amount of static RAM that is held up by the battery (nearly always, 3V, CR2032).

The static RAM normally holds various system configuration settings.

An error with these settings can cause absolute havoc and prevent the computer from booting, as I experienced recently on a laptop. Strangely, a keyboard fault can do the same thing. I'm sure the people who design computers have a sense of humor.:eek:

The only cure for the former is to remove the RTC battery, wait about 10 minutes, and then reconnect the RTC battery.

The cure for the second is to fix the keyboard, or fit a new one.

I do not know the answer to your question about PCM, but static RAM does need a battery to maintain its contents.

Normally, when the host is off, the RTC chip runs off 3V from the RTC battery, when it consumes a minute amount of current. When the host is powered on the RTC power line changes to 5V typically, which is sourced by the host.

spec

PS: interesting blog :cool:

From what I know, the Macintosh Powerbook uses PRAM to store various system settings such as time, volume, and other miscellaneous settings. I think the RTC module on this machine is just that, an RTC, without other features, unlike on a PC where BIOS settings are stored on static RAM that needs to be powered by the CMOS battery. I guess the restoration of the time when the system reboots is done by the OS (Macintosh System 7.5.5) when an RTC failure is detected. And yes, the system boots fine and works properly, except for the RTC issue. If the PRAM does not need battery to maitain its content, this explains the issue.

Another reason could be that the remaining battery power is not enough to oscillate the 32kHz crystal to keep the RTC running when on battery, but still enough to hold the PRAM memory contents. When the computer is powered on, the RTC works fine and the clock runs because it no longer uses the battery as power source. Not exactly sure, but just guessing.

On a side note, just a few weeks ago I repaired a Commodore 386SX-LT laptop. The system booted up with a 'CMOS Battery Bad' warning and prompted to press F1 to enter SETUP. However, entering the "General Configuration" setup screen (with date/time, floppy, and hard disk setup) would apparently hang the machine. I eventually discovered that tapping on the diode near the CMOS battery would temporarily fix the issue and make the BIOS setup screen responsive again, although the RTC would run for only 15-20 seconds before it stopped. The root cause turned out to be that the diode near the CMOS battery was somehow faulty and read as open. By tapping it I perhaps grounded the connections and allowed power to be supplied to the RTC. I replaced the diode with a 1N4148, soldered a battery holder, and the laptop is up and running again :) Whoever designed the BIOS setup program for this laptop apparently did not handle the case when the RTC would fail.

Just ordered the VL2330 from eBay. When the batteries arrive after a few weeks, I will replace them and see if the RTC will work properly again :)
 
Last edited:
Because of the microscopic currents involved with RTCs, contacts can be a problem, even solder joints which can get crystallized. The solution is often to reflow the solder joints and lightly burnish all contacts with a rubber eraser.

Real Time Clocks always seem to be a source of problems: I once worked on a huge computer controlled RADAR tracking system for the military. It would track numerous aircraft at high Mach numbers. identify them, send messages to a central control system, make the operators coffee... but could the processors maintain the correct real time: no!

spec

(have I mentioned before that when a kid I lived at Seletar RAF base in the late 1950s?)
 
Last edited:
After some further research, it seems that PRAM in the Apple Powerbook context simply refers to Parameter Random Access Memory, a type of static RAM that needs power to retain its data, mostly the Macintosh system parameters. It is not the Phase Change Memory (PCM) which I assumed earlier.

So I believe the RTC pause behaviour could either be caused by low battery voltage or because of the Macintosh OS caching the time on the disk drive and restoring the last RTC time upon startup if the RTC is not accessible. My Amiga 500 also does something similar - it uses the floppy disk last update timestamp as the starting point for the clock if an RTC cannot be detected.

As far as RTC issues go, many people do not bother to set the correct time in these clocks, even for apparently mission-critical systems. Here in Singapore on one of the subway lines (we call it MRT locally, short for Mass Rapid Transit), the LED display on the train would show the current time for a brief moment when the train reaches the terminal (it doesn't always show the time, only at the terminal for around 5 minutes when the train reverses its direction). More than ten years ago when this line was still new, I noticed that the clock was off for around 10 minutes (the date was correct, only the time was off). Years passed and despite many upgrades and maintenance, recently I came across the time display again, and it was still off by 10 minutes! Maybe setting the RTC correctly was never part of the maintenance procedures.

spec, yes you told me about that before. I am still living quite close to Seletar at the moment. Still wonder what Singapore was like back then. :)
 
Last edited:
After some further research, it seems that PRAM in the Apple Powerbook context simply refers to Parameter Random Access Memory, a type of static RAM that needs power to retain its data, mostly the Macintosh system parameters. It is not the Phase Change Memory (PCM) which I assumed earlier.
Ah, I didn't think PCM was used commercially.:)

As far as RTC issues go, many people do not bother to set the correct time in these clocks, even for apparently mission-critical systems. Here in Singapore on one of the subway lines (we call it MRT locally, short for Mass Rapid Transit), the LED display on the train would show the current time for a brief moment when the train reaches the terminal (it doesn't always show the time, only at the terminal for around 5 minutes when the train reverses its direction). More than ten years ago when this line was still new, I noticed that the clock was off for around 10 minutes (the date was correct, only the time was off). Years passed and despite many upgrades and maintenance, recently I came across the time display again, and it was still off by 10 minutes! Maybe setting the RTC correctly was never part of the maintenance procedures.
That sort of thing is common with technology- the number of times CCTV cameras are not operating properly when the police need a playback to investigate a crime.

I have yet to see a hotel TV system that works properly.

There is a super high tech information display system in the central bus station in Bristol. It displays what time the buses should be departing and which bay they should be departing from, not what time they are actually departing and which bay there are actually departing from.

The information system at Cardiff railway station tells you how wonderful the railway service is and makes all sort of other general statements. But it didn't mention that the train that we were waiting for was late and would be leaving from a different platform.

spec, yes you told me about that before. I am still living quite close to Seletar at the moment. Still wonder what Singapore was like back then. :)

Yes, sorry I remembered after the post. In those days Singapore was a fantastic place. We used to love the curries wrapped in banana leaves that the street stalls sold. And as a kid, when school finished at about 12:30pm, it was sunshine, swimming and boating every day. I was heavily into aero modeling and the weather was ideal for flying.:cool:
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top