PDA

View Full Version : Soldered Dallas 1287. Replace or Mod?



twistedpneumatic
October 29th, 2017, 03:27 PM
There probably are a billion posts on this but should I mod it or solder a new one. There is NO room for a socket. Not even a single chance of it.

krebizfan
October 29th, 2017, 03:46 PM
What is the system? Plenty of systems had secret pins for attaching a normal battery.

Modding is much easier than removing a Dallas chip completely. Strange system that does not have the fraction of inch of headroom to install a socket.

twistedpneumatic
October 29th, 2017, 03:59 PM
The idiots at Magnavox put a piece of plastic protecting the LCD cable exactly above the chip. It is also a laptop so I HIGHLY doubt there is an external spot to place it.

It is a Magnavox Metalis 286. If you would like MOBO pics, I can get those. It happens to have a 4.8V 1200mah secondary battery but I doubt that is for powering the RTC.

twistedpneumatic
October 29th, 2017, 04:19 PM
Ok, there is like a 1/4" to play with. Will that still work for a socket?

SomeGuy
October 29th, 2017, 04:34 PM
At least on a 286 era machine you stand a chance of being able to de-solder it. Pentium era or later motherboards redistribute heat too well.

Either way you will have to de-solder the chip. You might just have to try fitting a socket in there without immediately soldering it and see if things still fit together.

twistedpneumatic
October 29th, 2017, 04:37 PM
I cut down a different sized socket and cut its pins to see if it would fit but it didn't. I just pushed it between the plastic and chip. The chances a real socket to fit is slim.

KC9UDX
October 29th, 2017, 05:11 PM
You might be able to fit some low profile turned female turned header pins.

Like this, if the link works https://www.mpe-connector.de/index.php?lang=en&menu=16&mating=0351&id_product=0081

Chuck(G)
October 29th, 2017, 05:25 PM
Well, you could always take a plain old DS12885 DIP and add a crystal and external battery. There would be more than enough room for a socket. Didn't someone also make a PCB DIP adapter for an SMT 12885?

twistedpneumatic
October 29th, 2017, 05:43 PM
Problem with that is the LCD connector is exactly beside the chip so no crystal would fit, and neither would a wire to run the crystal.

Chuck(G)
October 29th, 2017, 06:07 PM
Watch crystals are tiny--and you could mount one on top of the DIP. If you've got the SMT version, there's oodles of space.

pietja
October 30th, 2017, 06:09 AM
I had the same problem with an 486 laptop so i made this little adapter board.
416134165341654
You can order the board here: https://oshpark.com/shared_projects/VMMt8B7N

glitch
October 30th, 2017, 06:15 AM
I've got a replacement module that uses the Dallas 12885 in DIP form, there's a forum thread here:

http://www.vcfed.org/forum/showthread.php?59536-GW-12887-1-A-DS1287-DS12887-et-al-Replacement

I haven't ordered a production run of boards, but I have 100% functional prototypes if you need one. Fully assembled and installed on a new DS12885.

Chuck(G)
October 30th, 2017, 08:15 AM
Personally, given the opportunity, I wouldn't let a battery of any stripe near a motherboard. I worked with an engineer back in the 70s whose precept was "printed circuit boards and batteries should not exist together", when I suggested a battery-backed clock.

I've got several systems where I've moved the clock battery off-board, so if the thing gets all weepy and decides to spill its guts, no harm done to the PCB.

glitch
October 30th, 2017, 09:45 AM
Personally, given the opportunity, I wouldn't let a battery of any stripe near a motherboard. I worked with an engineer back in the 70s whose precept was "printed circuit boards and batteries should not exist together", when I suggested a battery-backed clock.

I've got several systems where I've moved the clock battery off-board, so if the thing gets all weepy and decides to spill its guts, no harm done to the PCB.

Not too much of a concern with coin cells (there's not much in there, and whatever it is, is pretty volatile), even in the rare case that they are persuaded to leak. I've only had Lithium coin cells leak from mechanical damage, or chemical damage that ruined the rest of the board anyway.

KC9UDX
October 30th, 2017, 02:45 PM
I can't entirely agree with that. The only two old things that I've worked on with coin cells in them had battery leakage damage.

One was a watch that had a cell in it for around 40 years. The other a drum machine with one in it for around 30 years.

In both cases the cell had expanded to the point that the seal was working it's way out, and a little fluid had leaked past it. This is not what I had expected to see.

glitch
October 30th, 2017, 03:26 PM
What kind of coin cells, though? There are a variety of chemistries that come in the format.

KC9UDX
October 30th, 2017, 04:37 PM
What kind of coin cells, though? There are a variety of chemistries that come in the format.

The one in the watch was a mercury cell LR44 size or similar, the other was a CR2032.

Chuck(G)
October 30th, 2017, 04:41 PM
I wonder if "leakproof" potted-in-epoxy cells are available anywhere.

glitch
October 30th, 2017, 04:44 PM
Not surprising about the mercury cell. I can't speak for all coin cells, but the organic solvent in Panasonic and Renata CR1225 and CR2032 coin cells is quite volatile and smells a lot like acetone. I wouldn't expect a *major* release of it to do any real damage to a circuit board, unless it managed to carry all of the cell's lithium salt with it.

glitch
October 30th, 2017, 04:49 PM
I wonder if "leakproof" potted-in-epoxy cells are available anywhere.


You can get Lithium Thionyl Chloride in hermetically sealed welded case cells with glass bead lead seals. I'm only familiar with the Eagle Picher "Keeper" variety:

https://www.eaglepicher.com/technology/battery-chemistries/lithium-thionyl-chloride/

They're not cheap, but they're common in industrial control and embedded stuff

KC9UDX
October 30th, 2017, 05:02 PM
Not surprising about the mercury cell. I can't speak for all coin cells, but the organic solvent in Panasonic and Renata CR1225 and CR2032 coin cells is quite volatile and smells a lot like acetone. I wouldn't expect a *major* release of it to do any real damage to a circuit board, unless it managed to carry all of the cell's lithium salt with it.

This one was a 2032 with legs welded on, which were soldered to the circuit board. I don't recall the brand, whatever a Japanese manufacturer would have used in 1986, National or Panasonic maybe. It left a pretty good puddle and turned some traces green, but due to the sparcity of components in the area it was a fairly easy cleanup.