• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Lithium Polymer battery for PB100 (and maybe portable)

techknight

Well-known member
Yea.. its called clamp diodes. designed to protect the inputs and outputs, and they all go back to the main rail. So it makes sense. 

 

360alaska

Well-known member
I haven't quite had a chance to redesign the board. I'm actually renovating my house right now and that does take a lot of my time...

 

daveosx

Member
About 15 years ago, I drove a fellow home (my job at the time was "professional designated driver") whose company had developed a method of building lead-acid batteries which, though having completely traditional chemical composition, had greatly improved energy density by reason of using "foamed" (extremely porous) lead electrodes, which naturally had vastly greater surface area than the traditional simple lead plates… or so he proudly claimed, at least. It would be interesting to find out if they ever succeeded in commercializing that. You could in theory run your Portable for twice as long with minimal or no alterations.
I think I know this guy

It may be Robert Aronsson

He has a battery company in Pompano florida now.

The try polar and lead foam batteries do have larger density at lower weight.

They are gearing up to produce them.

http://www.apolloenergysystems.com

 

360alaska

Well-known member
So I've taken a few minutes to refine my idea, I was hoping the LM358 comparitor could drive the relay by itself but really I need a transistor.  I also added another diode to generate sufficient voltage drop for comparison. I'm considering how well the circuit will work if the comparitor is replaced with a 7.5v Zener Diode.

Capture.JPG

 

360alaska

Well-known member
I prototyped the circuit and it works! I adjusted the pot so it will trigger at 7.3v. I also ordered a 7.5v Zener diode so we'll see if that is a better setup.

IMG_26661.jpg

 

just.in.time

Well-known member
Very cool! How many of those smaller Li packs will it take to get equivalent run time to originAl batteries?

 

360alaska

Well-known member
Those 1200 mah are just for testing purposes, Once I complete the design I can use much bigger cells, but obviously they will take longer to charge.

 

360alaska

Well-known member
While I'm waiting for my custom pbcs and correct batteries for my pb100 battery, here is a test unit I made to fit in my portable, it seems to work well enough, it will switch into charge mode when the charger is connected and switch into operation when the charger is disconnected, the only problem is that the computer will reset, this is like due to the relay not switching fast enough. That would mean that unless I can find an electrical solution, one would have to reboot the computer to disconnect from the charger, not a bad compromise for having a working pb100 battery. 

IMG_2730.jpgIMG_2731.jpgIMG_2735.jpgIMG_2736.jpg

 

360alaska

Well-known member
Here is the completed prototype, it will probably be 90% of the full design. It will also have a cover. I discovered through use that only one diode should be used. So I've made that adjustment. I've also got a faster acting relay on the way and I'm considering replacing LM 358 with LM393 which will require a slight redesign to accommodate LM393's open collector output.

IMG_2768.jpgIMG_2769.jpg

 

ravuya

Active member
That's an impressive bit of work, and so clean too. Is it possible to use a giant capacitor to keep things alive while the switchover happens, or is the relay just super-duper slow?

 
Last edited by a moderator:

360alaska

Well-known member
It's probably possible to do that. I've also thought of incorporating a mosfet or transistor somehow. The old relay part "TX2-5v" has a 4ms release. The new one "EC2-5NU" has a 2ms release. I doubt that will make a difference at all but who knows. I'd like to fix that but I really don't see it as a major detractor given that there essentially are no other options for this little laptop. Right now if it's running and you yank the charger it resets.

 
Last edited by a moderator:

CC_333

Well-known member
How trivial do you think this would be to adapt to other PowerBooks?

I would think that, now that you've mostly come up with the design, it could be used as a reference for other similar designs targeting models such as the 14x/150/170 and 500 series.

They all use NiCd or NiMH as opposed to SLA, but I wouldn't think it would require much of a change to adapt your design?

Nevertheless, this is a fantastic development, and if I had a PowerBook 100, I'd totally want to get one of these once you've finished it!

c

 

techknight

Well-known member
Oh yeah, I forgot about that.  Has anyone been able to reverse engineer that "intelligence"?

c
Having that intelligence offloaded to the battery itself is actually to our benefit. That way your not fighting against the internal charge controller when changing battery chemistry types and using your own BMS. 

Yea, it needs reverse engineered, and then a new BMS made that has the "smart" features built into its microcontroller 

 

CC_333

Well-known member
@techknight OK, that makes sense, because the BMS in the battery can be modified to basically look like whatever the computer expects, regardless of the actually chemistry of the cells.

I'm sure someone can reverse engineer the 5xx battery easily enough.  I mean, all kinds of things have been successfully re-implemented that were once thought to be prohibitively difficult to reverse engineer (Custom ROM SIMMs!  FloppyEmu!  SCSI2SD!  And now even a complete clone of the entire SE logic board!), so why would this be any different?

c

 

360alaska

Well-known member
Today I tested the voltage curve. I wanted to make sure the laptop would display a low power message before the BMS cuts off. I ran it for about 3 hours of mixed use and got a low battery notice I ran it for about 30 minutes between low power and critical power and the laptop was still going strong. My final design incorporates 3600 MAH batteries, the original battery had a 2500 mah battery good for 2.5 hours so I figure my lithium battery should be good for about 3.6 hours.

IMG_2780.jpg

 
Last edited by a moderator:

techknight

Well-known member
It's probably possible to do that. I've also thought of incorporating a mosfet or transistor somehow. The old relay part "TX2-5v" has a 4ms release. The new one "EC2-5NU" has a 2ms release. I doubt that will make a difference at all but who knows. I'd like to fix that but I really don't see it as a major detractor given that there essentially are no other options for this little laptop. Right now if it's running and you yank the charger it resets.
If you are doing switching with this relay from charge to discharge, vice versa. this really needs to be a series pass transistor set to do this. 

Case in point: I made a dual-port BMS for 24v. Charge comes in on a different port (solar applications). for LiFEPo4. This one is loosely based on a chinese version which when my supplier started screwing up the parts that were being assembled onto the board, I had to take matters into my own hands and study its design plus engineer a new one. This one even does cell balancing as well. 

image.png

 
Last edited by a moderator:

360alaska

Well-known member
It's not that simple,

There simply is not enough voltage to charge the pack as a 2S so the relay switches the battery between 2S and 2x 1S for charging purposes. During operation cell 1's negative is connected to cell 2's positive but during charging cell 1's negative is connect to ground and cell 2's positive is only connected to the charger. The relay is connect to a comparitor which compares battery voltage to the voltage of the charge and energizes the relay whenever the charger is connected.

 
Top