[Ground-station] Yes, you can buy a cubesat, but...

[Phil Karn]

On 4/23/18 22:30, Douglas Quagliana wrote:
>
> Anyway, if anybody know *ANY* of the details of Larry's battery
> matching, please
> share them.  I'd like to reproduce his results one day.

I vaguely remember talking with him about this. I believe one of the
steps was to X-ray each battery and disqualify any that looked different
from the others.

Batteries aren't a black art, though it can sure seem like it. There are
few pretty basic concepts, such as the need to balance individual cells
in series strings to have the same state of charge, and to prevent any
of them from going outside their do-not-exceed voltage ranges. (This is
absolutely critical in electric vehicles, which typically have string
voltages of 300-400V).

Lithium-ion is both better and worse here. It's better in that coulombic
efficiency is 100%; there's no water in the electrolyte to steal
electrons. This makes it difficult for the cells to get out of balance.
But they're worse in that if they ever do, or if changes in individual
cell capacities cause an individual cell voltage to exceed allowable
limits, the results are much more catastrophic. Discharge a Li-ion cell
below the minimum voltage and you'll ruin it. Charge it above the
allowable maximum and it's quite likely to catch fire or explode.

So every battery pack containing a series string of li-ion cells has a
battery monitor circuit built in. Its sole job is to monitor the voltage
on each and every cell and open up the external leads if any one exceeds
an allowable limit. They may also open on temperature, I'm not sure. The
disconnection is usually via a MOSFET switch.

All of our old spacecraft batteries were nickel-cadmium, as I recall.
These things could take a lot more abuse, including taking them down to
0V. In fact I believe they were stored shorted. They can take moderate
overcharging; water electrolysis would occur, but the hydrogen and
oxygen would recombine in a controlled manner if the rate wasn't too
high. (Otherwise a vent opens and the cell is pretty much toast).

But because they have water electrolytes, it's much harder to keep them
balanced in series strings indefinitely. I think the usual way around
this was to deliberately but gently overcharge them to ensure they were
all at 100%. Again, this is something you simply can't do with li-ion.

So the bottom line is that I think Larry's tests were mainly about
ensuring that individual cells in series strings stayed close in voltage
over time as the pack was cycled. This was fine for NiCd, but may not be
all that applicable to Li-ion.

Also, because of li-ion's higher terminal voltage and advances in low
voltage power transistors, it is entirely practical to power a lot of
devices from just one li-ion cell with good efficiency. This completely
avoids any and all problems associated with series strings.

Phil
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0xF14E4354ED463446.asc
Type: application/pgp-keys
Size: 3301 bytes
Desc: not available
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20180423/36ec5a19/attachment.key>