This statement gets used often like its a fact. It may be but I do not get it. The only way this is possible if Li cells fail open circuit as a rule. If they do they are risky to use. Nicads on the other hand we know to fail closed circuit or open circuit, depending on the failure mode.

We know nicads wired in parrallel will, if one cell fails closed circuit, cause a high inrush (charging current into the shorted cell) causing loss of voltage and capacity.
We also know that when it fails open circuit voltage will be maintained but capacity will halve and internal resistance double.

What do we know of Li packs except for marketing hype and generalisations in our application. Any vendor prepared to state the facts in verifiable scientific terms?

Backup power systems are of course a good thing in principle but they do add to risk of failure. The risk of failure in the case of batteries are however probably higher. Where these units provide an independent regulated and noise filtered RX supply they would be helping quite a bit in a multiple digital servo installation.

Andre Baird