Batteries suck, chargers suck, monitoring sucks: surely we can do better than this
Last week, ESR opened up the work-in-progress on GitLab: the Upside project is currently defining requirements and developing a specification for a “high quality UPS that can be built from off-the-shelf parts in any reasonably well-equipped makerspace or home electronics shop”.
ESR’s original post, “UPSes suck and need to be disrupted”, set down his own complaints about what’s sold to consumer/SOHO users: batteries with “so little deep-cycle endurance” that they can’t last beyond a few years, and whose dwell-time is oversold by vendors.
UPSs lack the kind of sensor information that protected car batteries, Raymond wrote: they measure output voltage, discharge current, and battery temperature, to predict its likely dwell time and identify when batteries are failing.
Finally, Raymond wrote he want the UPS to provide text-based alarm messages (rather than flashing lamps), and to provide decent monitoring information over USB.
Since that post, a handful of others have joined the project: Eric Baskin is leading hardware engineering for Upside, Jay Maynard is developing firmware, and Jeremy Mitts is copy-editing the documents; and the group has expanded the requirements.
The newrequirements document has expanded to include:
- Smart charging – to preserve the battery, the charging circuit needs to be more than a constant trickle-charge that degrades the battery;
- Batteries – should be more sophisticated than lead-acid; LiFePO (lithium iron phosphate) is suggested instead;
- The open source UPS should be able to deliver 300W for 15 minutes; and
- The UPS should fail to a bypass mode (that is, pass mains power through) rather than “off”, which Raymond wrote is too often the default for today’s devices.
The Register has heard enough UPS horror stories to consider the project well worth watching. ®