Feature idea: Quality value based on Cell test

General Discussion of the 4 Cell Charger that Paul is working on.
Post Reply
RLBrooks
Posts: 13
Joined: Fri Nov 29, 2013 4:37 am
Location: Walla Walla, WA (USA)

Feature idea: Quality value based on Cell test

Post by RLBrooks » Fri Nov 29, 2013 5:16 am

Problem: All cells are slightly different. In use (my use anyway) pairs of cells may not stay paired as I grab them from a pile of available charged cells. Over time cell differences may increase based on usage habits and recharge levels. (Is this a valid assumption so far?)

Idea: Would it be useful to be able to test a cell and assign a Quality Value to it. This would make it easier to select recharged cells with similar values and hopefully expect a more balanced performance from that pair. (Again, valid assumption?)

Since the battery charger is software controlled and has discharge capabilities could a Cell Test function be provided that was primarily designed to charge/discharge (maybe multiple times) a cell and measure it's quality (as well as leave it charged)? That value could then be written on the cell and help me select matched pairs of cells (and also serve as an indicator of a failing cell as the quality value decreases over time).

Comments?

Russel

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Feature idea: Quality value based on Cell test

Post by Mark » Fri Nov 29, 2013 6:25 am

A single quality value wouldn't be that useful in my opinion. The main things to be concerned about with a cells health is the actual discharge capacity (which can be determined by a cycle test) and the internal resistance value which is automatically reported at the start of charging and discharging.

It's possible to have a cell with low internal resistance that has lost some of it's capacity and another that has developed high internal resistance without losing much capacity. With a single quality value, both cells might get the same score even though they would be a very poor match.

If you're wanting to match cells, you should do it primarily based on capacity (so that one cell doesn't run empty before the other) and then try to get them with similar internal resistances. At higher loads, the internal resistance affects the effective capacity, so that's why that should also be matched where possible.

Paul.Allen
Site Admin
Posts: 98
Joined: Tue Aug 06, 2013 5:33 am
Location: Utah, United States

Re: Feature idea: Quality value based on Cell test

Post by Paul.Allen » Fri Nov 29, 2013 6:38 am

We could have the quality value be a three digit number or something like that. So a cell might get assigned a number something like 746. 7 for capacity, 4 for internal resistance, and 6 for estimate of overall health... or something like that. That way in a glance you could get an idea of a cells overall performance. Another cell designed for high drain might get a score like 597. Not that great overall capacity but really low internal resistance. Get the idea? What do you think?

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Feature idea: Quality value based on Cell test

Post by Mark » Fri Nov 29, 2013 6:44 am

Sounds a bit like Intel's numbering scheme!

Personally, I think that looking at the actual numbers make sense. For someone that can't understand the raw figures, they're likely to be confused by a system like that and think that a 915 is better than an 898.

Paul.Allen
Site Admin
Posts: 98
Joined: Tue Aug 06, 2013 5:33 am
Location: Utah, United States

Re: Feature idea: Quality value based on Cell test

Post by Paul.Allen » Fri Nov 29, 2013 7:00 am

Mark wrote: likely to think that a 915 is better than an 898.
Good point.

Post Reply