1. Home
  2. Electronics Repair
  3. improvised voltage reference

improvised voltage reference

Does anybody have a weblink for the temperature coefficient of voltage for commercial (Energizer) size 357 silver oxide-zinc button cells? Nominal is 1.55 volts, but knowing the temp effect is (maybe) important if I'm stacking up nine or ten of them.

I'd like to check a trio of inexpensive Radio Shack DVMs for use in charging relatively-expensive lithium iron phosphate batteries. On the same battery the three meters report 14.41, 14.51 and 14.58 volts, so they certainly disgree among themselves. I suspect but can't yet prove that the middle one is about 2% higher than actual. Yes, I probably should buy a good DVM, but the Radio Shack units are on-hand and except for charging batteries absolute accuracy isn't usually of much importance.

Ten millivolts is plenty good for present purposes. If there's a better/cheaper improvised reference in this range please post! CR2032 cells are readily available, but voltage details are scarce.

Thanks for reading,

bob prohaska

Reply to
bob prohaska
Loading thread data ...

Eveready states in their literature that the "open" voltage (unloaded) of such a cell is 1.5 to 1.6 volts - as a matter of the specific device in question. Suggesting that the loaded voltage will be lower. They state, typically, at 1.55V for a 1.6V battery. There is also a temperature chart in their literature, but it is applied only to storage and longevity-in-storage.

9 x 1.6 = 14.4, so you are definitely within the range even with the lowest reading given. Typical accuracy (DC) on an RS meter (today) is +/- 0.5% - as advertised. Meaning some may be better, but should not be worse. As yours are likely older models, +/- 0.5% is likely optimistic. 0.5% x 14.4 = 0.072 Meaning that your range-of-readings should be somewhere between 14.472 and 14.328 for that particular group of nine (9) button cells.

Oh, RIGHT! You are comparing apples (liFe) cells to oranges (ZnAgO). Simple enough: Go to the Manufacturer and get the nominal unloaded and loaded values. Measure with the meters-in-hand. That which is the closest is the benchmark. If the others are consistent, there is your margin-of-error.

Peter Wieck Melrose Park, PA

Reply to
Peter W.

10mV on a single 3.5V cell is 0.3% - so you're going to have to calibrate SOMETHING, if you intend to provide that accuracy. One calibrated reference is enough - you can set up cheap meters to mimic it and calibrate microcontrollers to store correction factors using it.

Using batteries as a reference is no longer a viable method for most labs to maintain.

Get one good 6digit calibrated meter. If it's built into a data logger, even better.

RL

Reply to
legg

You want a precision that is doomed by the technique you intend to use - stacking cells. Instead, build a voltage reference using a single IC: ADR01 available from Digikey

formatting link
provides 10V output at +/- point one percent (.1%)

Calibrate your meters to that.

Ed

Reply to
ehsjr

Bad idea. If a silver oxide battery goes below 30% charge, there will be about a 100mv step increase in terminal voltage:

formatting link
formatting link

I've seen the step in my cheap Chinese digital calipers, where I need to re-zero the indicator when the cell voltage crosses the transition. Of course, the alkaline batteries are somewhat worse, where I need to re-zero the calipers nearly every time I use them.

Reply to
Jeff Liebermann

Oops. That should be "goes below 70% charge". I'm accustomed to capacity being state of charge (SoC), not state of discharge.

Reply to
Jeff Liebermann

There are ready to use voltage reference PCB's available.

formatting link
formatting link
formatting link
formatting link
formatting link
formatting link
etc...

Reply to
Jeff Liebermann

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.