test1.x-series: This post is about discharge tests with my
10440's.
• Parameters, tentative:
- starting voltage: fully charged, 4.200V+ offline
- program: "DCHG 00 NOR 100W", i.e. Discharge Test
- discharge rate: "00.20A", i.e -200mA
- target voltage: "02.80V", i.e. 2.80V
• my historic
reference values from genuine Imax B6, with 3.00V cut-off voltage and -100mA discharge rate:
grey protected UF10440, marked cell: 254mAh (repeated tests: 261mAh, 260mAh), or more typically: 245mAh
grey protected UF10440, unmarked: 225mAh (repeated test: 229mAh)
blue unprotected UF10440, marked cell: 334mAh (repeated test: 338mAh)
blue unprotected UF10440, unmarked: 278mAh (repeated test: 282mAh)
• my EBC-A results:
test1.1: grey protected UF10440, marked cell: "0240mAh"@-200mA/2.80V
test1.2: grey protected UF10440, unmarked: "0199mAh"@-200mA/2.80V
test1.3: grey protected UF10440, marked cell: "0253mAh"@-100mA/3.00V
test1.4: grey protected UF10440, unmarked: "0212mAh"@-100mA/3.00V
test1.5: (test1.3 repeated): "0258mAh"@-100mA/3.00V
test1.6: (test1.4 repeated): "0212mAh"@-100mA/3.00V
test1.7: (test1.3 repeated again): "0256mAh"@-100mA/3.00V
test1.8: blue unprotected UF10440, marked cell: "0345mAh"@-100mA/3.00V (starting from 4.21V)
test1.9: blue unprotected UF10440, unmarked: "0246mAh"@-100mA/3.00V
test1.10: (test1.8 repeated): "0328mAh"@-100mA/3.00V (starting from 4.17V)
test1.11: (test1.9 repeated): "0250mAh"@-100mA/3.00V
test1.12: (test1.3 repeated and logged): "0258mAh"@-100mA/3.00V
• comments on the above tests:
test1.1: that's lower than Imax B6 result (~240 < ~260). Given that the cell is exactly 24 months old and discharging at double the rate, the number is plausible. promising result!
test1.2: shows consistency because Imax B6 result was lower too! promising.
test1.3: to get directly comparable data, let's adopt Imax B6 parameters. lol, missed the Imax B6 reading by 1mAh (253 vs. 254), fantastic result. promising!!
test1.4: sounds plausible (212 vs. ~227) because of cell abuse and capacity degradation. in all fairness for sure we're going to repeat test1.3&1.4 because we did the same 2yrs ago with Imax B6 tests. no doubt the Imax B6 data isn't the best reference simply
because it is 2yrs old data. so when done with the 10440's tests, our next post will be on new Eneloop AAA for which we have a fresh reference: C9000 data!!
test1.5: higher than test1.2 (258 vs. 253) but definitely in the range of the Imax B6 data, i am impressed. cell degradation over the 2yrs should have been higher but maybe this is simply a good cell! test1.6 will be interesting now.
test1.6: un****ingbelievable, it's the same result as test1.4!! this is the first real indication of accuracy (or say, consistency) of the device. it would also mean, in theory, that in future i would
not have to repeat tests
again to ensure data integrity. let's repeat test1.3 (=test1.5) again for a final word on this preliminary promising result.
test1.7: totally plausible (256 vs. 258). i am really beginning to like the performance of this thing!! from this test result it becomes clear that in practice i could have saved the repeated tests because the device returns the same mAh-reading for the identical cell. we will use this knowledge in future tests where we
don't repeat the identical test for each and every cell. next up, let's
try to measure the max capacity of my pair of unprotected blue UF10440's. That's a true challenge because the cells
are abused and only 1 of my chargers, DC1, is able to get them to 4.200V after repeated re-insertions; other li-ion chargers terminate these poor cells at 4.11V or so. The cells suffer from superhigh self-discharge rate at the 4.20V level. Since the cells are so poor, it wouldn't make sense to measure their max capacity; on the other hand, since we have Imax B6 reference data, i am curious to learn how far off the capacities are after 2yrs of abuse in power LED flashrights! Depending on the result, i might repeat the test1.8, but i guess not. We should proceed to Eneloop tests soon imho.
test1.8: wow, this is totally unexpected, the value is higher than Imax B6 (345 vs. ~336). instead of repeating the test, for now let's go over to test1.9, that is the other blue 10440 and see what comes out! Something's off here. Not by much, but let's note it.
test1.9: looks plausible enough (246 vs. ~280), capacity loss after 2yrs of abuse. for the sake of completeness, i'll repeat the test in test1.11
test1.10: i could hardly get the voltage to the 4.20V target. value is lower than test1.8 (328 vs. 345), oh well. it's an abused cell and not suitable for reference purposes.
test1.11: wow, fantastic result (250 vs. 246). this really confirms my conclusion from test1.6: if one is able to create identical test start conditions (which will be a challenge with Eneloops!), then a
single test run is needed to get a fairly accurate and consistent mAh-capacity reading for the cell under inspection. in theory it'd be a waste of time (from now on) to repeat identical tests. I had 4pcs of 10440's. In the above series of test runs i tested each cell at least twice in order to be able to draw some conclusions on charger's tendency with respect to consistency and accuracy. I am done with testing 10440's and up to this point, we already have a good picture of this device's performance.
test1.12: meanwhile i figured out how to install the driver and get the PC software working. using neodymium was never my preference, so i tried a diy battery holder and redid test1.3 in order to see if the changed electrical resistance (metal contact plates instead of neodymium) altered the test result. see attached screenshot, i got the same value again (258 vs. 253/258/256), absolutely fantastic! this makes my life of testing small cells much easier because a battery holder holds the cell in place! For big cells, e.g. 18650's, the diy battery holder isn't too comfortable and involves some fiddling, so i might still revert to neodymium for them. In any case, i am confident now of EBC-A's performance (i.e. accuracy and consistency) and ready to proceed to the more challenging (or say time-consuming) tests: Eneloop!!
• Summary:
The EBC-A tends to deliver amazingly consistent mAh-readings, as consistent as the Imax B6 readings. This can only be possible, if the device is also accurate. Very accurate (for the price paid; it costs half the price of a genuine Imax B6!). Our next step to knowledge and wisdom is to compare the reading to an
absolute mAh-reference scale. For this, we take new (and also old) 3rd gen Sanyo Eneloop AAA/AA (rated nominal 800/2000mAh) and their mAh-readings off the Maha MH-C9000. I already know the 'C9000-capacities' of these Eneloop test cells and the exact conditions under which i produced them, so i could comment on the EBC-A readings right away and interpret them with confidence.
So far, my C9000 was the only reliable battery analyzer in my household (i had given away my Imax B6!). My hope is that i can replace it by the EBC-A as battery analyzing tool because C9000 can analyze NiMH chemistry only and it accepts only NiMH cells with low internal resistance. My personal C9000 unit is very consistent and very accurate, i established this as fact already. Am looking forward to the Eneloop tests from tomorrow on!!
My motivation is: Knowing the exact real actual capacity of cells helps me (the tester and experimenter) to evaluate the quality or assess the performance of
other products with battery analyzing feature, such as the forthcoming dreamchargers. The point is the question: Can the inexpensive battery analyzer EBC-A replace expensive proven high-quality analyzers such as
- genuine Imax B6
- Jun-si iCharger
- West Mountain Radio CBA IV
- Handheld Scientific BA-1
- Schulze chargers
- a. o.
We'll find out. Next up are Eneloop tests!
