With the Whizbang giving true end amps capability to the Kid and Classic charge controllers, it would be nice to see a new charge termination feature that is consistent with US Battery's recommendation (http://www.usbattery.com/usb_images/charging_instruction_2011_3.pdf) of terminating the absorption charge two to three hours after the return amps drops to 3% of the battery's C/20 rating. It seems to me that this is a significant charging improvement over a timed absorption or terminating absorption on end amps alone.
Another way to read that is to say there is a minimum Absorb time of 2 - 4 hrs and/or go to float after reaching less than 3% . Just doesn't quite yield the same result...
I'm thinking that the algorithm might be a bit complex, since there are so many chemistry's for FLA's.
Quote from: Kent0 on February 02, 2014, 12:24:26 PM
it would be nice to see a new charge termination feature that is consistent with US Battery's recommendation (http://www.usbattery.com/usb_images/charging_instruction_2011_3.pdf) of terminating the absorption charge two to three hours after the return amps drops to 3% of the battery's C/20 rating.
You can sort of almost do that. What is the current into your US Batteries 2 or 3 hours after the 3% point? On my interstate batteries (made by US Battery) I can get the return amps down to about 1%.
If your batteries behave that way, just set the Classic to terminate absorb at the 1% point.
--vtMaps
I totally agree with this idea. But do it not every day, but every so often as programmed by the
Skip days of Bulk/Absorb OR have it do a Bulk/Absorb at a programmed state of charge.
Re-Bulk on SOC% I mean... That's coming.
Ending absorb on Ending Amps alone does not guarantee that you get an Anti Sulfation Stage Full (A.S.S.F.)
charge that you would get if you run the absorb for a timed amount. Ending amps in addition to time sounds
like a great idea too. Time alone ~should~ do this as long as you know how long at least to have it
go for. But this way, you don't have to go extra time and wear the batteries out faster and have to
add water as much maybe.
boB
The thing that I like about the time delay after 3% ending amps algorithm is that it takes the guess work out of the decision about the absorption time. I think US battery has settled on this strategy as a way to make sure that the batteries have adequate time to get fully charged and still make sure that the charging process terminates.
Kent, have you monitored a charge cycle to see just how low an input Amperage is achieved in that 2 to 3 hr period? Just curious.
Im not sure that you can rely on WB data getting too close to the long end of the EA taper, because theres quite a bit of jitter. eg see todays curve (remember its upside down).
I do have a customer that has a system that can be remotely monitored with a Link 10. This file shows a typical charging situation. The absorption time needed seems to change a lot with temperature as well as the battery age.
(//)
What amount of PV input is that system getting and how big are the loads over the day?
Looks like there is some pretty heavy use... estimate ~7 Kwh at 12v?
That graph is from a transmitter site. The load is 60 to 65 watts continuously (maybe a little more than 120 amphours per day). Notice that the battery current is about -5A after the generator shuts down.
There is a 1000-watt PV array that provides nearly all of the charging. But occasionally the owner has to run up there and charge the batteries with the generator - PV modules don't work very well with snow on them.
HMMM, you are having to put in, bulk phase, ~ 12v x 75A avg (900W) for ~ 5.5 hrs so ~4.9Kwh. divided by 24 hr = ~205 w/hr...??
Saw that ~-5 and assumed it was the CC load....
Quote from: zoneblue on February 03, 2014, 06:42:33 PM
Im not sure that you can rely on WB data getting too close to the long end of the EA taper, because theres quite a bit of jitter. eg see todays curve (remember its upside down).
You may try this firmware it has more averaging on the WBjr current http://www.fusion.midnitesolar.com/MidNiteSolarUpdate_1756_2-4-2014_v4.12.exe (http://www.fusion.midnitesolar.com/MidNiteSolarUpdate_1756_2-4-2014_v4.12.exe)
Thanks, will give it a whirl on the weekend. Or maybe therell be another by then!
Quote from: zoneblue on February 04, 2014, 01:30:00 PM
Thanks, will give it a whirl on the weekend. Or maybe therell be another by then!
This has as much averaging time as the normal battery amps readout.
I can make it longer if needed but it ~should~ stop the spikes you were showing...
However, I don't know what your sample time is on your graph or exactly how you were
acquiring that data.
boB
Im sampling there at 1 second, on the second, system time. Bob, for WB, i reckon 1sec is just fine! Im not concerned about those spikes unless they were a bug.
The benefit of a 1s sample rate is that it matchs the 1s averaging for many of the datapoints in the classic. That way math operations for derived datapoints from the source data have more integrity.