 |
February 24, 2014, 08:26:03 AM |
|
Just to prevent time-wasting: I can't spend more than $1000 on this (either as BTC or items off Amazon shipped to US). That'd stretch my budget to its painful maximum. If spending that little is a pipe dream, please let me know.
I've been wanting to get back into testing batteries for a while. ... I've had a minor obsession with it for years. The end-goal is to create a consumers' guide site posting collected data, though I'm sure companies and hobbyists would really appreciate an easy-to-navigate page covering different sizes and designs with experimental data rather than the not-fact-checked advertisements manufacturers push. I was using a cheap DC wattmeter and a lab's ancient analog voltmeter to double-check data (it always failed to match, and voltmeter was calibrated properly while wattmeter cannot be calibrated, so...), with the battery connected to a case fan. It's an inadequate solution, obviously. I can't find a satisfying solution already-made. They're either too imprecise, don't log data automatically, don't log enough data, and/or have too small of an operating range.
I'm looking to measure the following:
*Watt-hour consumption over time
*Amp-hour consumption over time
*Voltage output (under load)
*Amperage output
*Wattage output
*Time running
Device must meet the following operating specs:
*Precise within .02V, .005A (wiggle room with higher output)
*Input range: .25V, .01A - 45V, 5A
*Able to adjust current supplied to "output" device within input range (must be able to step down voltage and decrease amperage supplied - no need to step up)
*"Time running" able to be started, stopped, and paused
*Polls, at most, once per three seconds -- however often it polls, it needs to poll at exactly that time with virtually no latency or discrepancies.
*"Set it and forget it" data/power USB connection for PC (data could be stored internally and later extracted via USB, or output directly to PC storage device). It can output the polled numbers into a shell-less file so long as all recorded numbers are separated and where each measurement is separated into different rows (I just need to be able to copy-paste into an Excel sheet without much fussing)... so like:
[Wh] 000.100 000.195 000.290
[Ah] 0000.0500 0000.0980 0000.1452
["VOUL"] 0000.0050 0000.0049 0000.0049
[A] 0001.0000 0000.9987 0000.9968
[W] 0000.5000 0000.4986 0000.4976
[time in seconds] 000003 000006 000009
^would be fine
Stretch goals:
*Device able to internally "consume" the energy, outputting it safely as heat and perhaps measuring it more accurately... heat cannot interfere with measurements!
*LCD display outputting current watt throughput
*Measure temperature of battery at terminals, ambient temperature, and relevant components inside the measuring device
*Ability to be calibrated easily by my stupid, aloof monkey brain without access to maintained lab equipment like adjustable DC power supplies (I could at least check recorded DC voltage of PC power supply lines against what the motherboard claims.... no idea how accurate they are)
|
