Subject: Re: Reducing contact resistance for low volt use?
Summary: What do you think of these ping times eh?
Reply-To: You can't see me, and I pass right through planets...
References: <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com>
X-Newsreader: Forte Agent 1.91/32.564
Date: Tue, 07 Jan 2003 10:10:24 GMT
NNTP-Posting-Date: Tue, 07 Jan 2003 05:10:24 EST
On Tue, 07 Jan 2003 02:57:28 GMT, jmuchow@SPAMMENOTcamlight.com (John
Muchow) Gave us:
>OMG, all these separate posts are killing me!!!
>>>> What do you think the shunt resistor in your ammeter has a value of?
>>>> What I offered you is the proper way of measuring current.
>>>> A low
>>>>value, high wattage precision resistor in series with the load line
>>>>can be read directly with a voltmeter for current usage.
>That's the way we're plotting our data...but with a 24-bit (0-2.5V)
>data logger....mentioned in one of my previous posts.
Then the resistor IS the load, as well as the pick point, or do you
use a load bank somewhere?
>>>>It is likely
>>>>less resistance than the shunt in your el cheapo meter.
>My what? LOL!
>>>> So, it offers MORE valid data, not less. For a longer period.
>>>>That is called repeatability.
>Who the heck said we were measuring the current with a DVM? Please,
>please, PLEASE READ EVERY POST IN THIS THREAD!!
They have data logging DVMs now, with software, quite cheap. Fully
capable of deriving every bit as much information from the tests as
your overkill sampler, guaranteed. :-]
The need for the resolve just isn't there as the events occur fairly
slowly. Unless you think that you can actually garner information
from failure mode event captures. I doubt it as a picture of the
battery failing yields nothing (the dvm could grab it too anyway).
You are plotting events so long, you could sample once a second, and
get the same data, right up till it avalanches or fails. After ten
percent of that roll-off, it doesn't matter. Especially down at near
zero volts already.
>>>> One more thing. If the "wires" that lead to your motor controller
>>>>are smaller than those in you tester, what "valid data" do you get
>>>>from your test?
>We get data that says for wire gauges that contribute a certain known
>voltage drop at various current levels, here are the current-vs.-time
>discharge curves for various cells....EXACTLY the info we're looking
>for. This is valid data!
For the wire, the tester... some for the battery too. If the data
logger has multiple ports, you could set up a string, and test in
parallel to see if series batteries discharge evenly. (light bulb in
head). A meter could be placed on
>The data we gather will allow the users to select the cell they want
>to use, knowing the cost, availability, and performance under typical
>(for combat robot and R/C users) load. The tests will all be run
>under almost indentical conditions for each cell...critical for
>achieving "valid data". Since we can't simulate the load and wire
>resistances encountered in every robot and R/C machine out there, we
>can only do what everyone else does....test with a known set of
Sure you could "simulate the load". You could construct a
dynamometer using... Oh, my god, no!... a motor. One that you will
not blow. One that is too big for the design, but able to give
stalled motor data, by merely putting a motorcycle disc brake on the
thing's shaft. Hell, turn that bastard backwards. RECHARGE! :-]
>Each R/C place and car builder/drive and combat robot builder will
>have to take the gauge of the wires they are using into account...just
>like ANY user of ANY battery. The manufacturer will test their cells
>under certain conditions (hopefully described thoroughly in the data
>sheets) and can only state that the data is good for these conditions.
>We will be doing nothing other than this.
No. the QA shipping dude at the Mfgr will test outgoing product
samples that way... However...
The qual engineers have labs where batteries get tested to
destruction, you can bet.