John Larkin a écrit :
On Tue, 01 May 2007 13:33:20 +0200, Fred Bartoli
r_AndThisToo wrote:
John Larkin a écrit :
On Sat, 28 Apr 2007 22:01:29 -0700, Don Bowey
wrote:
On 4/28/07 6:26 PM, in article ,
"John Popelish" wrote:
SuperM wrote:
The DEFINITION of the VOLT, and the AMPERE DEFINE what one OHM MUST
be.
I agree. An ohm is defined as a volt per ampere.
Linear resistance (constant ratio of voltage to current,
regardless of the magnitude of either) is not an actual
property of anything in nature, but an idealized concept
that is approximated to varying degrees by various real
materials.
Practically speaking, what is the typical error of trusting Ohm's Law to be
accurate in General electronics? Much ado about nothing.
If you're building audio amps, very little. If you're designing
thermocouple conditioners, or high-voltage power supplies, or
picosecond electronics, the non-Ohms-law behavior of real resistors
may be *the* limit on how good your stuff can be.
Just had the issue with a batch of RNX 3/8 1G 1% Dale/Vishay resistors.
They are factory adjusted to be 1G at 100V (that's a huge 10uW dissipation).
Under a few volts I measure them between 4% and 6% over the value.
At 100V they're just all spot on.
Not even the same variation among all the batch.
Yeah, Ohm law :-)
A 5% change over 100 volts? That's awesome. I wouldn't have expected
that at 1000 volts.
Surprising, isn't it?
As my app didn't involve high voltage I didn't check any further, but
I'll do this when I have time, just out of curiosity.
OTOH, I'm now testing some Caddok 1:1000 high voltage dividers for a
precision high voltage application. They're spec'd at 0.02ppm/V and seem
OK but measuring this isn't straightforward since most meters don't go
over 1kV and they are as suspect as the resistor you test.
The only thing you can rely on is repeatability and you have to devise a
set up that only needs this.
--
Thanks,
Fred.