Panel meter problem

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: David Speck <dave@xxxxxxxxxxxxxxxx>

Adam,

You didn't mention of your panel meter has a compressed scale at the 
lower end.  If it does, then it is a pretty common rectifier type 
meter (It's a lot cheaper to make a basic DC movement and feed it 
with rectified AC  than one that works directly of AC.)

The only problem with a rectifier type meter is that it shows a zero 
reading until the voltage across its input terminals exceeds about 
1.2 volts, when the diode bridge inside starts to conduct.
There do exist premium AC meters for voltage and current that have a 
linear scale (usually starting at about $50 on eBay, as opposed to 
$5-10 for a compressed scale meter).  Inside, they have a network of 
resistors and Zener diodes that allows them to respond with near 
perfect linearity.  If one of these linearizing diodes is blown, then 
the response will be decidedly non-linear.
To be honest, you meter sounds about as good as can be expected.  As 
previously mentioned, most of the best panel meters are rated to 5% 
accuracy at full scale.  10% is more common.  Digital panel meters 
off eBay are usually closer to two percent, but can be driven crazy 
if not totally blown out in a TC environment.  Most coilers stick 
with moving coil analog meters for TC work.
Try eBay searches for "Yokogawa", "Crompton", "GE meter", 
"Westinghouse meter",  "KA-241", or "KX-241" to find the nice 270 
degree switchgear panel meters that give somewhat better resolution 
than the 90 degree movements.  A lot of them come up with regularity, 
so don't fall for the ones with the $60 - $200 "Buy it now" 
prices.  If you're lucky, you can get them for $10-25 each.
HTH,
Dave