Calibrating Volt Meters

Do you mean

1V +- 0.1V thats 10% or 100 +- 0.1V thats 0.1%

martin

Glenn jacobs wrote: > Anyone have any good ideas for acurate volatage sources, with out the need > for standards. My hope is to get accuracies to plus or minus 0.1 Volts. >

Perhaps you can find something here.

I've used the AD588s (not cheap) in projects and discovered my second Fluke 8060 (4 1/2 digit) was out of cal when checking out new boards.

GG

Or zero that's +-0.1 volt.

John

Sorry should have been more clear. Plus or minus 30 Volts DC is the range that I am interested in.

JakeInHartsel

Or just use a random voltage source and an accurate meter to calibrate.

You will definitely need something that qualifies as a standard, although you may not need laboratory grade instruments. For 0.3% accuracy you need a standard about two or three times more accurate, or about 0.1%. Most DMMs are at least that good, but you need to have it certified traceable to NIST to be sure. You can buy 0.1% resistors, and build a divider from 30 VDC to a commonly available standard voltage, such as 1.2 V or 2.5 V, but you will need to make sure it is within the 0.1% accuracy. Then you would need a sensitive galvanometer to determine that the voltage difference between your standard voltage and the divider is zero. In this case you would need about 2 mV resolution on the galvo. This is the old way to do it. A certified DMM is the easiest, but certification costs about $40 or more.

Paul

I like the following voltage references, as i can use a box with 9V batery compartment to run the part which sources or sinks current. No need for a power switch due to the *very* low supply current, and the 9V battery can run down to where it would be useless for anything else.

If you need 30V, then connect 6 of them in series using them as a

2-terminal shunt regulator (Vout and ground). Or if you can stand the burning of more power and a higher supply voltage, use one as the reference to an opamp that has a gain of six, and use common 0.1% resistors (make sure that the supply voltage to the reference is limited to spec).

I believe that a well charged car battery is a quite accurate voltage source, at about 12.68 volts or so. Am I mistaken?

i

Yes.

It's barely accurate to a half volt.

I recently build a cheap +/- 10V reference using a 0.1% reference from National, some 0.1% resistors and AD688 instrumentation amplifiers as a buffer.

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

a fresh carbon cell was used in the old days. i don't remember the the exact voltage but something like 1.567 etc.

they do make reference diodes.

--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5

Yes

A mercury battery like PX675 (though may be kind of hard to find) loaded with 1.5K resistor(1mA load) will give, if memory serves me, 1.348V. Very precise. Four significative digits.

Tom

It isn't temperature dependent?

Good point! Here I'm not quite sure but the temperature only affects the battery capacity, not its voltage. I'm gonna investigate further. On the other hand, my suggestion is not really a good one because production of mercury cells has almost stopped.

Take that mercury battery, put a 5 1/2 digit meter on it and now you have an accurate voltage source.

*ALMOST* ???? AFAIK they have not been made for at least 10 years...

Have you looked on Google?

Several places selling them on the first page: Results 1 - 100 of about 3,630,000 English pages for mercury batteries. (0.30 seconds)

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

The # of google hits is about the crappiest form of statistical reporting imaginable. A site with an eveready ad and an article about mercury poisoning will be a hit. A site saying that mercury batteries should be banned would be a hit.

You didn't even put the search phrase in quotes. I got 174000 hits for "Terrell eats lead". You gotta quit eating lead!