Ultra super meter...

Super meter specs: 10 Megawatt input impedance, measures to 1 Gigawatt

The old-fashioned way. It's just a solid-state "VTVM". 1G will be all crammed up on the left side of the resistance scale so you can barely tell it from 1.5G. It's really measuring 1/R (conductance) and displaying it with a nonlinear scale as resistance.

My beat-up old Fluke 8026B handheld meter has 0.1nS resolution on the lowest conductance range, which represents 10G ohms. So you can get

10% resolution measuring a 1G resistance.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

W corresponds to Omega in the alphabets- you will start seeing W all over the place to indicate ohms.

Yeahbut - howinthehell is it going to measure over 1Gohm?

Ed

Such leakage paths are highly over-rated, and in practise don't exist at the 1G-ohm level for clean plastic. Well, maybe if your instrument is covered with a layer of grime and the like. I've also found that finger-prints on 1G resistors don't affect accuracy either, although following conventional wisdom I avoid touching and clean them as well. Now, if you want to talk about 1000G resistors, and fA currents, maybe we can talk. Still, I've found little difficulty in measuring 1000G resistors to a 0.1% accuracy level, which implies insulation exceeding 1000000G for my measurements. It's not at all hard to achieve in air.

Returning to the topic of measuring high-value resistors with cheap instruments, it's surprisingly easy. Start with a 10V power supply, and a standard hand-held multimeter. The meter's 200mV input-range probably has infinite input impedance, so add a measured 10M resistor across its terminals. Now add your 1G resistor, powered from the 10V source, and observe Vx = about 100mV. Measure that and calculate, Rx = 10M * 10V / (10V - Vx), using accurate values for 10M and 10V. Assuming a 0.1% meter, you should be able to measure Rx to ~ 0.25%.

Should you want to measure a say 100G resistor, use a 100V supply and a pre-measured 100M sense resistor. You get the picture.

Returning to the subject of meter insulation, note that the meter terminals in these 1G measurements have 10M across them, not 1G. So even your grimy old grease-covered cast-off too-sucky-for-eBay garage meter would be up to the task. But use a fresh battery!

--
 Thanks,
    - Win

It's the path along the outside of the component that may be lower than a Gohm, and the path across the face of the instrument and the path inside the instrument and the path on the PC board etc. etc. I don't see how the meter shown at the site can do it. And assuming it can, how would it be used? Dump everything into something like flourinert after scrubbing & deionizing & ???

Ed

1G is a relatively low resistance. No special techniques are required, just normal (for analog circuitry) care.

Try measuring between a couple of points on a typical FR4 board (say a blank board, 100mil pitch pads, soldered with RMA flux, cleaned and uncleaned) and see what kind of actual leakage you see at room temperature and with, say, a volt...

If you use the lowest voltage input range of a DMM and know the input resistance, with a voltage source..

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Not too swift; bad precident. If the symbol is not available, then the units should be spelled out.

Maybe by forcing a picovolt and counting how many electrons pass by??

I think it depends on the character set you have installed.

I believe the W occurs when the document was made using the proper Omega symbol ( whichever 'ASCII' code that is ) . If your PC / printer doesn't have a character set active that can display an Omega it displays W instead. That's my understanding anyway.

That's a general problem with having only 256 'characters', many of which were long ago assigned as 'control codes' and many others to useless 'graphics' thingies.

There is a 'new' 16 bit IIRC character set available ( forget its name now ) that includes 'everything' but of course it's not backward compatible.

Graham

Thanks fred, that's correct. Rx = 10M ( 10V/Vx - 1 ).

--
 Thanks,
    - Win

Should be Rx / 10M = Vx / (10V - Vx)

You're calling Vx the voltage across the 10M and Rx the 1G unknown so Rx/10M=(10V-Vx)/Vx ...

Also a bad precedent, but whatever.

If that's true it's broken, especially for electrical symbols. However this happens, it explains the fairly recent mention on rec.audio.pro of a webpage describing a mic that had an impedance of

200 watts.

You're talking about the IBM PC extended character set, which became the de facto 8-bit set (there had been innumerable 8-bit sets in microcomputers and terminals before the IBM PC). ASCII is still 7 bits (the lower 128 characters of most any 8-bit set), and is the standard for Usenet (binaries use a 7-bit encoding). I recall hearing that some Usenet servers strip off the high bit of

8-bit characters.

That would surely be Unicode, it's been around for at least a decade, though I'm not sure what its status is as far as character set standards, I think I heard it was lost in space...

-----

It is very much alive. It is used as UTF-8 (ASCII compatible), UTF-16 (where every character is 16 bits) and UTF-32 (UCS-4).

UTF-8 is used by both Windows and Linux. Java uses UTF-16.

Thomas

That's a different problem than here. It's a web page, so it uses HTML, and in HTML the omega is represented with the string Ω. (That's ampersand Omega semicolon to those reading with a newsreader that automatically does HTML!)

Of course, the web page was probably created using some crappy HTML editor and cut-and-paste from some other document.

Tim.

At the Kernel level, Windows also uses 16 bit Unicode (exclusively). At the user application level, it's used more often than you might guess, as anybody building "enterprise" level applications (requiring ports to multiple languages) is already using it, as is anybody using the .Net languages, Visual BASIC, etc. The largest single chunks of folks still using plain old ASCII, in my estimation, are those building small Windows applications in C/C++.

Nope, they used a W:

---------- Electronic FET Analog VOM 10MW input impedance Resistance measurements from 0.1W to over 1GW 4-12" mirror scale meter Meter can be set for zero center scale Reversing switch and "power on" indicator

-----------

Cheers! Rich

Hell, the source code reads " Resistance measurements from 0.1W to over 1GW" so there was no attempt to create the correct symbol.