Weird Multimeter Behavior???

Hello,

Is it normal for a multimeter when it's set to measure voltage and it's probes not touching anything to show a floating reading of fractions of volts?

This happens with Agilent U1253B.

Any idea?

Yes. It's a floating input.

Perhaps, if it is set to read AC at a low voltage -- there is 60Hz (or 50Hz) EMF (and other forms of RF) that is picked up by the probes (acting as antennas).

Try shorting the probes together and see if the reading settles down at or near 0.00[00...]. If not, there might be something wrong with the meter.

What is your meter set to? Or, on what setting(s) do you see this effect?

There is a vast amount of electrical energy (electromagnetic radiation) in the air, from radio station and mobile phone transmissions, and radiated energy from all the mains cables and electrical devices in your house.

The multimeter leads act like a radio aerial, picking up this energy and producing a reading on the meter.
You will see this effect even more clearly with an oscilloscope.

Rarely seen now, there are simple radio receivers, called "crystal radios", that are powered purely by the radio signal transmitted from the station But they only work on AM signals, which are all being switched off in this digital age.

As Steve says, this effect should normally only be seen on low AC voltage ranges, and touching the probes together shorts out this energy by putting the two terminals at the same electrical potential, so an absolute zero reading would then be normal.

It's set to read DC voltage V. But the numbers go up and down very unstable reading until I probe some voltage.

But even with the atmospheric noise around and with the high internal impedance of the meter, it should be stable at least showing something close to 0 value.

You sound like someone who, like me, was first exposed to analog meters. I have one meter that annoys me as it is difficult to see what range you are actually measuring in. It is very easy to misinterpret as the range indicator is only a series of dots that are very difficult to see and almost impossible if you are in direct sunlight. It makes it very easy to misinterpret 2 volts for 20 volts. If you want to really want to see it jump around move a magnet within a foot or two of the leads. I have even seen stray indications with the leads touching. If you see that, then try twisting the leads gently together and touching them. That should cancel out most stray voltage indications. You may have a source of EMI in your area you are unaware of and you must remember that your meter is sensitive to one millionth of a volt. Personally I would like to see a range switch that could lock the meter in a one tenth volt range. Reading one tenth volt is accurate enough for 95% of the measurements I take. I know others will disagree but they might like the ability lock the range in another scale. There may be a meter out there that is already capable of this but none of the four I own do.

This is interesting isnt it ?
http://www.brianhoskins.co.uk/website/agilent-u1253b-input-impedance-problem/

Yup I'm new to the DMM world.

Actually it's. How bad we feel when we pay a lot for something that's supposed to be close to perfection or at least bug free, and then realize it's just over-hyped and lied about. The problem with the support we got which sucks, the printed manual that comes with this super expensive bastard is low quality, no colors, very thin, and printed in Asia!

Lesson learned!

And, what happens when you short the probes together?

If your meter is a moving coil type (not indicated) then the problem is RF from a nearby source being picked up by the meter leads and rectified by the meter reverse protection diode. Otherwise on a digital meter, if your leads are floating your meter could be reading the static buildup on the leads. I have done experiments to prove this by holding on to one meter probe metal tip and swing the other rapidly in the air so you can hear the wind from it. You will see on the display tens of millivolts.. The other explanations I have seen here are equally valid i.e meter turned of RF in the semiconductors in the meter etc.

Ops. Should have read at the end. meter turned on to AC range or RF getting into the semiconductors within the meter etc

The floating unstable reading goes from +-mV to a +-1/10 V.
When the pobes are shorted on DC voltage it shows 0, and on AC it still shows 6 to 7 mV. When it's set to current A, and shorted or not, it shows ~1/10 mA and depends on how I move the probes it reads 0.

Is your meter (or leads) near a transformer or other source of EMF/RF/MMF?

It's in a closed room with wireless modem and a computer. I'm not sure how much close is considered close in this situation. But it's within the room space, 6 - 10 feet.

Winner Winner, Chicken Dinner! Try turning the wireless modem and computer off and see if it helps :~) I bet it does. Especially the modem (transmitter)!

The fact that your meter is reading something other than 0 may be a tribute to how good it is. It's sensitive enough to pick up the stay voltages. Readings in the mv range are typical, and as long as it goes away when the probes are shorted no prob. The readings on the AC scale with the probes shorted might be due to RF and the probes behaving like a loop antenna. Try bringing the wires together (while shorting the probes together) so they are parallel to each other and in close proximity (and even try twisting them together loosely and see if that reduces the reading.

Readinga on the *ma* scale with the probes open is a bit puzzling, but might be explained by standing quarter waves on the probes. At the meter end of the probe would be the high-current section of the antenna [but, the meter would have to be able to read fairly high frequencies for this to be the case]. Again, try twisting the probes together, but NOT shorting them and see if that goes away. Also, try removing the probes and see it that makes a difference. [but if it's on the _DC_ ma scale, then quarter-wave antenna effect is probably not the answer]

If the DC ma readings persist during all of the above experiments, then perhaps the meter is not adequately shielded.

BTW: All of my meters indicate voltage on the AC range. My cheaper meters go as high as 1 volt and even 2 volts. My more expensive Protek meter shows around 10-20mv. All go to zero when the probes are shorted.

So if it's not adequately shielded would it be a problem in the design of this specific multimeter model, or due to a defect/damaged equipment that I should take back to the store?

Thanks a lot.

Hi Rodney

Answer easily revealed he quoted the model number "Agilent U1253B. "

A search on the net reveals address as follows
http://www.home.agilent.com/en/pd-1765026-pn-...t-with-organic-led-display-oled?&cc=GB&lc=eng

So its a DVM OLED "organic light emmiting diodes "

FP meaning floating point , IEEE standards Floating point math’s springs to mind here especially when dealing with micro's for measurement

This might help further in understanding how FP or floating point numbers are represented in digital micro controllers also on early mobile phones example the Sony Erickson T610 upwards before smart phones arrived which required extra libraries and classes to be written so that we could use J2ME for scientific calculators

Link as follows

Should help you understand a lot more about DVM “Digital Volt meter”

http://steve.hollasch.net/cgindex/coding/ieeefloat.html

Here is some of the maths behind this which will help explain in more depth also a good explanation

http://kipirvine.com/asm/workbook/floating_tut.htm

That should help a few of you as well


If you want more information on what OLEDS are here a DIY how to make your own Oled "Interesting article this because it will change the way that glass windows might be manufactured in the future to incorperate OLED technology " Example the future of shop signs

http://education.mrsec.wisc.edu/Edetc/nanolab/oLED/index.html

Terrible I agree wait till you see some of the schematics which you cant even read without so much as component marking or what it is never mind readability

That’s another subject all on its own if your luckey to even get such informations out of them which is normally diabolical