4
\$\begingroup\$

I am using this this desktop Multimeter - 6 and half multimeter

Whenever I measure a voltage or current in the auto mode, the last three digits continues to decrease and keeps on decreasing till the fourth digit also decreases.

For example, if I measure a diode drop to be +0.805919V at time t=0, it goes from that value to +0.805900V... +0.805851V....+805230V........+0.804780.. and goes so on. This happens around 5minutes.

Why does this happen?

I am measuring at only +25degC

\$\endgroup\$
  • 7
    \$\begingroup\$ A 6 1/2 digit multimeter will show even the smallest change. The voltage across a diode is temperature sensitive, the multimeter applies a current to the diode so the diode will heat up slightly which changes its forward voltage (typically by 2 mV / degree C). Are the other voltages that you measure really that stable? Few devices generate a voltage that remains stable in the micro-volt range. Suggestion: make the multimeter display less digits like 3 1/2 digits. Few applications need more than that. \$\endgroup\$ – Bimpelrekkie Oct 12 at 7:26
7
\$\begingroup\$

Note 9 on page 12 of the manual states:

Specifications are for the voltage measured at the input terminals. The 1 mA test current is typical. Variation in the current source will create some variation in the voltage drop across a diode junction.

enter image description here

Figure 1. The meters in question. Image taken from the manual.

I think you're seeing two issues:

  1. While the meter is 6.5 digit, the current source stability is not. It will probably drift a little due to thermal effects.
  2. 1 mA and a diode Vf of 0.7 V gives a power dissipation of 0.7 mW in the diode's junction. This will cause a little heating and this will cause the Vf to drop. At some point the diode will reach equilibrium when the heat lost to ambient equals the power input.

You could check which is the larger effect by allowing the temperature to stabilise for some time before connecting the diode.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Thank you for the answer. How to take a proper accurate reading in these circumstances? \$\endgroup\$ – Newbie Oct 12 at 10:13
  • 4
    \$\begingroup\$ What's the point? Are you going to temperature control the diode in use? Are you going to calibrate each diode before soldering it into circuit? (They will vary within the same batch.) Will soldering affect the diode and change the reading? Why do you need to know the value to 0.1% at all? What problem are you trying to solve? \$\endgroup\$ – Transistor Oct 12 at 10:20
  • \$\begingroup\$ Hold the diode temperature constant. Watercooling it may help (though you can cool the case, so you want a diode with a good thermal connection to the case. SOT23 probably, or one of those old fashioned metal cased rectifier diodes. \$\endgroup\$ – Brian Drummond Oct 12 at 10:36
  • 7
    \$\begingroup\$ "I am just trying to get the value of the forward voltage of the diode at various input voltages." You have something wrong there. You're applying a test current which is not adjustable. The diode will determine the voltage at 1 mA test current. You only get a single point on the I-V curve. What is the overall objective of the exercise? \$\endgroup\$ – Transistor Oct 12 at 11:52
  • 1
    \$\begingroup\$ That information is missing from your question. Everyone will assume that you are using the diode test function and the internal 1 mA current source. If you're powering from 12 V to 40 V then you have some other circuitry to limit the current. Again, this is missing from your question so we have no way of knowing what it is, how stable it is and how it might change with temperature. \$\endgroup\$ – Transistor Oct 13 at 6:31
1
\$\begingroup\$

6.5 digits are enough to see some otherwise subtle effects:

  1. Unintentional thermocouples between different metals in your setup. You do touch them by fingers, don't you? You fingers are hotter than the environment and the couples gradually cool down for a while.

  2. A semiconductor voltage references drifting away because of self-heating and even aging.

  3. Load of the electrical grid changing and influencing even a good stabilized power supply.

  4. The multimeter itself still heating up and slowly stabilizing. A lot of measurement tools are reaching their advertised accuracy only after a prolonged start-up period (say, 30 minutes). See the manual.

If you are doing some serious stuff, you have to account for all these things (and some more).

| improve this answer | |
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.