Fluke resistance reading abnormally low

[dedonaldson]

I have a Fluke 87 V True RMS Multimeter. I did buy it used and believed it to have been working properly. Lately however, while inspecting a PCB, I have become suspect of the resistance readings. I also own a Proster BM4070 LCR Meter.
To test my theory I measured 30 resistors ranging from 10  to 1 M using both meters. I prepared a spreadsheet of the results (below) and was very surprised to find that in the range of 51k to 1M, the Fluke had an average -23% accuracy! The Fluke average for all measurements was -7%. By comparison, the Proster LCR meter was an average of <1% across the range.
I did also note that when place in ohms mode, the Fluke shows a reading of .857 M without probes touching and .3 when touching. The .3 seems normal but I’m not certain about the .857 because I never paid attention to it until my suspicion about accuracy arose.

Value   Fluke   % Diff   Proster   % Diff
10   R   10.0              0%   10.0             0%
22   R   22.0             0%   22.0             0%
47   R   47.0              0%   47.0             0%
100   R   99.0             -1%   99.0            -1%
150   R   150.0      0%   150.0     0%
200   R   198.0     -1%   200.0     0%
220   R   220.0      0%   220.0     0%
270   R   268.0     -1%   268.0     -1%
330   R   328.0     -1%   329.0     0%
470   R   469.0      0%   471.0     0%
510   R   505.0     -1%   509.0     0%
680   R   677.0      0%   680.0     0%
1   K   1.0             -1%   1.0            -1%
2   K   1.9             -5%   1.9            -5%
2.2   K   2.2              0%   2.1            -5%
3.3   K   3.3              0%   3.3             0%
4.7   K   4.7              0%   4.7             0%
5.1   K   5.0             -2%   5.0            -2%
6.8   K   6.7             -1%   6.7            -1%
10   K   9.8             -2%   10.0             0%
20   K   19.6             -2%   20.0             0%
47   K   45.0             -4%   47.6             1%
51   K   48.0             -6%   51.6             1%
68   K   62.4             -8%   67.8             0%
100   K   91.0             -9%   101.0     1%
220   K   176.0   -20%   220.0     0%
300   K   225.0   -25%   300.0     0%
470   K   318.0   -32%   470.0     0%
680   K   382.0   -44%   680.0     0%
1   M   0.5           -54%   1.0             0%
AVERAGE                     -7%                0%
AVG 47k-1M           -23%                0%

1.   Why the aberrant low readings?
2.   Does the Fluke need a recalibration/repair?

Sorry for the table formatting but the spreadsheet did not transfer over too well.

Since I am relatively new to the electronics field and am an old geezer, any help should be pretty basic in presentation.

Thanks to all for any assistance.

[Martin72]

Hi,

I did also note that when place in ohms mode, the Fluke shows a reading of .857 M without probes touching

This is suspicious, with an open input the fluke should display "OL", nothing else.

[TimFox]

A quick check of some readings with large resistors indicates that this unwanted 0.857 megohm seems to be in parallel with the external resistance.

[2N3055]

As you requested, simple:

On inputs, Fluke has certain components to protect it from overvoltage.
They can be damaged and can fail in a way to start leaking current.
It presents as if you have something already connected in Ohm mode.

Or if batteries leaked, that can also create problems.

I guess it is repair time.

In USA Fluke might offer Lifetime warranty. You should check with them before doing anything hasty...

[Caliaxy]

You probably know this, but just to make sure: you don’t touch the metal part of the leads with your bare hands while measuring resistors, do you?

Doing so would put your body resistance in parallel to the measured resistor, resulting in lower readings in the ranges you are concerned about.