Keysight U1272A V DC offset error

[Caliaxy]

My Keysight U1272A started behaving strangely.

All DC voltage measurements have an offset error of about 1.6 V (this happens in “normal”, i.e. non-relative mode). When no voltage is applied (or when the leads are shorted), it displays about 1.6 V. When attempting to measure any voltage (say, a 1.5V battery), it displays that voltage + the offset error (3.1 V, in this case). The offset error can be compensated in “Null” (relative) mode. All the other modes (DC mV, AC V, DC mA, AC mA, R etc.) seem to work fine.

Resetting all the settings didn’t help.

I tried to store a different reference value (by pressing “Null” while applying an arbitrary voltage). The Null mode still seems to work as it should (it subtracts the new reference from the measured value), but the offset error in “Normal” V DC measurements changed as well (now is about 5.3 V).

Has anybody encountered a similar issue??

[lowimpedance]

 I have not encountered this with my U1277a.
What firmware version is installed ?.

[wraper]

does it change when selecting ranges manually?

[coromonadalix]

Could a calibration correct this behaviour ?

[Caliaxy]

does it change when selecting ranges manually?

Yes, it does. The reading is higher as you go up the scale:

Reading - Scale:
2.5V on 3V
3.5V on 30V
3.5V on 300V
3.9V n 1000V

Two other things I noticed while checking that:

1. The actual value could be anything in the range of 3-11V when you power up the meter. The longer it's been off, the higher the reading.
2. The value decreases while the meter is on (as if a capacitor discharges), on all scales, until it stabilizes to values like the ones above.

I can swear this behavior is new...

[Caliaxy]

Could a calibration correct this behaviour ?

I don't know how to do that...

[Caliaxy]

I have not encountered this with my U1277a.
What firmware version is installed ?.

2.0

It couldn't be older, I guess... And it's an Agilent, not a Keysight (sorry)

[fiambre]

Similar issue here. Offset is only 0,25V negative though. If I leave the meter on reading increase slowly (as it starts negative, increase means reducing the absolute value and approach to zero).
Firmware is 2.04
I did s VDC calibration, only on "short" (0V) calibration, it helped, offset stabilize at around -20mV now.