Multimiter sanwa PC510a strange temperature reading

[haha123]

Hi I have sanwa PC510a multi.
the temp reading is offset by about -1000 C
it was fine few months ago and started acting up recently
so room temp will read ~970 C

it is not the sensor ( K type) as when I plug to other meter (fluke) it reads just fine
the sensor is sanwa original

all other readings like voltage AC&DC, current AC&DC, resistance, diode, and capacitance give normal readings within
meter's specification.

I ve tried searching internet for schematic, but no luck so far.
I guess it has sth to do with the cold junction compensation circuit or sth else, but i am really blind about servicing meter.

i ve tried cleaning and brushing pcb with alcohol, again no change.

any help is appreciated.
thanks

[BreakingOhmsLaw]

Most likely cause: It's a PT1000 sensor and the meter is set to PT100.

Connect a 100 Ohm  and 1kOhm resistor instead and tell us the readings. Should read around 0° C for one of them.

[haha123]

@BreakingOhmsLaw

Thanks for your quick reply.
I failed to mention that this k probe came with the meter, and I only have this one probe l, so the possibility of mistaking or unintentionally switching it with other type is nonexistent..

100, 1kohm or even shorting rest lead gave me -1023 to -1025 depends on the Meters mood...

I really wish it was that simple, but i guess its time to retire this one :sad

Thanks for the effort btw ! :smile

[lowimpedance]

PRT sensors are not thermocouples !. Not sure what the Sanwa uses for the cold junction maybe a diode junction  .
Some nice close up pictures of the input area on both sides if the PCB might help identify possible suspects.

[lowimpedance]

One other thought, does light flexing or prodding the input area of the PCB with an insulated stick  result in any obvious changes to the reading ?.

[haha123]

Surprisingly twisting the meter with hands affects by a couple of degree. Sticking sth into jack doesnt do anything.

Heres some of photos. I hope its worth sth
https://drive.google.com/folderview?id=1PvWGvbGEHpEcEWyMLNhYhlqqCE_COB30

[lowimpedance]

Can't see enough close up of the input jack area from those pictures to guess at possible cold junction comp component. I would have thought the cold junction component to be near the input jacks ,( just a guess here !!).
One thing to look for would be the obvious diodes and or transistors with collector-base connected together,  and then use another meter to test for open or short and a good PN junction etc . I would also check for cracked solder joints etc.

Edit: Just had a look at the Fluke 87V schematic and that meter uses a BJT connected as a diode and mounted close to the input jacks for its thermocouple reference junction, see Q3 on page 40 cct.
(manual readily available for download!).

If shorting the input terminal does not display the ambient temp. I would trace from the function switch 'TEMP' position to see which circuits are in use and perhaps with some luck locate a bad connection  .

[haha123]

Thanks for the reply

There is only protection circuit on the input daughter board.
The temp compen circuit is somewhere on main board.
Unfortunly with other things at hand, I am not going to chase the rabbit down.
I will keep this one as my sub meter.

Thanks for the help folks !