Maybe we shouldn't take Jprao's comment too literally?

in the below, some statements do not reflect on this specific thread, but general issues (e.g. Jprao has (tried to) answer questions.

1) the language barrier which makes I am using imported resistor divider (1M/1000M). It is giving entirely non-linear readings. The voltage coefficient is good. totally unclear, I read "The voltage coefficient is good" as "the thing is linear" and obviouisly that does not jive with "It is giving entirely non-linear readings".
2) the 'fear' of appearing stupid (as you all now I have nio such fear :) and trying not to be too verbose. Here my 'standard' is "more words does not make you appear stupid, they make you appear precise"
3) the refusal to answer questions.

now
to Jprao (how do you pronounce that?)
one way to overcome the language barries would be that you gave us the DVM readings for 4 widely separated voltages in this form

500V reads 1.17
2000V reads 1,7
3000V reads 2.0
6000V reads 2.7

Erik