Ceramic capacitors changing capacity over days

[mclarkson]

I'm reverse engineering an old board from the late 80's early 90's that was powered on right before disassembly.  There are 8 unidentifiable ceramic capacitors on the board which I have easily removed (hot air).
4 of the capacitors were easy to test and get the value with both multimeters that I have on hand (0.99nF, 099nF, 100.9nF, 102.9nF)
The other 4 capacitors however have been changing value day to day though

For example 3 of them follow the general degradation of
Friday - 154nF
Saturday - 139.9nF
Tuesday - 120nF
Wednesday - 115nF

Another one was
Friday - 122nF
Wednesday - 104nF

I'm trying to determine how to reliably replace these old ceramic capacitor?  Is this a situation where I would benefit from a LCR?  Should I try to find someone with a LCR with ALC and get them to read it (given the price of those things).
Or can I simply just replace it with the initial (or last) value and feel relatively safe?

Thanks for any help or insight into this matter.
--
Matt

[wraper]

Capacitance rise after heating and gradual dropping later is normal. read this document.

ageing process can be reset by heating the capacitors to a temperature above the Curie Point. The ageing process then starts again
from zero.

For example 3 of them follow the general degradation of
Friday - 154nF
Saturday - 139.9nF
Tuesday - 120nF
Wednesday - 115nF

Likely Y5V/Z5U crap. They noticeably change their capacitance even when you simply touch them with your fingers.

[sleemanj]

Looking at the values, I would assume all of them to be nominally 100nF, typical for decoupling capacitors.

[mclarkson]

Appreciate the responses guys.

It looks like this these are all used for power cleaning. 

I've attached a quick image of the different capacitors on the board that I'm looking at replacing.  It's a sound board for an old arcade game.



The chips in question are
AD1868R that utilizes 3 separate 5V inputs
-Pin1 - VL - Digital Supply (+5 Volts)
-Pin9 - VS - Analog Supply (+5 Volts)
-Pin15 - VS - Analog Supply (+5 Volts)

4558 that uses a single 5V VCC input.

Based on the comments, and this information, do you think it would be alright if I replaced all ceramics in question with 100nF?

Thanks again, those replies were fast!

[wraper]

Actual capacitance of decoupling caps for power does not really matter much. If you replace them with 10 times less or 10 times more, it will still work just fine.

[mclarkson]

Great!  Thank you Wraper!

[RoGeorge]

Most probably 100nF and 150nF.

Strange they all measure less with each day.
Did you checked the measuring instrument, or its battery? 

[The Electrician]

The other 4 capacitors however have been changing value day to day though

For example 3 of them follow the general degradation of
Friday - 154nF
Saturday - 139.9nF
Tuesday - 120nF
Wednesday - 115nF

Another one was
Friday - 122nF
Wednesday - 104nF

Take the capacitors that are changing value and heat them up again to the same temperature as when you removed them from the board.  Then measure them and see if their value has been restored to a higher value.  Measure them daily again and see if their value decreases every day like when you first removed them.  If they do, then the value to replace them with is their value after enough time passes so the rate of decrease is small; several time constants in other words.

[mclarkson]

Thanks everyone for the replies.

Yeah, I was thinking that it was my 90's era Radio Shack multi meter at first, so I ordered a brand new mid-tier multimeter.  Pretty much the exact same values being read.  So it's not the multimeter. 

I'll heat them up tonight and see what I can find.  Appreciate all the responses.  This is great.