Hanging a scope probe on the crystal pins will disrupt the oscillator.
For better accuracy use a TCXO - e.g. https://www.mouser.co.uk/Passive-Co...illators/TCXO-Oscillators/_/N-7jdmj?P=1yzmo0z

Will this mean that i would get 32768 hz from output from this device ?

https://www.mouser.dk/ProductDetail/Maxim-Integrated/DS32KHZN-DIP?qs=u4fy/sgLU9NOWMjpolj3aQ==

Then i could use this for input to mu MCU.

 

Your problem is not the devices you have. Buying even more won't help.

 

Will this mean that i would get 32768 hz from output from this device ?

https://www.mouser.dk/ProductDetail/Maxim-Integrated/DS32KHZN-DIP?qs=u4fy/sgLU9NOWMjpolj3aQ==

Then i could use this for input to mu MCU.

Yes.

 

Yes.

And then just take the output and connect to T0CKI.
No need for extra resistors or caps or external x-tals-

Setup timer0 to
T0CKPS<3:0>: Prescaler Rate Select bit
0111 = 1:128
Rollower at 256 = 32768 "ticks".

 

Some 32kHz crystals (like small cylindrical ones) are unstable if their shell is not connected to MCU supply common ground. To measure crystal frequency you need connect scope probes trough very small capacitors (couple of picofarads) and AC input mode. To make reasonable frequency measurements, scopes bandwith must be about 5 times higher than expected frequency.

 

Hi again,.
Somehow after setting a 1M Resistor in series on SOCS Out, i finaly got a working x-tal.
Now it's working with Timer1 , and i got a 1hz, have been running for 15 hours now, and not 1 second off.

Working with next step, later on.