Mikkel C. Simonsen wrote:
-------------------------------
But I thought that was ok. But then it turns out the chips behave very much like blank chips after programming the lock bits... If I skip that step the chip works fine in circuit.

I've now got one chip to work after programming the lock bits (an AT89C4051-24PI date code 0038).

I have tried with other chips from the same batch as well as various 1051U and 2051 chips, but only one success story so far...

The other chips appear to be either partially or completely erased when I write the lock bits. The chips work fine after programming, but after programming the lock bits they either do nothing or work only partially when installed in the application circuit. If I erase and re-programme the chips they work fine again.

Could the reason be that I've exceeded some time limit in the "write lock bits" process? Too bad the "write lock bits" part is the only thing not included in the timing tables and chart...

Best regards,

Mikkel C. Simonsen