Anyone know what this P9852AD chip does?

[Lorenzo_1]

Trying to repair this rather dated garage door controller (for a spare in case of failure - I've had a couple pop their clogs).  The power supply is good at all points and no obvious points of failure apart from two dodgy switches that I will replace. Oscillator works fine. It's not completely dead but doesn't respond properly to press-button commands. I've been able to get datasheets for all the ICs apart from the circled one in the photo. There's an LM358 on the door open/close control circuit, a 93C06/46 EEPROM (presumably for code read/write), an FM processor for the remote control, and then the P9852AD. This code comes up in searches as a possible Fairchild chip but no datasheet and no information about what it is/does. Number is preceded by ATA, which happens to be the manufacturer's name, so I wonder whether it's some sort of proprietary programmable logic. Haven't mapped the board but it looks as though this talks to the EEPROM (presumably for access code R/W), drives the relay coils for open/close, and takes inputs from some/all of the push-button controls. 

[helius]

9852 is likely to be just a date code—52nd week of 1998.
From the package and your description I expect it's an ASIC or CPLD

[Ian.M]

.... or a custom mask programmed MCU.  Whatever it is, the close proximity and interconnection with the memory chip, a ceramic resonator and the traces to the buttons and relay drivers imply with a high level of confidence that its the 'brain' of the system, and if its dead or damaged, the unit's unfixable without a parts donor.

[Lorenzo_1]

Thanks for the feedback. All of that makes good sense. Hadn't thought about date codes... Got some new microswitches today - will put them in and see if that fixes the bugs. If not, might try and decipher some of the traffic in and out and see whether it seems to be behaving sensibly.  Seems the only stock now available is refurbished boards at a highish price, so worth a try.

[CJay]

Nat Semi COP8 series processor perhaps...

[Lorenzo_1]

That looks like it might be right CJay. Quick look at a COP8 Designers Guide illustrates a COP8SAA7 tied to an NM93C06 (256k version of the 1k NM93C46 in my unit) for a rolling code RF transmitter door opener system.  Changing the dodgy switches didn't cure the problem.  Your lead looks like it will help me decipher whether it's alive or not. Thanks.

[CJay]

Handy tip is that it may be possible to read the code out from the COP8 relatively easily even if it's 'protected' however I hope it's not necessary.