Installing a PLCC EEPROM Socket onto a MOBO

Once you are happy with the state of the socket’s connectivity, it’s time for the electrical test.   For the specific socket that we used, the EEPROM chip seemed to pop in a little deeper than we wanted it to.   It still make connections just fine, but having the main leads of the IC so close to our soldering work made us nervous.   We ended up folding a thin strip of paper two or three times and laying it down in the bottom of the socket, so the IC would rest flush with the top of the socket as we expected it to.   This may just be a bit of engineering voodoo, it’s actual benefit is debatable.   But if you find you are now getting pin-to-pin shorts with the IC installed, you may try this technique.

Fig 4 – The EEPROM Chip, Installed in the Socket


Now for a little-known piece of engineering lore:   ESD diode test.   If you already know this, then you are a true nerd and we praise you.   But if you don’t, then this is the one piece of QA testing that will virtually guarantee your install is good or not.

Each pin of an IC must be protected against static zap.   The circuits to do this range from simple diodes to grounded-gate mosfets, to SCR junctions.   In any case, each of these techniques gives some sort of PN junction on each pin.   Exceptions to this include HV programming pins which require fancier techniques, but we’ll ignore them for now.   Most ESD structures or output drivers give a PN junction from GND to the pin.   Some may use a PN junction from the pin to VDD, but these are less common.

To ensure that each socket pin is making a good connection to actual silicon, you can use a DMM on “diode test” mode.   The anode of the diode is GND (DMM Red lead) and the cathodes are each pin (DMM Black lead).   By testing between GND and each pin, you should see 0.6v to 0.7v diodes at each pin.   The VDD pin will have a much larger diode with a lower Vf of 0.4v or so.   For pins that show nothing to GND, try checking for a diode between the pin (DMM Red lead this time) and VDD (DMM Black Lead).   If you don’t see any diodes and the pin is a normal pin (address, data, CE, WE, OE) , it’s not connected.

We strongly encourage the use of the diode test, and just to show you that we’re not bullshitting, here’s the results of the diode test on our EEPROM.


01) NC       x                       17) D3       0.724
02) A16     0.681         18) D4       0.723
03) A15     0.682         19) D5       0.722
04) A12     0.681         20) D6       0.723
05) A7       0.682         21) D7       0.723
06) A6       0.681         22) !CE     0.679
07) A5       0.680         23) A10     0.680
08) A4       0.682         24) !OE     0.672
09) A3       0.679         25) A11     0.680
10) A2       0.681         26) A9       0.681
11) A1       0.680         27) A8       0.681
12) A0       0.683         28) A13     0.681
13) D0       0.723         29) A14     0.680
14) D1       0.720         30) NC       x
15) D2       0.722         31) !WE     0.680
16) GND     x                     32) VDD     0.477

Last but certainly not least, it’s finally time to fire it up.   Here’s the results of our successful test on this system.   Done and done.


Fig 5 – It’s Alive!


Hope you’ve enjoyed the article, and please go back and read the first part on how to remove the EEPROM with a heat gun if you have not yet done so.