• Please review our updated Terms and Rules here

Help Diagnosing and Finding Ram Replacements for IIe

I

ignis-fatuus

Member
Joined
Oct 7, 2022
Messages
24
So I've been slowly going through and troubleshooting a recent Apple //e that I picked up on the cheap. It will boot into basic but has difficulty reading floppies. At first I suspected the the power supply, but after opening it up and testing all the caps with an ESR meter, everything checked out (I will do a full cap job ASAP--I did replace one 220 uF@16v that wasn't reading great). Whenever it tries to read floppies it makes a harsh stuttering sound different than the drive initialization noise that accompanies normal disk drive operation. I tried swapping the duodisk for a known working Disk II and I get the same result--the duodisk does work on another machine so I suspected that there is a fault on the motherboard. Please see the following error message which accompanies the attempted reading of a floppy (also note that numbers are different every time but the error message is consistent in its formatting):
20221106_231637_2.jpg

Running the self diagnostic I get the following result:

20221106_233748_2.jpg
I believe that the error message from the self diagnostic is suggesting that following RAM ICs are bad: F13 F12 F11 F10 F9 F8 F7 F6, which would be all of them...I honestly suspect that multiple RAM ICs have failed, but perhaps not every single one. Perhaps the failed RAM is associated with trouble in disk operation? That is my assumption for now. I am also wary of the fact that the RAM chips are socketed. Just about every IIe board I've seen has the ram soldered into place (did later revisions include sockets by chance?). I was hoping to use some II/IIe diagnostic software I have to help get a better idea what is going on but without working disk drive operation, I am at a bit of a loss.
20221107_073354.jpg
In thinking toward simply replacing the RAM, I am having some difficulty figuring out a proper replacement. The RAM ICs are labeled MT4246 with the Apple logo. According the data sheet for the IC it says there should be a dash and a number indicating timing speed (-20=200ns, -15=150ns, etc), but as you can see from the picture, no such indicator is provided.

If anyone has any insight toward testing the RAM further and/or finding suitable substitutes, I would be most appreciative.
 
You mean MT4264, right? 64Kx1 DRAM. 4164 equivalent; single +5 (Vbb is NC). Not all 4164s require Vbb.

I'd look elsewhere for the problem. Chances are very good that the DRAM is fine, unless something really terrible has befallen the motherboard.
 
ignis-fatuus said:
I believe that the error message from the self diagnostic is suggesting that following RAM ICs are bad: F13 F12 F11 F10 F9 F8 F7 F6, which would be all of them...I honestly suspect that multiple RAM ICs have failed, but perhaps not every single one. Perhaps the failed RAM is associated with trouble in disk operation? That is my assumption for now. I am also wary of the fact that the RAM chips are socketed. Just about every IIe board I've seen has the ram soldered into place (did later revisions include sockets by chance?). I was hoping to use some II/IIe diagnostic software I have to help get a better idea what is going on but without working disk drive operation, I am at a bit of a loss.
Click to expand...
I believe later revisions of the motherboard removed the sockets for cost savings as well as increasing reliability. First thing I would recommend is to pull the chips from the socket, maybe apply some dexoit (if you have some), and then reinsert them. You may want to do this with every socketed chip in the system. I recently diagnosed a II Plus where multiple chips needed to be pulled and reinserted. I decided just to pull them all and reinsert them. No problems since.

As both of you have concluded I doubt the problem is with all eight chips, possibly a problem with the decoding logic for the RAM.
 
I see lots of reports of bad Micron Technology DIP RAM, but I wonder if it's legitimately bad or if it's just that Micron made so much RAM, that you're just likely to find it everywhere.
 
oldpcguy said:
I believe later revisions of the motherboard removed the sockets for cost savings as well as increasing reliability. First thing I would recommend is to pull the chips from the socket, maybe apply some dexoit (if you have some), and then reinsert them. You may want to do this with every socketed chip in the system. I recently diagnosed a II Plus where multiple chips needed to be pulled and reinserted. I decided just to pull them all and reinsert them. No problems since.

As both of you have concluded I doubt the problem is with all eight chips, possibly a problem with the decoding logic for the RAM.
Click to expand...
Good call on the deoxit. I have plenty of it from working on receivers and amplifiers. Would you apply the deoxit to the sockets and work the chips in? I'm surpised I never thought to consider that sockets would oxidize and cause issues down the line. Will give it a go this weekend and report back with the results.
 
jafir said:
I see lots of reports of bad Micron Technology DIP RAM, but I wonder if it's legitimately bad or if it's just that Micron made so much RAM, that you're just likely to find it everywhere.
Click to expand...
Do you have any recommendations for more reputable brands and/or specific ICs to look out for?
 
My go-to DRAM of this era was the NEC uPD4164. Make sure that you get the faster parts. There were a lot of manufacturers of 64Kbit DRAM, BTW.
 
ignis-fatuus said:
Good call on the deoxit. I have plenty of it from working on receivers and amplifiers. Would you apply the deoxit to the sockets and work the chips in? I'm surpised I never thought to consider that sockets would oxidize and cause issues down the line. Will give it a go this weekend and report back with the results.
Click to expand...
Just spray it into the sockets. If you want to clean the pins I've found soft pencil erasers work well.
 
OK, I pulled every socketed IC and sprayed the sockets with Deoxit cleaned the pins pins of the ICs with an eraser. Unfortunately, when I run the self diagnostic I continue to get the same error. The screen displays the following and does not progress further:

RAM: F13 F12 F11 F10 F9 F8 F7 F6
 
Last edited:
I find it hard to believe that all eight chips would fail at the same time. While Micron parts have been problematic, it's extremely unlike that all eight would fail at the same time.
clean the MMU and CPU while you're spraying away with deoxit.
 
Success! And it turns out my initial hypothesis was correct. After pulling all the ICs and hitting the sockets with Deoxit, I decided to switch out the ram with a full set of HYB4164 chips I borrowed from an extra 80 Col card I haven't been using and it booted right up.
20221113_195106.jpg
Since I am now able to access the disk drive, I populated the 80 Col card with the MT4264 chips I pulled the motherboard and ran an extended ram test. Turns out two of the ram chips had failed (see pic below); I guess the MT4264s really are prone to failure!
20221112_215002.jpg
 
Last edited:
Chuck(G) said:
So not all 8 chips bad?
Click to expand...
Nope, only two. I used Computer Checkup to check the suspect ram I installed in to the 80 Col card. The software does a decent job of locating the bad ICs (as shown in a previous pic). The two bad chips must have interfered with the self test diagnostic making it unable to correctly determine the bad ram locations.
 
This is one of the main reasons I will snag an extra 80 column card whenever I notice one that is socketed. Having a full extra bank of ram on hand can definitely help shortcut the troubleshooting process; I fixed my own machine the same way. Of course, I'm also lucky enough that my IIe also had the RAM socketed to begin with. Not a job I can't do, but I'm always grateful when I don't have to desolder entire DIP chips just to check them.
 
ignis-fatuus said:
Nope, only two. I used Computer Checkup to check the suspect ram I installed in to the 80 Col card. The software does a decent job of locating the bad ICs (as shown in a previous pic). The two bad chips must have interfered with the self test diagnostic making it unable to correctly determine the bad ram locations.
Click to expand...
After reading post #12 this was my thought too. How can you trust the diagnostic results of a system diagnosing itself?
 
You must log in or register to reply here.
Back
Top