Microsoft Language Card

16 posts / 0 new
Last post
Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
Microsoft Language Card

Hello group,I recently repaired a faulty Microsoft Language card that was locking the system up, reading disks incorrectly or not reading them at all. System worked fine with the card removed. The card had 3 bad TTLs. Both LS367s and the LS86. I replaced both tantilums as well. Now the card passed diagnostics and worked fine. Today I noticed it was doing the same thing it did before. Not loading programs correctly, locking the system up, intermittently not reading disks it just read. I removed the card and the system works fine. I was able to get the diagnostics to boot and it says the language card is fine also. (Which I’m certain it’s not) I tested each TTL and they are all still showing good. Anyone ever have a similar situation? I can’t seem to find a schematic for this card. There are a few other things on the board I can’t determine the values.Thanks group

 
S.Elliott's picture
Offline
Last seen: 10 hours 54 min ago
Joined: Jun 23 2022 - 16:26
Posts: 20
Yes, I've been in a similar

Yes, I've been in a similar situation.  In fact I've got two 16K RAM cards with faulty memory.  One of my cards dependably passes diagnostics like Locksmith's "RAM Card Utilities" despite being verifiably faulty.

But the following test reveals the fault.  Try these commands:

  1. At an Applesoft prompt, enter the monitor by typing: CALL -151  (return)
  2. Put the language card into read-ROM/write-RAM mode by typing: C081 C081  (return)
  3. Copy ROM into main memory and the card by typing: 2000<D000.FFFFM D000<D000.FFFFM  (return)
  4. Put the language card into read-RAM/write-RAM mode by typing:  C083 C083  (return)
  5. Compare language card RAM against main memory RAM byt typing:  2000<D000.FFFFV  (return)

This will verify that the language card and main memory both preserved identical copies of the data from ROM.  If everything is working correctly, the last command won't generate any output...it will just return to the monitor prompt like this:

 

On my faulty UNITRON 16KRAM CARD-1 the monitor prints a list of memory locations that didn't match, as shown below:

If your language card (and main memory) are both working, the monitor should not print any mismatched addresses.  If there are mismatches, like my second example, then it indicates that the data was not preserved accurately -- so there's definitely a fault -- but it doesn't pinpoint the cause.

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
I will give that a try.i have

I will give that a try.

i have another language card in there now (It’s an Apple Card not a Microsoft card) and I’m running a game demo to let things warm up. Even though the diagnostics showed the ram was all good I swapped them out anyway and same results. I don’t see any damaged traces in fact they look really good. One of these capacitors has to be failing.  There is really nothing to this board other than production components. 

I’ll try those commands and post my results 

Thanks for the help

Offline
Last seen: 16 hours 38 min ago
Joined: Apr 26 2016 - 08:36
Posts: 387
Does your suspect language

Does your suspect language card fail on power-up and pass after it's been turned on for a few (like 5-10) minutes?

 

I have one that will fail 100% of the time when it's cold and pass 100% of the time (and work perfectly thereafter) after being powered up for as little as 3 minutes.

Now I just treat it as a quirk.  I've troubleshooted that thing to kingdom come and can't figure it out, so I just let it warm up now.

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
It actually does just the

It actually does just the opposite. Seems to work fine for a few minutes then it starts acting up. 

I performed the test with the Call-151. I tried to post a picture of my screen but it was enormous when it uploaded. Like not even worth posting.  When I entered  C081 C081 my results were C081-DC C081-AC.  C083 C083 resulted in C083- AC C083- AE. After the last command it just brought me to a monitor prompt. No listing of mismatched data at all. I tried another card I have and I got the exact screen you showed with a good card. I tried a third card, one that has never given me trouble and I got the same mismatched C081 and C083 as the bad card and after typing the last line the mismatches just never stopped scrolling!! And this card isn’t giving me trouble!

But those are my results. 

Offline
Last seen: 9 hours 1 min ago
Joined: Jun 12 2022 - 23:35
Posts: 55
Maybe you should warm up the

Maybe you should warm up the chips before you test each one.  They could be intermittently failing.  Maybe heating to 50-60 C will reveal the culprit.

 

S.Elliott's picture
Offline
Last seen: 10 hours 54 min ago
Joined: Jun 23 2022 - 16:26
Posts: 20
Oopsie!  I'm sorry, I should

Oopsie!  I'm sorry, I should've told you to ignore the output from everything except the last command.

 

And, from what you wrote, the card passed this particular test.  Quoting the important part of your message:

After the last command it just brought me to a monitor prompt. No listing of mismatched data at all.

 

So your card correctly copied data into RAM and verified that it matched.  This test doesn't accomplish anything more than that.

 

Those C081 and C083 commands just change the mode setting between RAM vs ROM, but output like " C081- A0 " is just random-ish data that it picked up from the data bus while changing the card settings.

 

As to the second card, which generated a lot of differences,  I can only guess...

  1. Maybe you just had an unlucky type while entering one of the commands.
  2. Or maybe that second card really does have a bad region of memory that isn't detected by other memory-testing tools.

 

Sorry, if I caused more hassle than help.

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
No not more hastle than help.

No not more hastle than help. I always appreciate input from members. 

Just confused. Both cards show good results with the diagnostic software. (Apples service software not Locksmith). The card in question shows data being transferred correctly based on your test. But something is clearly wrong with the card. The card that fails your test showing data is NOT being transferred correctly gives me no issues. Stays on for long periods of time, doesn’t lock up and reads disks fine. Tonight I’ll check the TTLS on the second card, maybe one of them is causing the data to not be transferred correctly.  But I have no idea where to look next on the card that I started this thread about. 

Offline
Last seen: 2 weeks 5 days ago
Joined: Dec 3 2015 - 20:45
Posts: 62
on one of my 3rd party 16K

on one of my 3rd party 16K cards, I found a certain brand of ram chips caused errors but were fine if used on the motherboard so just exchanged a row between them.

I don't recall which brand.

 

Larry G

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
Thanks Larry,i swapped the

Thanks Larry,

i swapped the ram on the Microsoft card and same symptoms. I haven’t changed out the ram on the Apple Card that’s basicslly telling me everything’s wrong but I just had it on for 45 minutes and no issues.  I tested all the TTLS and they show good. 

Does anyone know the command from monitor to see memory size? I saw it somewhere and I can’t remember where. I’m not sure if it will even show auxiliary memory but I’m curious if it does. 

Offline
Last seen: 2 weeks 5 days ago
Joined: Dec 3 2015 - 20:45
Posts: 62
on boot Applesoft scans the

on boot Applesoft scans the lower 48k region to determine how much ram and places the word at $73 and $74. It does not search for the upper 16k since it is not mapped.

The schematic for the Microsoft card is in the back of the manual. I can't figure out how to upload the pdf. I found it here

https://mirrors.apple2.org.za/Apple%20II%20Documentation%20Project/Interface%20Cards/Language%20Cards/Microsoft%2016K%20RAM%20Card/Manuals/

 

Apple-Cillin II does a good job of testing the LC. I uploaded a version I cracked from the original nib a few years ago to Asimov.  Looks like they moved things around. Look for Apple-Cillin II.dsk in here

ftp://ftp.apple.asimov.net/pub/apple_II/images/disk_utils/diagnostics/

 

PS - the Hot Rod crack (the applecillin with hr on the end) had bugs and the disk tests did not function at all, mine does.

 

Larry G

 

 

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
Thanks Larry for the link to

Thanks Larry for the link to that manual.  That's something that should probably be printed out.

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
Testing the cable this

Testing the cable this morning I noticed I had to twist the cable pretty good to get a reading on the #11 pin. I got continuity to the wire end but the solder point on the board the cable needed twisting. #14 was intermittent as well. Had continuity to the solder point but just the right position of the cable and I would loose it.  So it looks like my first move is to make a new cable. 

Would anyone have a link to these 16pin male IDC connectors? All the links I found seem to have the spacing too close. Looks like they’re 7mm.

Thanks group

 

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
And as far as the Apple Card

And as far as the Apple Card that was showing nothing but errors with those commands....I tried the card in a ii plus and performed the test and got zero errors. So the card tests all good, it just doesn’t like the Rev0. Could the problem with all the errors the language card is getting have something to do possibly with the Rev0 has integer basic ROMs?  

S.Elliott's picture
Offline
Last seen: 10 hours 54 min ago
Joined: Jun 23 2022 - 16:26
Posts: 20
mochatea396 wrote:Could the
mochatea396 wrote:

Could the problem with all the errors the language card is getting have something to do possibly with the Rev0 has integer basic ROMs?  

Oh ho, good thinking!  Yes, it could.

I forgot to account for empty ROM sockets.  Integer BASIC doesn't fill all the ROM sockets, so any empty sockets will just produce random-ish bytes.

So the comparison command was undoubtedly finding mismatches for addresses that had been copied from those empty ROM sockets.

Offline
Last seen: 3 weeks 6 days ago
Joined: Apr 23 2022 - 18:41
Posts: 32
Once I tried the card in the

Once I tried the card in the ii plus and performed that test and I got no errors I figured it must have to do with the integer roms. So the card is appearantly working fine.  A friend is making me a cable for the microsoft card and he's going to try that test in one of his integer machines and see if he gets the same results. Ill let you know what he finds for your reference.

Log in or register to post comments