SAMS card troubles

[+Vorticon]

Hi. My original 1M SAMS card from 1995 is acting up. Here's the result of the memory check. Any suggestions where the fault might lie?

 

[GDMike]

You can dl my SNP program and run it.

And it will init all UPPER banks with blanks, and you can page through each page looking for garbage.

Also, Turboforth will allow setbank. And you can do a similar thing.

I guess, I don't know what that error means except maybe it's doing a CRC at init time, which doesn't necessarily mean that the 32 k in lower banks is bad, but maybe an UPPER memory detection?

Maybe someone can chime in about this error. But wow, it lasted this long?!!!

Great!!

This program is a Fixed 80 program name is SNP.

Maybe the caps and voltage regulator need swap out.

SNP

Edited December 13, 2020 by GDMike

[+Ksarul]

I'd clean the edge contacts before you try anything else, some of them look pretty grungy. . .

[+TheBF]

I have had it happen that removing and re-installing socketed chips solves a problem. There are a lots of opportunity for that problem on that board.

[+FALCOR4]

1 hour ago, TheBF said:

I have had it happen that removing and re-installing socketed chips solves a problem. There are a lots of opportunity for that problem on that board.

Definitely try that first.  I had a similar problem that cleaning solved.  I ended up cleaning the pins of the ICs, using contact cleaner on the pins and sockets, then swiping a thin film of dielectric grease on the pins before reinserting them. The card has been rock solid since.  Hopefully something like that will work for you.

[+FarmerPotato]

@InsaneMultitaskermade a comment that he replaces single wipe sockets with double wipe. I think most low profile open sockets are not single wipe. I never knew about this before. 
 

are your sockets machine pin holes or the springy contact kind?

 

I read elsewhere that corrosion in the pins bends the contacts so new chips don’t make good contact. Folks recommend machine pin sockets. I hate those because it is hard to get a 32 DIP into one without bending a pin.  
 

I am going to pay more attention to the socket contacts in future. 

[+InsaneMultitasker]

21 minutes ago, FarmerPotato said:

I think most low profile open sockets are not single wipe. I never knew about this before.

Here's a site I saved along the way that may be of interest.   The first few pictures of the sockets represent what I have found along the way in many of the cards I repaired - whether logic, ram, or EPROM - and often times the low height coupled with age tends to lead to sporadic connections with the chips. Not all of them are problematic though certainly a good thing to keep in the rework/troubleshooting arsenal.  (the comment about new chips flaring outward is something that caught my attention back then as well)

 

http://arcadecontrols.com/BBBB/ic.html

[+TheBF]

I can't say if this is 100% true.

A tech I worked with once told me that H/P had discovered fungi that can grow on metal, even gold. Over time the fungi can actually get between contacts.

 

We had 3 early vintage colour cameras by RCA that were forever going bad sometimes in the middle of a live show!

The quick fix was to remove a card and re-insert it. 

He suggested using a cleaner with contact oil like Nutrol.  We did all the cards by wiping them with a cloth and nutrol.

Those problems never recurred.

My theory is the contact oil prevented re-growth if it was in fact this fungus. The cameras were almost 20 yrs old at the time.

 

 

 

[GDMike]

Encase the whole thing in cement!

Damn fungi, take that!

[+FarmerPotato]

4 hours ago, InsaneMultitasker said:

Here's a site I saved along the way that may be of interest.   The first few pictures of the sockets represent what I have found along the way in many of the cards I repaired - whether logic, ram, or EPROM - and often times the low height coupled with age tends to lead to sporadic connections with the chips. Not all of them are problematic though certainly a good thing to keep in the rework/troubleshooting arsenal.  (the comment about new chips flaring outward is something that caught my attention back then as well)

 

http://arcadecontrols.com/BBBB/ic.html

Ha! That was the first google hit I got using my phone. It's much better looking now that I am at my desktop.

Thanks for posting that. Lot of practical experience there. 

 

The single wipes that I have are wire-wrap, some of them reused!  The new ones of those are machined-pin, which I hate. There's only one example of the boxed up (double-wipe) on page 2 here: https://www.peconnectors.com/sockets-wire-wrap/

[+Vorticon]

Welp I cleaned all the contacts with 91% IPA, re-seated all the chips after spraying Deoxit on the machined sockets, and also deoxidized the PEB slots, but unfortunately with no change. I think I have a bad chip somewhere and the trick is to figure out which one... I wish there was a SAMS testing program which pinpointed the bad chip.

Any suggestions for a testing strategy would be much appreciated. I have all the likely needed testing equipment.

 

[RickyDean]

I'd start with the sram chips first, then the buffer chips and work around the board. An eprom tester will generally have the ability to test the ttl and sram. The 612 mapper is a hard one to test, at least for me as none of my equipment can read it for testing.

[+Ksarul]

Do you have an EPROM burner with integrated LS logic testing capability? If you do, just pull the ones it can test one by one and see if any show up bad. . .

 

[+TheBF]

1st thing to check is for correct power.  Is the 7805 regulator making 5 volts?

 

This can be hard to test without a card extender.

You might have to solder a wire onto the output pad of the regulator chip. Insulate the end you will measure with tape to prevent accidentally shorting it. (Murphy's law)

 

Or just replace it with a new one.

 

 

 

[+Vorticon]

On 12/13/2020 at 8:21 PM, Ksarul said:

Do you have an EPROM burner with integrated LS logic testing capability? If you do, just pull the ones it can test one by one and see if any show up bad. . .

 

I have a TL866CS programmer, and the 6840008 SRAM is not on the list of supported chips unfortunately. Is there an equivalent chip I could use?

[+Vorticon]

On 12/13/2020 at 8:22 PM, TheBF said:

1st thing to check is for correct power.  Is the 7805 regulator making 5 volts?

 

This can be hard to test without a card extender.

You might have to solder a wire onto the output pad of the regulator chip. Insulate the end you will measure with tape to prevent accidentally shorting it. (Murphy's law)

 

Or just replace it with a new one.

 

 

 

I used an extender and checked the output of the 7805 and it's a steady 5V. So that's not it. That would have been too easy, wouldn't it ?

[+TheBF]

I am not sure what the software test disclosed. 

When it says "data not kept correctly" does that mean everything? For example are the registers retaining data? 

 

Sticking with the power question. I suppose the next thing to confirm is; is that 5V getting to all the Vcc pins of each chip?

There are a couple of tantalum caps on the board. They can sometimes short although that was common years ago. Modern ones may be flawless.

 

Solder joints can open up over time as well if they were not quite right in the first place. 

So touching up all the solder joints to those sockets can fix strange problems.

 

And then of course old chips can fail. That's where replacements are that fastest way out the door.

 

 

 

 

 

 

[+Vorticon]

The error message is very vague unfortunately, so it could be anything. I'm working on testing all the chips I can test, then order replacements for the rest although it might actually be cheaper to just buy one of the new SAMS cards from Jim... That said, I just hate throwing away a fixable piece of equipment, particularly since all the chips are socketed.

We'll see how far I get.

[GDMike]

Does the computer work? I mean is the base 32K useful? I didn't see where you said that it didn't or did function. Sorry if I missed that part. I just saw the test results.

[+Vorticon]

22 minutes ago, GDMike said:

Does the computer work? I mean is the base 32K useful? I didn't see where you said that it didn't or did function. Sorry if I missed that part. I just saw the test results.

Seems to but I have to test that more extensively. 

[RickyDean]

3 hours ago, Vorticon said:

I have a TL866CS programmer, and the 6840008 SRAM is not on the list of supported chips unfortunately. Is there an equivalent chip I could use?

From a Google search you have 512Kx8 sram, I do not see an equivalent listed in the minipro menu, when selecting standard sram, but there may be an equivalent in one of the other listed manufacturers.

[+TheBF]

Not sure is this is of use Vorticon. 

I had this in the can from some time back and I can now compile EA5 programs so here is lexicon of test commands. There is a HELP command.

It needs Editor/assembler cart.

 

I have not tested this on real iron yet.

It tests the card in 64k segments. 

It does not try to test the bottom 64K of memory.(segment 0) 

 

If the card does not work at all then of course it will not run.

 

If there is any specific test you would need let me know I may be able to build something for you with the tools I have.

Edit Bug fixed code.

 

Spoiler

\ test sams card in 64K segments

MARKER /SAMSTEST   \ command to remove everything when you are done.
NEEDS DUMP FROM DSK1.TOOLS
NEEDS PAGED  FROM DSK1.SAMSFTH
NEEDS SDUMP FROM DSK1.SAMSDUMP
NEEDS COMPARE FROM DSK1.COMPARE
NEEDS ELAPSE FROM DSK1.ELAPSE
NEEDS U.R    FROM DSK1.UDOTR

HEX
7FFF CONSTANT 32K
FFFF CONSTANT 64K
1000 CONSTANT 4K

VARIABLE FATAL

: ?FATAL ( flag -- ) FATAL @ AND IF CR ." Stopped on failure" ABORT THEN  ;

: ?BREAK  ?TERMINAL ABORT" *HALTED*" ;

CREATE PBUFFER   4K ALLOT  \ holds the pattern we are testing

\ : ERASE    0 FILL ;
\ : BLANKS  BL FILL ;

CODE MOVE16 ( addr1 addr2 len -- ) \ fast move 16bit cells
    C036 , C076 , C104 , 1306 , 0584 ,
    0244 , FFFE , CC31 , 0644 , 15FD ,
    C136 ,
NEXT,
ENDCODE

: BINARY   2 BASE ! ;

: SETPATTERN ( char -- )
        BASE @ >R
        CR ." Bit Pattern = " BINARY DUP 8 U.R
        R> BASE !
        PBUFFER 4K ROT FILL
        CR ;

\ move pattern-buffer into entire segment
: FILLSEG ( seg# -- )
           SEGMENT \ set the segment
           64K 0
           DO
             PBUFFER  I PAGED  4K MOVE16
           4K +LOOP ;

: PAGETEST ( addr -- ?) PBUFFER 4K ROT PAGED 4K COMPARE DUP ?FATAL ;

: TESTSEG ( seg# -- )
          CR ." Fill 64K seg# " DUP .  ." with" PBUFFER C@ HEX 4 .R
          FILLSEG
          64K 0
          DO
              CR ." Address" HEX I 5 U.R  ."  ... "
              I PAGETEST
              .S
              IF   ." FAIL"
              ELSE ." PASS"
              THEN ?BREAK
          4K +LOOP
;

DECIMAL
: SAMSTEST
      PAGE
      CR ." SAMS Card memory test"
      CR ." ---------------------"
      16 1
      DO CR
         CR ." *SEGMENT*" DECIMAL I  3 .R
         CR I TESTSEG
      LOOP ;  ( 58 secs)

HEX AA SETPATTERN

: REPTEST ( SEG# -- )
     BEGIN
         PAGE
         CR ." SAMS 64K Repeat Test"
         CR ." --------------------"
         CR ." *SEGMENT*" DECIMAL DUP  3 .R
         CR DUP TESTSEG
         ?BREAK
      AGAIN ;

HEX
: REGTEST

      1E00 'R12 !       \ select SAMS card CRU address
      SAMS-OFF          \ mapper off
      0SBO              \ turn on register memory
      4000 PAD 20 MOVE  \ save the registers
      4000 20  FF FILL
      4000 20 DUMP
      CR ." Should see FFFF"
      CR ." Restoring registers
      PAD 4000 20 MOVE  \ restore registers
      0SBZ
      SAMS-ON
;

: HELP
  PAGE ." * SAMS 1Mb TESTER Help *"
  CR ." --------------------------"
  CR ." HEX ( xx) SETPATTERN"
  CR ."    sets bit pattern used"
  CR
  CR ." ( seg#) SEGMENT sets segment"
  CR
  CR ." ( seg#) REPTEST Test seg# until BREAK"
  CR
  CR ." SAMSTEST  test all segments"
  CR
  CR ." ( addr len) SDUMP dump SAMS memory"
  CR
  CR ." FATAL ON  breaks on error"
  CR ." FATAL OFF no break on error"
  CR
  CR ." REGTEST " 
  CR ." HELP   show this screen"
;

: COLD   WARM
         PAGE ." SAMS Tester "
         CR ." HELP to see commands
         ABORT ;

LOCK

INCLUDE DSK1.SAVESYS

' COLD SAVESYS DSK2.SAMSTESTER

 

 

 

See next post for corrected version.

 

[+TheBF]

Very sorry.  I added a FATAL error halt feature and of course I added a bug to go with it.

 

When you run some of the tests they don't return properly and the command interpreter is left in limbo. 

 

This version has the bug fixed.

 

 

SAMSTESTER2.VIP.zip

[+arcadeshopper]

you could also try matt's tester here: https://github.com/jedimatt42/ti994a-32kmemtest it does sams and tests more extensively than the amstest4 does

 

 

[+Vorticon]

14 hours ago, TheBF said:

Very sorry.  I added a FATAL error halt feature and of course I added a bug to go with it.

 

When you run some of the tests they don't return properly and the command interpreter is left in limbo. 

 

This version has the bug fixed.

 

 

SAMSTESTER2.VIP.zip 9.41 kB · 4 downloads

Great program! You've got to love Forth!

I am seeing some inconsistencies however:

• AMSR4 reports only 512K of RAM (after I reflowed the solder joints, the RAM size went from 1024 to 512 irrespective of the position of the SRAM chips), but SAMSTESTER seems to see all of the memory. The REGTEST passes.
• SAMSTESTER goes only to segment #15. Shouldn't there be a segment #16 to account for 1024K? It starts at segment #1, not #0.
• AMSR4 is locking up at random memory pages. Not so with SAMSTESTER

I think the problem lies with the 612 memory mapper. I have managed to test almost all of the support chips and they pass. I think one of the SRAM modules is bad as well but I am getting inconsistent results with the testing (swapping the SRAM chips sometimes gives an error, and sometimes not).

I'm just going to order replacements for the 612 and SRAMs and see if that solves the issue.