giberish desplaying

To post a picture, you click on "Media browser" after typing your post.

Choose (or drag) a file to upload, then select the display style. The default style is to show the image inline, but the browser can also generate a "preview" or "teaser" with a link.

Your screen looks like there is a problem with video generation, maybe with the character rom or the 74LS166. I'm not sure.

It is definitely not a problem with the character ROM. Most likely it's due to a problem with the memory.

That's not gibberish. It's lores graphics. When you were getting the beep it meant that the CPU was functionng correctly (along with ROM, RAM, and a bunch of other stuff). I assume this is an Apple II or II+ (you didn't say).

Try reseating/replacing the 74LS259 at location F14 on the motherbaord. This usually causes this symptom. Once you get a TEXT display you may find other issues.

1ajs wrote:

its a apco clone

That's probably good news -- it might explain why the bit patterns correspond to the letters  "A P C O"  with space characters between them.

Jeff' Mazur pointed to the 74LS249 (Addressable Latch) at F14 because that's the component that stores the choice of video modes.  The computer is behaving as if all the video mode settings inside the Addressable Latch are stuck at OFF.  (ie: Text mode off, Text window off, Page2 off, Hi-res off)

Also check the 74LS151 (8-to-1 selector) at A9.  That's the component that selects which signal to send to the video output.  A fault in either A9 or F14 can cause these same symptoms, since both components are necessary to choose the correct video signal to send to the video output.

1ajs wrote:

after steeling some 4116 chips off the ram expson and swaping the 1st two i get to a blank screen with apco at the top of the screen

 

pluged the spkr in and the kb but its not responsive to the kb. i know it was respnsive when it was a garble as u could type so prolly more chips need to be reseated?

 

 

Do you have a disk controller plugged in?  If you do and there is no bootable disk in a drive attached as drive 0 it will sit there forever like that.

What happens if you press reset?  You should get a BASIC prompt.

Are you pressing Reset on its own?  Try pressing CTRL-Reset.  

When a reset is pressed it directly forces the reset pin on the CPU high.  You should be able to see that change with a voltmeter on pin 40.

Is the keyboard connected?  Hitting reset should do something.  If it doesn't you've got another fault somewhere.

As baldrick noted, it may require ctrl-reset.

1ajs wrote:

ctrl reset it beeps eventualy throws up another error 

e005-  a=023 x=38 y=1b s=f1

 

guesing more bad ram chips?

Could be bad RAM, but that address is in the ROM area so it could be a corrupted EPROM too.  Or something in address decoding logic causing thigns to be off.

softwarejanitor wrote:

Could be bad RAM, but that address is in the ROM area so it could be a corrupted EPROM too.  Or something in address decoding logic causing thigns to be off.

 

If we apply  Occam's razor  to your observation about decoding logic, it's worth noting that a single faulty 74LS138 decoder in socket F12 could cause all the current symptoms.  In a previous thread tony359 discovered two faulty 74LS138 decoders that failed by activating multiple address regions simultaneously.

Assuming the APCO is similar to an Apple II, there should be four 74LS138 decoders on board:

• F12 - distinguishes between IO and ROM.  A faulty decoder in this socket could generate spurious keyboard input as seen in the video, or trigger those crashes (E002, E003, E005, E05e...etc.) by selecting multiple ROMs concurrently.
• H12 - distinguishes between slot ROMs and motherboard IO.  A faulty decoder in this socket could disable the keyboard or cause conflicts between devices in concurrent slots.
• F13 - decodes motherboard IO.  A faulty decoder in this socket could disable the keyboard or speaker, or generate a high-pitched whine while the computer is awaiting input.
• H2 - selects device IO for the slots.  A faulty decoder in this socket will prevent expansion slots from performing IO, even if their firmware is otherwise ok.  This decoder can be omitted if the expansion slots are not being used.

So a sensible test is to empty the expansion slots, borrow the 74LS138 decoder from socket H2 and use it to replace the one in socket F12.  If the symptoms change, then one of the decoders is faulty.  (NB: Verify that these chips are are all marked "74LS138" just in case APCO moved things around on their motherboard!)

S.Elliott wrote:

softwarejanitor wrote:

Could be bad RAM, but that address is in the ROM area so it could be a corrupted EPROM too.  Or something in address decoding logic causing thigns to be off.

 

If we apply  Occam's razor  to your observation about decoding logic, i

I like your plan S.Elliot.

1ajs wrote:

ill look for more chips to steel and continue with sourceubf more 4116s as i do prolly need to change them on other systems eitherway

The computer can work with just one row of good RAM, row C.  16K models were sold with rows D and E empty.  A common troubleshooting strategy is to take the chips out of rows D and E and swap them into row C until you find a workable set of eight chips.

Unfortunately that picture raises a troubling issue: there are only 3 ROM chips, but the original circuit is designed to select ROMs in 6 separate sockets.  If the clone manufacturer had followed good practice, they would have used diodes to combine the signals to safely drive the three sockets (ie: DTL) -- that's how Apple consolodated the CD and EF ROM sockets on the later Apple //e. 

But your motherboard doesn't have the six additional diodes that would have been needed, which suggests that the manufacturer might have lazily joined the ROM-select outputs from F12 in pairs.  Such a kludge might work temporarily, but it would put stress on the 74LS138 in socket F12 and shorten its useful life.

If you have a continuity meter, check for the following short-circuits at socket F12:

• Is pin 7 shorted to pin 9?
• Is pin 10 shorted to pin 11?
• Is pin 12 shorted to pin 13?

If these pins are shorted together, then the sad solution is to replace F12 with the newest 74LS138 you can obtain and keep spares on hand.  If these pins are not shorted together, then it'll be necessary to figure out how the motherboard combines those signals in order to fix those ROMs.  (Can you read the part numbers on the ROMs?  Maybe those are an uncommon type of ROM that has a different chip-selection scheme?)

1ajs wrote:

sorry but i have no idea where these sockets are  i see 4 sockets with the chip in them  but i have no idea witch ones witch theres 2 to the right of the roms and one on each end of the cpu

The sockets are identified by a letter-number coordinate system -- row letters are silkscreened along the left edge of the motherboard, and the column numbers should be silkscreened along the bottom edge.

Here's an annotated crop from your photo, identifying two of the decoders I mentioned and the addressable latch that jeffmazur mentioned earlier.

Socket F12 pins to test.png

PS: I managed to decipher the part numbers on those EPROMs in the ROM sockets.  They're Hitachi 2732 4KB EPROMs, which are no longer common but not too rare to find replacements if necessary.  But if those pairs of pins really are shorted together then it might be simpler/safer to remove the old EPROMs and install a ROM Replacement Board, or MultiROM card, or ROMX.

softwarejanitor wrote:

1ajs wrote:

ctrl reset it beeps eventualy throws up another error 

e005-  a=023 x=38 y=1b s=f1

 

 When it drops into the monitor like this, try typing E000L <return>

Probably will show E003 is  00

Send a screenshot if you can, but most likely the E0 ROM is bad.

S.Elliott wrote:

Unfortunately that picture raises a troubling issue: there are only 3 ROM chips, but the original circuit is designed to select ROMs in 6 separate sockets.  If the clone manufacturer had followed good practice, they would have used diodes to combine the signals to safely drive the three sockets (ie: DTL) -- that's how Apple consolodated the CD and EF ROM sockets on the later Apple //e. 

But your motherboard doesn't have the six additional diodes that would have been needed, which suggests that the manufacturer might have lazily joined the ROM-select outputs from F12 in pairs.  Such a kludge might work temporarily, but it would put stress on the 74LS138 in socket F12 and shorten its useful life.

 

On many clone motherboards, it was quite common to see three 2732s instead of six 2716s.  There is a butterfly connection to cut and maybe one or two half-moon pads to solder together.  

Here's an example of the mods needed for a Unitron board (which is the exact board the earlier post is showing)  as explained in Raymond Kosmic's clone assembly manual "Apple Seed":

kosmic.jpg

I have never seen reliability issues doing this modification.  I also sort of see the mod done on this board, but it's a bit blurry.  The jumper at "Z" is definitely there.

There they are:

Just like the books says.

board.jpg