Apple IIe Failing RAM Test

11 posts / 0 new
Last post
Offline
Last seen: 10 years 2 months ago
Joined: Apr 19 2011 - 15:11
Posts: 6
Apple IIe Failing RAM Test

Hi,
I need a help for my apple IIe, starting the self-diagnostic test I obtain on my monitor the following string:
RAM: F13 F12 F11 F10 F9 F8 F7 F6
Or sometimes
RAM:

I have changed the following ICs:
SN74LS125N
74S109
SN74S10N
and ONE 4264 (DRAM)
I have also cleaned the pins of the ROMs and MMU
But the problem still remain.

Note that I'm able to write basic instructions and the apple IIe run them, but none of my software on floppy works:
-UNABLE TO LOAD PRODOS
-sometimes I obtain
A1DD- A=28 X=FA Y=40 P=38 S=E6
*

Please can you help me,
Alex

Offline
Last seen: 4 days 8 hours ago
Joined: Dec 19 2003 - 18:53
Posts: 903
Re: Apple IIe Failing RAM Test

Based on what you describe, the computer CPU seems fine. It seems like your disk or drives or both might be the problem. Do the head cleaning, drive speed adjustments (both procedures have been previously described in this section of the forums), and use a good known disk to test the drives. Or try a drive and disk that are known to be already working.

Mutant Pie

Offline
Last seen: 10 years 2 months ago
Joined: Apr 19 2011 - 15:11
Posts: 6
Re: Apple IIe Failing RAM Test

I tried with 3 different drives,2 controllers and about 20 floppy disk.
I've obtained the same results,i've changed all the electrolytic capacitors too.

Offline
Last seen: 7 years 5 months ago
Joined: Aug 3 2011 - 07:50
Posts: 33
Re: Apple IIe Failing RAM Test

What led you to diagnose the problem as that one DRAM in particular?

Try running a memory test: http://www.willegal.net/appleii/6502mem.htm

It'll tell you what addresses are failing, and what the expected versus actual result was (so you can see what bit, and therefore what chip, is failing.)

Offline
Last seen: 10 years 2 months ago
Joined: Apr 19 2011 - 15:11
Posts: 6
Re: Apple IIe Failing RAM Test

I thought that if really was a ram problem, changing one of the chips, the message would have to change.
(not all the ram chips should be displayed at the end of the test)

Now I'm going to follow your link. tx. I will report the results asap.

Offline
Last seen: 7 years 5 months ago
Joined: Aug 3 2011 - 07:50
Posts: 33
Re: Apple IIe Failing RAM Test

The DRAMs are one bit wide.

So, you can have a failure at a certain address, and the bit within the byte that is wrong points to the chip that has failed.

Offline
Last seen: 10 years 2 months ago
Joined: Apr 19 2011 - 15:11
Posts: 6
Re: Apple IIe Failing RAM Test

But...
What exactly means the message "RAM: F13 F12 F11 F10 F9 F8 F7 F6" ?
That all the RAM chips doesn't works?
If the answer is "YES", changing one chip the message MUST change...
If the answer is "NO", it's a nightmare... this means that the TEST it's a fake.

amauget's picture
Offline
Last seen: 1 week 3 days ago
Joined: Jan 3 2011 - 11:34
Posts: 317
Re: Apple IIe Failing RAM Test

Hello,

According to Apple website, there is no doubt about what the message means.
But I agree with you : if you change one RAM chip, one error should go.

So it seems more complicated than it is. There is obviously another problem.
The RAM is tested at the end of the diag so apparently all the other chips are good.
Are the chips into sockets ? If so and if you have another known good Apple //e, I suggest a MMU/IOU swap.

Offline
Last seen: 10 years 2 months ago
Joined: Apr 19 2011 - 15:11
Posts: 6
Re: Apple IIe Failing RAM Test

Ok, so may be the RAM has no problems (but we are not really sure)
Obviously I have already removed and reinserted all the chips into they sockets.
The RAM chips are soldered directly on the motherboard, so for testing the RAM, I replaced one of them using a new socket.

But we must go forward.
I have bought one IOU from ArcadeComponents.com, but not exactly the same because seems impossible to find: in my Apple IIe there is a 344-0022-A, but I have bought a 344-0021 (they told me this is an international version of the chip)
After changing the IOU chip the computer is not able to start the Self Diagnostic Test.
Trying to start the Diag pressing the keys or leaving the keyboard disconnected result in a blank screen or sometimes displays a random number like C431* or C423*.
Unfortunally I'm not able to find another Apple IIe for other tests.

If someone has other ideas or tests to suggest me, I tell you that I have also a 4 tracks oscilloscope and a signal generator. Thank you in advance

Offline
Last seen: 9 years 1 month ago
Joined: Mar 30 2011 - 22:00
Posts: 7
Re: Apple IIe Failing RAM Test

Alex:

Try changing the RAM I have had many of these go bad. It could be one of the Chips on the RAM has gone bad not unusual for them to go bad.

amauget's picture
Offline
Last seen: 1 week 3 days ago
Joined: Jan 3 2011 - 11:34
Posts: 317
Re: Apple IIe Failing RAM Test

Hello Alex,

There is a lot of confusion about the IOU.
Check out Tezza's article. The dead link "this other site" was pointing to this article :

The Case of the Missing Pixels By David Wilson Copyright (c) 1995 Apple Users'
Group, Sydney Republished from Applecations, a publication of the Apple Users'
Group, Sydney, Australia.

david@cs.uow.edu.au

With versions 4 and 5 of the popular AppleWorks integrated package requiring
256k of RAM, many Apple //e owners replaced their 64k extended 80 column cards
with a combination 1MB RAM and 80 column card from Sequential Systems called the
Meg80z. Any owner of a PAL //e immediately noticed that the top left hand corner
of their screen just wasn't being displayed correctly. David Wilson has found
out why.

The Case of the 7 missing pixels or Meg80z + PAL //e = trouble

The Problem: When used with a PAL Apple //e, the Sequential Systems Meg80z 1MB
Auxiliary Slot RAMcard suffers a minor but annoying problem: the first 7 pixels
in 80 column or Double Hi-Res mode are not displayed properly.

How to demonstrate the problem: The following commands will quickly show the
problem on a 50Hz Enhanced Apple //e equipped with a Meg80z:

]PR#3
]INVERSE
]HOME

The screen should be a white (or green or amber depending on your monitor)
rectangle with a ] on the second line. Instead, the rectangle will have a small
notch taken out of it in the top left corner. To illuminate the missing pixels
type:

]CALL-151
*C073:3 I C055:0 I 400:20 I C073:0

If you now type control-C to re-enter Applesoft and then type

]NORMAL
]HOME

You will find that the missing pixels are now white while the rest of the screen
has returned to black.

The Explanation: The Apple // series of computers are designed to display their
video output on a colour TV set. This would be fine except that there are at
least 3 different colour TV standards in use in the world: NTSC (used in USA &
Japan), PAL (used in Australia and most of Europe) and SECAM (used in France).
As the original Apple ][ was designed in the USA it used NTSC to generate the
colours and it output a video signal compatible with the TV sets used in America
(using a vertical refresh of 60Hz). When the Apple ][+ was introduced Apple
decided to support users in Europe (and Australia) by selling the Europlus. This
was an Apple ][+ with jumpers changed on the motherboard to select 50Hz vertical
refresh, a new main crystal to more closely match the 15625Hz horizontal refresh
and a PAL colour card to encode the colour information using the PAL standard.

When the Apple //e was introduced, Europe and Australia got an entirely
different motherboard to the American version. This //e had the PAL colour card
integrated into it so that slot 7 was not wasted for the colour card. Also,
rather than having jumpers to select the vertical refresh frequency, the IOU
chip was built in two different versions - one for 60Hz and one for 50Hz.

Now comes the problem - the Meg80z was designed using an American //e and was
apparently never tested on a PAL //e before being released. So why does the
video refresh frequency affect the Meg80z? The answer again has its roots in the
past.

In order to display a picture on a TV screen the video display must be
continuously refreshed. Dynamic RAM chips also need to be refreshed to prevent
the information they hold from fading away as the tiny capacitors that make up
each memory cell discharge.

The Apple ][ utilized the video refresh to also refresh the dynamic RAM. By
carefully arranging the address lines used to access the video memory it was
possible to get the dynamic RAM refresh for free. This explains the seemly
bizarre non linear screen memory map.

The //e automatically refreshes both the main 64K and the auxiliary 64K as it
creates the video display. This is however insufficient to refresh the full 1MB
so the Meg80z must do some refreshing of its own. The only time available for
this to occur is during the vertical blanking interval.

Auxiliary slot memory cards treat their memory as a number of banks of 64KB (16
for a 1MB card like the Meg80z). To access a different bank the program must
select it by writing the bank number to location $C071 (or $C073 depending on
the card). The only tricky part is that the video display data must always come
from bank 0 otherwise the 80 column display would be messed up each time the
bank was changed.

The problem occurs because the NTSC and the PAL standards have their vertical
blanking interval at different stages. It appears that the Meg80z logic is
confused by the extra 100 lines that a PAL //e has per screen.

The Solution: To fix the problem, you'll need to install the IOU used in the
NTSC Apple //e (part number 344-0020). If your Apple //e has a socketed IOU
chip, you can simply purchase the new chip and replace the old one.

Platinum //e owners will find that their IOU is soldered. The process of
removing the old chip is better left to the experts which unfortunately means
that the labour cost will probably make the solution unfeasible for you.

The part numbers in question are:

344-0020 //e 60Hz NTSC IOU 344-0022 //e 50Hz PAL IOU

Do not confuse them with the Apple //c IOUs - they are different again.

344-0021 //c 60Hz NTSC IOU 344-0023 //c 50Hz PAL IOU

These chips are listed on the Apple Parts list. So any Authorised Apple Service
Centre should be able to help you. The chip will cost approximately $45. The
labour charge for replacing a soldered chip will be approximately $80.

Replacing the IOU chip with the 60Hz version does have some disadvantages. The
video output will no longer be able to be recorded on standard PAL VCRs and some
colour TVs may not be able to lock onto the picture giving a rolling display.

The IOU chip can be obtained from an Authorised Apple Service Centre. One such
service centre is: IC Technologies 124 Forest Road Hurstville NSW 2220

For those who are interested in upgrading their Apple //e's and who don't mind
missing seven pixels, Sequential Systems' Meg80z is a 1 megabyte, auxiliary slot
memory expansion and 80 column card for Apple //e computers. The Meg80z is
designed to be fully compatible with software programs that are designed to
recognise large extended memory additions. Sequential Systems 1200 Diamond
Circle Lafayette, CO 80026
303-666-4549

BTW, with an oscilloscope, you should be able to track down the problem using the Apple //e SAMS book as a guide.

Log in or register to post comments