More shooting in the dark with Apple IIs

[tezza]

In a previous post, Mike Willegal says the following about my Apple II+s...

If the video display is stable, your DRAM should be ok, though there is the still the possibility of bad bits. Once the machine is up, you can run a DRAM test, such as the one included in the programmers aid ROM. Connect a speaker and see if you are getting a beep when powered on. The power on beep routine is called almost immediately after boot. If not beeping, the processor is having issues reading and writing memory.

Most likely source of problems is the address decoding logic, but this can be tricky to figure out, without the right tools. I've had lots of issues with flakey sockets on these old boards due to the poor quality sockets that Apple used. This is more likely than a having a bad chip, though either or both is possible.

With all 3 boards the video display is indeed stable. I'm starting to wonder if it is a bad socket or two? The system on startup doesn't even get to the beep on any of the boards.

The screen below shows the pattern on the Europlus. Anyone recognise it? It's stable and doesn't flicker. If I remove the CPU and switch the machine on I get the same pattern. I've tried subsituting those chips around the CPU and PSU but to no avial.

The other two machines show just the white vertical bars. It's unlikey, but maybe the same chip(s) are damaged in all 3 of them!

Tez

 

[tezza]

There are some useful hints in this google groups post.

Maybe it's a transistor?

Is a logic probe any different from a standard multimeter measuring voltage?

Tez

 

[david__schmidt]

There are some useful hints in this google groups post.

Maybe it's a transistor?

I was going to suggest the transistor manning the reset line to the 6502. But I see I suggested that in the above usenet post too. ;-) Here's another diagnostic procedure relating to that reset line.

 

[mwillegal]

The picture you attached to your post is pretty unusual with a mixture of color and text modes interfering with each other. I would look first at the systems with the plain white vertical bars. Plain white bars may indicate that dram video data is not being serialized into text properly by the video circuit. First places I would check are the 74166 at A3, the 74LS194 at A10, 74S86 at B2. Also could be the 2513 character generator which takes multiple voltages, including -5 and -12. Are you using same power supply to test all the boards? If so, check all voltages. Could also be problems closer to the DRAM, but I think that this is less likely because data is in parallel format and you would more likely see less consistent patterns.

Regards,
Mike Willegal
www.willegal.net

PS. later rev boards use a PROM for character generation, not sure which boards you have, but similar issue could exist

 

[tezza]

I've checked the voltages on the three PSUs I'm using. They are all fine. However, I have not checked them under motherboard load. To tell you the truth, I'm actually not sure how to do this? Where on the board should you put the +ve and -ve probe so it will truely reflect what is being delivered by the PSU under load.

Sorry if that seems a pretty basic question. It probably shows how much I don't know. However, I figure If you don't ask, you'll never know.

I'm going to have a good look at the apple technical manual tonight, which might actually tell me, but if anyone knows off the top of their head, I'd welcome a post.

Tez

P.S. I find if I flex one of the clone boards slightly the display changes from vertical white bars to snow. Some dodgey connections/sockets somewhere to boot.

P.P.S Am I ready to give up yet? No way!

 

[tezza]

I've checked the voltages on the three PSUs I'm using. They are all fine. However, I have not checked them under motherboard load. To tell you the truth, I'm actually not sure how to do this? Where on the board should you put the +ve and -ve probe so it will truely reflect what is being delivered by the PSU under load.

Having just studied some schematic diagrams I think I've answered this question. There are a number of points on the board which indicate 5V+ and GND and I guess this is where I put the leads.

Tez

 

[mwillegal]

The easiest place to check voltages may be right on the chips. Put one meter lead on ground pin and the other lead on the pin connected to the supply voltage.

If you have 2513 character generator, ground would be either pin 10 or 11. You can check -12 volts on pin 1 (bottom left), -5 volts on pin 12 (bottom right) and +5 on pin 24 (top left). The only other voltage in the system is +12 which can be checked on pin 8 of the DRAMs.

 

[tezza]

Success at last!

Success at last!

Ok, I've had a good night.

I took your advice Mike and started to work on one of the clones rather than the Europlus which until now has been my main subject of attention.

First i measured the voltages on the board to check the PSU was working properly under load. These where fine. Then I measured the voltages on the CPU to determine in a RESET was happening. It didn't seem to be, but then I remembered that with some keyboards, CNTRL/RESET rather than just RESET is necessary. Indeed, doing this showed a drop to 0v. A RESET was taking place.

Hmm...looking at the pins on the CPU while measuring the voltage I noticed many were very rusty and corroded. On a hunch I swapped the CPU for a non-corroded one. This time, switching on produced a (feeble) beep and the picture below....

Progress! Still not out of the woods by a long shot though.

I was getting video, and could hear a strangled beep. Occassionally I'd even see characters on the screen. I figured this was likely now to be a RAM problem. Which chips were faulty though?

Then I remembered that I had 2 language cards that came with the machines. These had 9 and 8 DRAM chips respectively and because they were sitting vertical in the cases, had escaped being drenched in mouse pee. I found the cards and extracted their RAM (see pic).

I then replaced the RAM in the Apple II with this RAM, making sure I used the same type of 4416 for each of the three banks, just to give myself the best chance. For the final bank, I selected the cleanest chips I could find from the ones I had been using. I flipped the switch and I saw this...

At last!! It works!

Note the message. "READY GO" rather than APPLE ][+ (lol). Well, I guess it is a clone after all.

So now I had a test board. I put it to good use testing the RAM I had been using in the clone and the Europlus prior to the chip transfusion from the language cards. Two of the clone RAM chips and 4 of the Euorplus RAM chips were faulty . No wonder I was making no progress on the latter!

I filled in the Euorplus with known good RAM and fired it up. Hmm..all I got was a single white screen (No coloured blocks)? Well, perhaps that's progress. Now I have a working machine I can tackle this tomorrow, and the other clone also. Just before I called it an evening, I substituted the character generator just in case. It made no difference, so it wasn't that.

So, a good night's work. Thanks to everyone both publically and privately that have helped with this. Still a long way to go, but progress is being made.

Tez

 

[Druid6900]

Tezza,

You have just learned two valuable traits when troubleshooting a computer; Deductive logic and Intelligent substitution.

I expect that you will have them all up and running in no time.

 

[tezza]

Yep, you also learn patience.

I've now got the second clone working. It had 4 defective chips excluding the RAM. Mouse urine and time must really do something to them, although they could have been non-working before being stored.

I've checked the chips in the Europlus. Found a handful of defective chips there as well including the character generator. It has some chips that are not mirrored in the clone that i suspect are faulty too. Working on replacements.

One question for anyone who might know. Should individual clone ROMS be interchangable with geniue APPLE Roms? I would have thought yes, but all bar one of the APPLE ROM chips fail to work in the Clones when substituted one at a time to their corresponding positions. Only one (ROM 4) works. Either they are faulty or they can't be subsituted. Disappointingly, the Auto-boot ROM is one that doesn't work, which means I may never get to see that APPLE ][+ boot text.

I can't use the whole set of 6 Apple ROMS as I broke a leg (including the lug) off ROM 2 when I was extracting it. The APPLE ROM pins are pretty corroded.

Hopefully it will boot with a full set of clone RAM once I've got it going. "READY GO" just doesn't quite have that impact value though. :D



Tez

 

[Druid6900]

Well, Uric acid, as the name implies, IS an acid and it's going to corrode (rust) whatever metal parts it comes in contact with.

Copying ROMs, especially when the machine is current, is, of course, a Romper Room No-No, so, the board could be wired so that ROM2 is actually E0, et c., ya know?

However, I HAVE heard rumours, only rumours, mind you, of people downloading images of the ROMs used in emulators and burning them to the appropriate chips, but that, of course, would be illegal, wouldn't it?

 

[mwillegal]

clone ROMs or PROMs may or may not be electrically compatible with Apple original ROMs. It really depends upon the specific PROM chip selected to be used by the manufacturer. There are several variations of 2Kx8 ROMs and PROMs used during those years and not all are equal. You may be able to build simple adaptors for some of the variations.

see this page for some links and additional information.

http://www.willegal.net/appleii/appleii-integer.htm

Regards,
Mike Willegal

 

[tezza]

Well, Uric acid, as the name implies, IS an acid and it's going to corrode (rust) whatever metal parts it comes in contact with.

It does a most effective job. i could salvage only about 1/2 the chips in the machine that was obviously the "family" room for for the mice. The pins simply fell off the chips or remained in the sockets. A transistor also fell off the board when touched.

However, I HAVE heard rumours, only rumours, mind you, of people downloading images of the ROMs used in emulators and burning them to the appropriate chips, but that, of course, would be illegal, wouldn't it?

Surely not! Tsk tsk tsk....

see this page for some links and additional information.

http://www.willegal.net/appleii/appleii-integer.htm

thanks Mike, interesting page.

The auto-boot ROM is now the limiting factor in ressurecting these machines. Out of the 5 Apple II + -type machines I have, only two of the Auto-boot chips pulled from two of them actually work (the Apple one may not work because it expects all 5 other Apple ROMS, not a clone substitute in the bunch). So this means I can only restore two machines to a working condition. Pity, as once the Europlus is repaired I'm likely to have a third board with everying working except for that autoboot chip.

As I'm not hard-core enough (yet) to burn eproms I'll just have to add APPLE II+ ROMS to my ebay search string

Tez

 

[tezza]

Tonight I attached two disk drives from the haul, to two of the Apple II+ clones I'd repaired to see what would happened.

The good news is that both drives worked, and loaded DOS 3.3 without a problem. That being the case I started to go through the stack of disks that came with the load. As expected most were too degraded to get much off and after I had to clean the heads due to a particularly sticky disk I decided it really wasn't worth it. I didnt want to damage the drives. I might just dump the old disks rather than persisting with them.

I grabbed some on my (new) Apple IIe disks to see if my copy of Apple Panic would load.

On the first clone it looked like this...

Arrg!! The zigzag screen seems to only appear in graphics mode. Text mode the display is fine. Any ideas what IC is not working properly here?

On the second clone it looked like this...

Better but no colour? Should I be tweaking one of those pots on the motherboard to see if it clicks in?

Both seem off centre to the screen. I guess this can be adjusted somewhere too. I've already twiddled the monitor knobs and that orientation was as good as I could get.

Still a few more sessions before full health is restored it seems.

Tez

 

[tezza]

Ok, I think I have both these clones fixed now. The "snowy" hi-res picture was caused by a faulty RAM chip servicing the memory location of one of the graphic pages. Neither machines seems to show colour but maybe it's the software I'm trying to run or the Apple IIe monitor I'm using? Hard to believe BOTH machines would be faulty in this area.

Here they both are. I wish I'd taken a before and after picture. You wouldn't believe how disgusting they looked when they arrived! The picture below shows a clone with an APPLE II-type case. It's lost it's badge but it was a Redstone (according to the guy who gave it to me).

This pic shows the insides.

This next one is a clone called an RX-8800. This one is a good candidate for the peroxide treatment, but I can imagine it would have made a nice machine in it's day with its numeric keypad and all.

There were two other RX-8800 in the haul but these have been scavanged for working power supplies and chips to make the two complete units above (plus the Europlus).

Now I'm just waiting on specific chips for the Platinum and the Europlus and these Apple IIs will be done. Won't have finished with the haul even then though as there are dozens of cards, books, software and other machines to shift through.

Tez

 

[Tinkerer]

Fantatsic! Good on ya Tez.

 

[Druid6900]

There is another valuable lesson.

Just because a computer boots doesn't mean that the RAM is all good, just that the amount that it needs to boot is good.

This old stuff doesn't have power-on diagnostics, so, only when you load something that uses more memory do you find out you have a stuck bit.

 

[tezza]

There is another valuable lesson.

Just because a computer boots doesn't mean that the RAM is all good, just that the amount that it needs to boot is good.

Yes, actually it was a lesson. I had made that very assumption i.e. the computer booted and I could do a few things in APPLEBASIC so all RAM was good. Consequently the first thing I looked at was all the chips that handled all aspects of video. After a couple of hours of frustration exchanging all those chips, and then ALL the chips EXCEPT the RAM, I was at a loss. I was starting to think it was a static component, maybe a capacitor or transistor?

Then, thinking through the problem, I remembered I'd read somewhere that part of the RAM is dovoted to Hi-RES pages. The problem only appeared with Hi-Res graphics. Ah ha, a light in the brain flicked on! I turned to the "Apple Reference Manaual" and sure enough, a bank of RAM occupying location 32,000 and above is devoted to Hi-RES graphics. Figuring this was in the middle row of the RAM bank I started substituting chips there. On chip four, I hit the fault.

So it took me much longer than it should have due to a faulty hypothesis. One the other hand, I have learnt a valuable lesson I can take to other diagnostic problems.

Fixing these old machines has taught me a lot. I'm not sure how much of it is useful to modern day living, but it certainly is satistfying restoring these items to their former working glory.

Tez

 

[willowmoon93]

But the really really cool thing is that you have that game "Apple Panic" playing on the RX-8800 -- definitely a classic game for me, a personal favorite (even if it pretty much is a blatant clone of "Space Panic"). Cool !!!!

 

[Druid6900]

I learned, the hard way, a long time ago to always rotate ALL the RAM in a system through the first bank to make sure that it's working all the way up.

It might take a little more time right then, but, as you found out, it would have saved a lot of time later on.

Unfortunately, the chip tester I have, which does DIP DRAM, only goes down as low as 64K x 1 chips and I have spent a lot of time searching for either a device or schematic of a device that would test 16K x 1 chips.