Any Osborne I gurus?

[NutmegCT]

Hi folks.

In a different topic here several people gave me some good suggestions on troubleshooting my 1981 Osborne I. The power supply unit, and possibly some main logic board bits, had gone bad.

After a week of searching, I found a working (yes - I saw it work!) Osborne I. Watched it boot up into cp/m on A, then load SuperCalc on B.

So I paid my shekels and drove the six hours to bring it home today. Set it up, power on, saw the opening Osborne screen perfectly, with the Insert System Disk in Drive A text.

Inserted the System disk, hit Return, and watched the *top two lines* of the "Loading cp/m and Help" screen load. At that point the drive stops reading, and the screen slowly starts a random display of a vertical block (a graphic?) character.

Several power cycles, reboots, etc., all result in the same thing. It never gets to the A: prompt.

Tried a copy of the System disk, and the same thing happens, altho' the random graphic is different. As I can't get to the A prompt, I can't check the B drive.

Any suggestions - before I throw myself (and the new OCC) in the path of a moving train?

Thanks.
Tom
PS - interestingly, when the "new" OCC wasn't successful, I pulled the PSU from it and put it in my original OCC. That didn't change a thing. Weird.

 

[dave_m]

Tom, I am not familiar with the Osborne, but since it experienced a six hour drive, did you reseat all connectors, modules and socketed parts?

If that doesn't help, consider swapping the drives. It seems like it is almost working so hang in there.
-Dave

 

[NutmegCT]

Tom, I am not familiar with the Osborne, but since it experienced a six hour drive, did you reseat all connectors, modules and socketed parts?

If that doesn't help, consider swapping the drives. It seems like it is almost working so hang in there.
-Dave

Dave - thanks for the positive vibes.

Would you believe it, I actually got it to run perfectly - ONCE! Booted from original Osborne cp/m disk in drive A, loaded perfectly, and I could even do a DIR of drive B.

With a big smile on my face, I ran some cp/m utilities on the B disk, and - unfortunately - decided to try the boot process again.

Back to square one. Saw the opening Osborne screen, started loading the system files from A, screen changed to "Loading Osborne CP/M and Help" on the top two lines, then drive stopped, drive light went off, and a few random graphic characters started slowly appearing across the screen.

Powered off, unplugged, reseated every single removable chip and plug in the entire system, even swapped the two disk drives. No change.

This is strange!

Thanks.
Tom

 

[barythrin]

I'm not sure what'd be causing the bad characters to slowly go on the screen. I wonder if the floppy drive just needs to be cleaned. Gibberish on the screen is often bad RAM or video ram though (on most common computers, not necessarily an Osborne).

Could also be an overheating problem which they were prone to do which might explain why it works first time but warms up and doesn't. You may be able to get further performance by having a fan blow on the system just to check.

 

[tezza]

Hmm, yes I love my Ozzy but they can be temperamental. I've had three issues with mine...

• Capacitor blowout in power supply (alarming bangs, smoke)
• Bad RAM (garbage characters and non boot)
• Buzzing disk drive causing power fluctuations (and hence a crash)

You can read about all those three things from the links futher down on this page.

However, your issue doesn't sound like any of those problems. Osbornes can get RAM issues as I found with mine but (like most ICs) RAM is usually damaged or it's not. I'm wondering if it's a dry solder joint on the circuit board somewhere or maybe, just maybe a faulty capacitor? I had an Apple clone that had a problem where it worked for about 15 seconds, then totally lost the plot. I tracked the problem down to a small bluecap capacitor which failed once it warmed up. Your Osborne starts to boot then fails, so I'm wondering if it's a similar thing?

These kind of problems are not easy to diagnose and the only way I discovered mine was to go back to first principles and trace the fault from the CPU. I was lucky in that, given where the faulty component was, I could isolate the circutry involved quite quickly. That would not be a case with a lot of things though.

If interested, you can read about this Apple clone/capacitor adventure here.

Good luck. Having two Osborne 1s I'm sure you can get at least one working machine out of the both of them.

Tez

 

[NutmegCT]

Hi Tez - there's a goodly number of ram chips and capacitors on the board. How did you find the faulty one? As they're soldered in, I'd imagine they're not all that easy (!) to trace and replace.

Tom

 

[NutmegCT]

An interesting tid-bit. I tried booting from the B drive (the SHIFT+") combination. No change.

So on the off chance the drives were failing, I swapped with the drives in my first machine (the one with the dead PSU).

Now after getting the power-on Osborne "insert system and press Return", regardless of which drive I boot from, I get "BOOT ERROR" scrolling down the screen when the drive tries to read the system disk.

I *guess* this is progress, as my original Osborne (with the failed PSU) wouldn't even get to the initial Osborne screen.

Both machines are single-side single-density.

Any thoughts?
Thanks.
Tom

 

[NutmegCT]

PROGRESS!

I went back to the original drives, and no longer get the BOOT ERROR message. Don't have a clue why, but now I get the "Loading CP/M and Help" message again when I boot from the system disk.

As usual, the second screen has "cp/m and help" at the top, and "freezes".

BUT I discovered that regardless of what key I hit when that screen is "frozen", I get a rapid display of the Osborne "Help" screen. Hitting any key seems to change that help screen. If I hit the ESC key I actually "drop" to cp/m - and can do a DIR of the B drive, change to the B drive, and run xdir on the Utilities disk in the B drive.

Reading the manual, I now remember that the Help screens are the "initial menu" when cp/m is loaded.

So it seems there's something hanging in that Help section. By the way, I tried this with a second (backup) system disk, and get exactly the same result.

This is progress - just not very far down the road.

Does this "hanging" in the Help system indicate anything obvious to you experienced folks?

Thanks.
Tom

 

[dave_m]

If I hit the ESC key I actually "drop" to cp/m - and can do a DIR of the B drive, change to the B drive, and run xdir on the Utilities disk in the B drive.

This is great news that you discovered ESC gets you to the system prompt. At this point perhaps you should try to test your RAM. Do any of your floppies contain a memory/RAM test or some diagnostics? Also before you get too far, you should check the +5V power to ensure it is well regulated and with minimum ripple/noise.

 

[tezza]

Hi Tez - there's a goodly number of ram chips and capacitors on the board. How did you find the faulty one? As they're soldered in, I'd imagine they're not all that easy (!) to trace and replace.

Tom

Well, regarding identifying faulty RAM I used the old hardware hackers technique of piggybacking a good RAM IC over each RAM on the board, switching on and seeing if it made any difference. Whether it works with faulty RAM depends very much on the nature of the fault in the IC itself. It worked that time for me though.

More about it (and other Ozzie repairs) here (halfway down the page).

Tez

 

[NutmegCT]

Aha! Now we're getting somewhere - a memory test.

I can borrow a ram test on a double-density disk. I also have an Osborne friend who can lend me pair of double density disk drives, and a double density card.

What do I need to do to change my OCC1 to double density? I see where to put the DD card on the logic board, and where to put that short cable that goes from the DD card to the main board. Are there other changes needed? Any jumpers to change? Does another chip need changing? My two ROM chips say "Rev D" and "REV 1.4 7/26/82" (one in my original OCC and one in this "replacement" OCC).

Thanks.
Tom

 

[dave_m]

Troubleshooting

Troubleshooting

Tom,
I'm not sure that you should make so many changes to the current configuration until you know more about the problem. Can your friend make a single density copy for you with the test program on it?

If not, there are plenty of other things to check. You should make sure you have good clean +5V and also check the DC power to the disk drives. Keep a good log of all changes that you make so that you can return to the original configuration if necessary.

 

[NutmegCT]

Dave - he's got the drives and the card, but doesn't have a working Osborne. So, unfortunately he can't make a single density disk for me.

I did, however, get the two DD drives and DD card (and cable) installed, and along with the 1.4 ROM chip I've now got a working DD machine. Still has the same boot problem, but now I can boot from a double density Osborne system disk.

Symptom remains: Power on, screen asks for system disk. Insert system disk and hit Return: screen shows Loading CP/M and disk reads. Screen changes to top two lines reading "Osborne Computer and Help 2.1", then hangs there, with various random graphic characters slowly appearing on the screen.

I tried the original single density drives and boot sequence this morning after a long night of rest for the Osborne. No change at all, so it seems that heat has nothing to do with the problem.

I'm really feeling this is a RAM problem (boot, drive, video weirdness) as others have described. But 'til I can get a clean boot again I can't load that diagnostic disk. And of course, Osborne discontinued the built-in diagnostics when they started using ROM 1.3. Used to be you could hit CTRL+D at the "insert system disk" prompt. No longer.

I am making progress - but now have to figure out how to troubleshoot faulty RAM chips.

Edit: Dave - where exactly do I check for that +5V? I do know that the PSU is putting out +5V - if that's what you mean? Thanks.


Onward through the fog!
Tom

 

[NutmegCT]

Gents - an update with good news.
I found one of my old DD disks from the early '80s. It was marked "Diagnostics SYS" on the label.

I figured that SYS meant that I had put the cp/m OS on that disk along with whatever the diagnostics were. I put the disk in A, powered up, hit Return when the screen asked, and voila!

I got a clean boot - with NO help screen or garbage. I'm now running a FOG utility called MemTest. I don't have any docs on that app, but I'll report back after it's finished.

So far - 5 minutes into the run, screen shows 4139 refresh failures in something called Location UB - 27. Have no clue how to interpret that.

Stay tuned.
Tom

 

[NutmegCT]

OK - final result of the memory refresh test: no failures *except* Location UB:27. That location had 99999 failures (I assume that's not good).

So, looking in the OCC Tech Manual, and reading that memory is physically located on the logic board in four ranks (UA, UB, UC, UD) with eight files (20 - 27), I *assume* that the eighth chip in the second rank is the failure.

Could someone confirm I'm interpreting this correctly?

If so, I assume that the next step is to replace that chip. I have plenty of extras - from my original Osborne's logic board. I'd need to figure out how to remove a chip, and how to insert it to replace the faulty one in my replacement board.

Thanks.
Tom

 

[dave_m]

Very good. Yes I think you have the right chip location. The system board may have some markings to confirm the IC locations. The chip in question seems to be a dynamic RAM meaning it needs to be refreshed (read out) once in while by the system hardware to maintain its data contents. It seems that chip at UB27 is not retaining its data. Hopefully the chip is in a socket. If not, you will have to remove it. After that you may want to install a chip socket to make it easier to install the new chip. What type of RAM chips does the Osboune use? Is it a type 4116 (16K X 1) or similar?

Are you good at soldering? If you are not an experienced hand at soldering, I would advise you to play it safe. By that I mean it may be better to cut all the leads off towards the top of the chip before you attempt to unsolder the individual leads from the system board. This will reduce the amount of heat necessary to unsolder each lead.

Others on this forum can explain better than me about the use of solder wick, solder suckers and other aids to remove the leads and clean the plated through holes on the system board.

 

[tezza]

Tom,

Sounds like you're making progress. A RAM problem could certainly account for all those weird problems. It looks like you've identified the correct IC but I'm not really an Osborne guru and it would be good for someone who is to comment. You don't wnat to be pulling out the wrong IC.

For desoldering ICs I've found solder wick works well for me. However, it's easy to damage the tracks around the IC when extracting chips from spare board to place in working ones. This is no problem if it's a junk board but if you want to keep it as a possible "good" spare, I would see if you can source a replacement from a junk board or the net. From memory they are 4116 ICs yes? These were common in many machines of the era.

To extract the faulty IC, just carefully clip each pin off the the chip then extract the pins one at a time using a soldering iron and pliers. Use solder wick to clean up the holes.

Similar to what I did for my IBM PC as described here.

Tez

 

[dave_m]

Tom,
Before you unsolder the chip, carefully examine the solder joints on that chip with a magnifying glass in a good light. If you see a joint that is not shiny, it may be a cold solder joint that is making intermitent contact. Reflow the solder joints to make good connections.

 

[dave_m]

You don't wnat to be pulling out the wrong IC.

For desoldering ICs I've found solder wick works well for me. However, it's easy to damage the tracks around the IC when extracting chips from spare board to place in working ones. This is no problem if it's a junk board but if you want to keep it as a possible "good" spare, I would see if you can source a replacement from a junk board or the net. From memory they are 4116 ICs yes? These were common in many machines of the era.

Similar to what I did for my IBM PC as described here.

Tez

Tez, that's good advice from a guy who has proven he can do it. As for myself, I usually make a ham-fisted muck of such repairs.

 

[NutmegCT]

Gentlemen - your advice and encouragement are priceless. Thank you.

All RAM chips are soldered into the board.

The "bad" chip is marked NEC D416C
The replacement (same position on the other board) is marked AM9016FPC 8228GMP

Is that a match?

T.
PS - by "piggybacking" do you mean simply holding one chip over another so all the upper chip's legs touch the lower chip?