• Please review our updated Terms and Rules here

Osborne 1 resets itself continuously, but only when the keyboard is plugged in!

wowbobwow

Experienced Member
Joined
Mar 20, 2015
Messages
110
Location
Bay Area, California
As mentioned in a previous thread, I'm working to get my three Osborne 1's up and running. All three power on currently, but one of them (the one with the oddball "MICRO CZEK" ROM noted in my previous thread) is now being super weird:
  • If I power it up while a keyboard is connected, the machine just resets itself continuously
  • If I power it up with no keyboard connected, it behaves normally (but of course I can't hit RETURN to actually boot without a keyboard plugged in)
  • If I power it up with a keyboard connected and it goes into the endless-reboot-cycle, sometimes when I mash down on the keyboard it'll stop resetting and either freeze, or boot long enough for the CP/M prompt to appear, at which point the screen fills with garbage characters
Is this a known behavior with the O1? I'm a beginner-level technician so forgive me if I'm using the wrong lingo here, but it sure "feels" like there's some kind of short circuit happening with the keyboard controller - is that a typical fault? Here's a short video I made showing some of this odd behavior:

Thanks for any guidance or suggestions here, I'm stumped!

Huxley
 
The keyboard is a matrix. My Osborne 1A does the same. It seems to be a keyboard issue. If you mash the keyboard it might boot. I haven't looked into the key issue yet - Another working O1 keyboard from the Osborne 1B should work with it.

Remove the keyboard and jump the enter key to get it into boot mode. If you have a failed controller to the disk drives, it will take minutes to come up with the boot error issue. I have yet to fully solve that problem and am still waiting on parts to arrive to fix mine.

They keyboard is just a matrix between the data lines and the address lines, so all kinds of bad stuff can happen when the keyboard isn't working properly since it's on the data and address lines directly, which can cause issues. There is no keyboard controller.

On mine, I manually activated the FDD and was able to boot even though the disk select line had failed on my machine, but if you're not getting the boot error after a minute or so, it's not a select issue - it's that the enter key isn't working.

You can also try Shift Quote to switch the A: to the right and initiate boot without pressing enter.

Hotwiring enter is another method - Third pin from the left, Second pin from the right, both on the top row is enter. You can use dupont connectors.

DONT try this without a proper connector - eg, with a bit of wire - because there's 12v next to one of them, and if you short out 12v to a signal line you will blow the chip at best, and potentially worse.

Here's a photo.

hotwire.jpg
 
You can also check the key matrix with a multimeter.

The most likely cause of your issue is a shorted switch contact - so you are going to have to fix the keyboard anyhow!

Dave
 
You can also check the key matrix with a multimeter.

The most likely cause of your issue is a shorted switch contact - so you are going to have to fix the keyboard anyhow!

Dave

True, but it's nice to check if it's booting first.

Oh, and if you haven't checked the filter caps on the PSU, turn it off and check them. They WILL blow if they have not already. The RIFA filter caps will not survive old age... And they can cause damage to other components if they hit it with a molten jet of metal, which is what they shoot out when they smoke.
 
The keyboard is a matrix. My Osborne 1A does the same. It seems to be a keyboard issue. If you mash the keyboard it might boot. I haven't looked into the key issue yet - Another working O1 keyboard from the Osborne 1B should work with it.

Remove the keyboard and jump the enter key to get it into boot mode. If you have a failed controller to the disk drives, it will take minutes to come up with the boot error issue. I have yet to fully solve that problem and am still waiting on parts to arrive to fix mine.

They keyboard is just a matrix between the data lines and the address lines, so all kinds of bad stuff can happen when the keyboard isn't working properly since it's on the data and address lines directly, which can cause issues. There is no keyboard controller.

On mine, I manually activated the FDD and was able to boot even though the disk select line had failed on my machine, but if you're not getting the boot error after a minute or so, it's not a select issue - it's that the enter key isn't working.

You can also try Shift Quote to switch the A: to the right and initiate boot without pressing enter.

Hotwiring enter is another method - Third pin from the left, Second pin from the right, both on the top row is enter. You can use dupont connectors.

DONT try this without a proper connector - eg, with a bit of wire - because there's 12v next to one of them, and if you short out 12v to a signal line you will blow the chip at best, and potentially worse.

Here's a photo.
This is all quite interesting and helpful, thank you! I should have noted in my first post that I did try swapping keyboards and got the same result, which makes me feel like the fault is more likely in the Osborne itself rather than the keyboards.
Also, make sure you use a SSSD disk and not a SSDD disk to boot from as the earlier models most likely do not have the Double Density adapter.
I actually ordered a couple bundles of single-density disks from an eBay vendor when I first started working on these machines - that's what I'm using for the boot-test in the video I linked in the first post. I do need to sort out which (if any) of my Osborne's have the Double-Density upgrade!
You can also check the key matrix with a multimeter.

The most likely cause of your issue is a shorted switch contact - so you are going to have to fix the keyboard anyhow!

Dave
See the above comment - only this machine exhibits this behavior, even after I swap between keyboards, which I think indicates the fault is in the system vs. the keyboard(s).
True, but it's nice to check if it's booting first.

Oh, and if you haven't checked the filter caps on the PSU, turn it off and check them. They WILL blow if they have not already. The RIFA filter caps will not survive old age... And they can cause damage to other components if they hit it with a molten jet of metal, which is what they shoot out when they smoke.
I've just received some replacement cap-kits for Osborne PSU's, hope to get them installed this weekend!

Full disclosure: the reason I'm scrambling to get one of my Osborne's working semi-reliably is that I'm going to be lecturing about + exhibiting several of my vintage systems for Golden Gate University next week. My goal is to have each of the machines I'm speaking about during my presentation working well enough for in-person attendees to try out and learn about firsthand, but of course with vintage hardware and software, "rock solid reliability" is rarely a given. I'm grateful for the advice you guys are offering, and given the impending deadline I'm facing, please forgive me in advance for when I come back with more urgent questions!

Huxley
 
>>> See the above comment - only this machine exhibits this behavior, even after I swap between keyboards, which I think indicates the fault is in the system vs. the keyboard(s).

Ah, you didn't state that previously...

There is no 'keyboard controller' as such - just buffers UE12, UE13 and UE14.

Even if there is a short circuit on the keyboard somewhere - that should not cause it to constantly reset. Generate a character causing the BIOS bootstrap to look like it is rebooting - yes.

I can't see any power on this connector though (other than GND).

Do you have an oscilloscope or logic probe to test the logic out?

If you connect the keyboard, you should be able to look at UE12 pins 2, 4, 6, 8, 12, 14, 16 and 18 and ensure all of the pins are HIGH.

Dave
 
Last edited:
Even if there is a short circuit on the keyboard somewhere - that should not cause it to constantly reset. Generate a character causing the BIOS bootstrap to look like it is rebooting - yes.
It does look that way because on power on it waits for you to press a key. If that key is A ( or another indicating to try to boot from B ) it tries to do that. If you ( or a busted keyboard ) press anything else, it repaints the menu - which looks like its continuously resetting. Detatching the keyboard and mapping out the scan lines ( as mentioned above ) is the way to approach it. Some times just poking / prodding the shorted key membrane will revive it. I have also used a hypodermic needle to inject a little air to separate them.
 
Quick update: I just took the front panel off the "MICRO CZEK" O1 that's doing the endless-resetting thing, and it does look to me like it has a ScreenPAC in there, yes?

IMG_9538.jpeg

IMG_9539.jpeg

Huxley
 
Technical manual can be found here: http://dunfield.classiccmp.org/osborne/index.htm

Yes, that is a ScreenPAC and a DD adapter - :) Nice model.

Jumping the enter key is a quick way to check if it's the circuitry or the keyboard - it makes for a quick test and if the DD light comes on, you're good to go.

The PCB is the same in all Osborne 1's so if you have a few, mix-and-match is a valid outcome also.

If the ScreenPAC is working, you don't need to mind it much - It can cause issues, but generally, it would cause screen related issues - though it could also potentially affect the z80 socket affecting some processor function - unfortunately, it's not easy to remove or disable for testing, so if you do go to remove it, the sockets will all be pretty damaged beneath it - I desoldered my z80 and screen ROM socket and replaced them with 0.1" header sockets which are more closely aligned to the ScreenPAC pins.

Your first goal will be to get the "Enter" key recognized by the computer, even if it's just a wire jumper. If it's not recognizing it, there are a few chips related to it's function that you may need to look at, but the ScreenPAC will make it difficult to troubleshoot, so try everything you can externally before going to work with a ScreenPAC in there.

The outcome of that goal is a regular repeating message that it couldn't boot that slowly scroll the screen away. ( or it boots is valid also )

Also, old disks can shred and affect your FDD reading. That happened to me... Booted once, then not... Had to clean the heads and installed a Gotek.

If you can't get it to put up the boot message by jumping the Enter key, then plan a big repair session and I'd recommend replacing the sockets on the main board. If you solder some 0.1" pin headers to a 40 pin socket and a 24 pin socket, you can disconnect the ScreenPAC while testing and troubleshooting the main PCB and just plug in the CPU and Video ROM on extended legs :) But trust me that the sockets on the main PCB will be so badly damaged once you remove the ScreenPAC that you won't be able to reuse them. They are just too bent up. They were never intended to facilitate such large pins.
 
Well, I'm coming in hot with another update: the problem is almost certainly my keyboard(s).

As best I can tell, both of my O1 keyboards are suffering from "membrane rot" (or whatever the proper term is). I've cleaned them out (they were FILTHY inside) and carefully re-tested but it's pretty easy to see that the plastic layers of the membrane are warped and separated, and connecting them to my otherwise-functional O1's sends the system into the reset loop.

Occasionally I can get an O1 to boot by mashing keys, but then it flips out with constant beeps and garbage characters filling the screen. I'm bummed out (I really want a functional original O1!) but I guess my next step is just to focus on finding or building a replacement keyboard for these machines. I've seen a few threads here about projects like that, but most of them are older and aren't clear if the original poster actually has drop-in replacement keyboards available (or even DIY kits). If anyone has something like this, I'd be ready/willing/able to work out a deal - please let me know!
 
The only one I can recall seeing was in New Zealand and of course, the one that's been in these threads lately.

There's a lot of good O1 keyboards around though, so finding one may not be too difficult.

I recently bought one here in Australia so I can fix an earlier O1... I am hoping the insides of a later O1B model can be fitted to the O1A but haven't checked yet. All my O1Bs are good and it's the O1A keyboard that seems to have issues.

I don't know enough at this stage to offer advice, but I will follow your progress as you fix yours since it may be relevant to me also.
 
Well, I'm coming in hot with another update: the problem is almost certainly my keyboard(s).

As best I can tell, both of my O1 keyboards are suffering from "membrane rot" (or whatever the proper term is). I've cleaned them out (they were FILTHY inside) and carefully re-tested but it's pretty easy to see that the plastic layers of the membrane are warped and separated, and connecting them to my otherwise-functional O1's sends the system into the reset loop.

Occasionally I can get an O1 to boot by mashing keys, but then it flips out with constant beeps and garbage characters filling the screen. I'm bummed out (I really want a functional original O1!) but I guess my next step is just to focus on finding or building a replacement keyboard for these machines. I've seen a few threads here about projects like that, but most of them are older and aren't clear if the original poster actually has drop-in replacement keyboards available (or even DIY kits). If anyone has something like this, I'd be ready/willing/able to work out a deal - please let me know!
I have fixed mine by pointing the hot air gun on my SMD rework station at the back of the keyboard (the metal support) for ~30s at ~300 degC until the short cleared. The first time I did it needed more time. This does require using a multimeter to find out which two keys are shorted, so you can direct the heat.

I also found that once I'd fixed it, closing the computer up with the keyboard would end up in the same keys getting shorted again. This was because the cable's fold pressed on those keys (7&8) when closed, so that must have been causing it.
 
I have just re-read the posts in the thread (and watched your video - now I realise it was a video)...

The assumption was made that one of the three keyboards must be working and, therefore, that the fault must be with the machine. However, that assumption is actually a guess unless you either (a) have checked the keyboard out with a multimeter (or a keyboard tester) or (b) got a keyboard that works fine on another machine.

The keyboard is quite simple to test with a multimeter as it is just a key matrix. Sit down with a multimeter, the keyboard and a pen and paper and test out all of the key press combinations - checking for both short circuits and open circuits. It is pretty boring to do - but it is fairly easy and you can sit there with a glass of wine and watch the TV in the background while you do it!

I wouldn't leave your 'so called' known good boot disk inside the floppy drive when you power the machine OFF or ON! Please remove all floppy disks before you do this. You can corrupt data on the floppy disk if you do this! I suspect your 'known good boot disk' is not any more...

Also, I would put a write protect tab on your boot floppy disk to save the machine accidentally erasing it. I speak from experience here!

Dave
 
Back
Top