UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers


Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
Thread Tools
Old 27th Aug 2016, 10:26 pm   #1
ricard's Avatar
Join Date: Mar 2006
Location: Lund, Sweden
Posts: 1,583
Default Toshiba T1000 display

I have this Toshiba T1000 computer, which is one of the first laptops to hit the market. Even today it's actually reasonable compact. It does lack amenities such as a hard drive though (drive C: is a ROM disk, holding the OS). I've held on to it not only for historical reasons, but because a) Running DOS, it is compatible with my ALL-03 EPROM programmer, and b) I've added MIDI ports on it, and essentially lacking an OS, I know that it won't miss a MIDI byte when capturing incoming data, nor get the data mangled by an OS driver.

For about a year, the display contrast has been rather dim when starting up, but over a 10 or 20 minute period got better and better. Finally when I turned it on the other day that didn't help either. I could type commands like DIR A: and see the FD light come on so I figured it was only the display which was at fault.

The Hardware Maintenance Manual is freely available on the net, unfortunately there are no schematics or detailed technical information on the main board itself. One tidbit of information was mentioned about the display however: it uses a -20V bias voltage. Measuring the connections to the display yielded nothing less than -5V, so there was something to go on.

Poking and measuring my way around the (switch mode) PSU section, I found the -20V regulator, and from what I could tell, it seemed to be working: the switch transistor was working, sending a positive pulse through the switching coil, and the coil was generating a negative voltage pulse just not enough of it! Cooling spray caused fluctuation in the readings, more than I'd expected. A faulty semiconductor perhaps? More examination revealed that the actual regulator chip was a TL1451 which was mounted fairly far away from the switch transistor and other components that I'd cooled.

Had to call it a day, and the next day, the regulator was happily providing -20V. Damn. I could just mount the whole thing together, but the problem would then surely manifest itself sooner or later again. Cooling spray still caused more fluctuation than I'd expect (several volts). Finally I started looking hard at a voltage divider, connected between the -20V output and a 2.5V reference voltage, yielding 1.24V, which in turn was connected to the TL1451 sense input. When cooling one of the resistors (R262, 56k) the output would go towards -16V, but the 1.24V was stable. So, probably the resistor was unstable. My spraying it with cooling spray the night before possibly (temporarily) cured it.

Didn't have an SMD resistor on hand, but there was enough space to put in a standard 1/8W . Cooling spray now only changes the -20V line by a couple of millivolts, so it's a definite improvement.

Not 100% sure I've actually fixed it as it worked tonight even before replacing the resistor, but the increased thermal stability points in the right direction.

After fixing it, I've noticed it has trouble booting, which I don't remember from before, so perhaps there's another unstable resistor in the PSU somewhere...

Incidentally, this machine is actually Y2K compliant, proudly displaying the date as Sat Aug 27, 2016 (not shown in the pictures below, as I took them before I'd set the date). Hats off to Toshiba for getting that one right!

(The grey and black patch wires running across the main board are not original; the black ones select an alternate master clock for the UART when running MIDI, and the grey ones piggyback the Rx and Tx signals on the serial port, connecting them (with appropriate interface circuitrly) to the DIN connectors mounted on the top of the machine.)
Attached Thumbnails
Click image for larger version

Name:	DSC08342.jpg
Views:	345
Size:	83.3 KB
ID:	129277   Click image for larger version

Name:	DSC08324.jpg
Views:	472
Size:	161.3 KB
ID:	129278   Click image for larger version

Name:	DSC08319.jpg
Views:	413
Size:	174.6 KB
ID:	129279   Click image for larger version

Name:	DSC08320.jpg
Views:	413
Size:	143.3 KB
ID:	129280   Click image for larger version

Name:	DSC08336.jpg
Views:	439
Size:	111.0 KB
ID:	129281  

ricard is offline  
Old 28th Aug 2016, 9:23 am   #2
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 8,526
Default Re: Toshiba T1000 display

I admire your persistence, and I have to admit a SM resistor would not have been high on my list of suspects given your symptoms (slow decline over a long period of time) - that would normally make me look at all the electrolytic capacitors first.

For the 'reluctance to boot' problem, I would still look in that area as the PSU may be unable to keep up its output during the initial heavy inrush on power-on, but OK after that. If you have access to a storage scope it would be interesting to know what happens to the supply rails in the first few tens of milliseconds after power-on.

Given the age of the unit I think it might still be worth inspecting all the PSU capacitors out of circuit with an ESR meter or just re-capping the power supply section with suitably low-ESR high temperature replacements.

Incidentally, like you, I have a Hi-Lo Systems device programmer (an ALL-07 in my case) and I also maintain a DOS Laptop (A Brother, in my case) in working order to keep that running.
SiriusHardware is online now  
Closed Thread

Thread Tools

All times are GMT. The time now is 12:40 am.

All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at
Failure to heed this warning may result in death or serious injury to yourself and/or others.

Powered by vBulletin®
Copyright ©2000 - 2022, vBulletin Solutions, Inc.
Copyright ©2002 - 2021, Paul Stenning.