Toshiba T1850: The Land Before Windows

One of my hobbies is Radios. It covers everything from building your own radios with a soldering iron to collecting old police radios and reprogramming them for your own use. It’s pretty fun. Keeps the nerds busy. However, as you dig deeper and deeper into the radio hobby, you will undoubtedly unearth some of the older radios. When the cops upgrade, they dump massive quantities of (for the most part, excellent) radios onto auction sites like eBay. Hams and other radio gurus sweep them up in great quantities. Until you find out one particularly sad fact: you need DOS to program them. And often, the slower the better. More on radios elsewhere, today we cover the DOSmonster.


The Specifications

Enter the Toshiba T1850. A stalwart of the laptop era: a relic from when men were men and boys were boys. Basically, it’s a 25 MHz Intel 386 Architecture Laptop. It’s a gigantic clunky beast which uses a NiCAD battery (remember those?) and has a glorious 3968KB of memory and an 80 MB (or optionally a 120MB) hard drive. It has lots of I/O: A serial port, a parallel port, a (optional) modem, and a PS/2 keybaord.

I first encountered this wondrous creature at my favorite hunting ground, the MIT Swapfest. For those who aren’t familiar, it’s a large festival of nerds buying and selling anything and everything electronic or metal or wood that they have in their vast collections.

The Troubleshooting Begins

I was able to boot my T1850 with ease, however it didn’t find an operating system and that’s where the trouble began. The BIOS (which I eventually discovered could be reached by pressing the ESC key at boot) did not reveal a hard drive, even though there was most clearly one plugged in. I promptly ordered a CF to IDE adapter and plugged it into the computer.

But then the puzzling began: it didn’t see the adapter. A tremendous google search on the (nearly zero) websites describing this laptop revealed my problem. The laptop uses the IBM AT bus, a pre ATA bus standard which, although it has basically the same pinout, is completely incompatible with the IDE ATA standard. The bus width is 16 bits as opposed to the IDE’s 32, and the problems continue.

The Battery (an aside)

Both of my T1850s came with original NiCD battery packs. But there was the problem: the computers didn’t recognize that they had batteries, and they didn’t charge them. While Lithium cells today will “die” if they are discharged below a minimum level, NiCDs do not have the same vulnerability. My hypothesis: the low voltage caused by 15 years of storage caused the computer not to recognize the presence of the cell. Strapping a charger to the cell, I boosted the voltage to its nominal value with a few minutes of charging. Low and behold, a full 45 seconds of runtime on a pack! (I’ll probably re-cell one of these packs. I have a desire to bring this computer into lecture to take notes. I’m that guy.)

The Fix Appears

So I gave in. To eBay I ran and found another T1850 with functional 120 MB (!) hard drive. I will admit to being slightly scared, as I now have one (1) functional hard drive for my computer. That hard drive was manufactured probably before I was born (10/1/92), and it, like me, is not getting any younger.

So all in all, between the two computers I now have one completely functional Toshiba T1850 laptop. Thanks to the readiness of at least one more eBay unit available, I’ll probably purchase another one just to make sure I don’t end up dead in the water. Unfortunately, one of my units is locked in my dorm for the next week, but I’ll most certainly update this once I get it running DOS. The hardware serial port will enable rapid and efficient programming of radios that require processors this slow (primarily the HT1000 and also GTX lab RSS). But more on that once the madness continues.

3 thoughts on “Toshiba T1850: The Land Before Windows”

  1. Hi, Nicholas. Great to see you trying to get this one up an running. I had a T1850 in 1993, “inherited” from my father who got an IBM or a newer Satellite back then.
    I used this one with Windows 3.1, DOS 5.0 probably, WordPerfect, lots of gaming and accessing BBS with an external modem, which would leave me without the mouse in Windows so I had to resort to DOS communication programs, possibly Telix, since I am pretty sure I used Terminate on a 486 and not on this one.
    Those were the days… 🙂

  2. The Toshiba T1850C does take standard IDE (PATA) 2.5″ drives. The AT-Bus you mention refers to the memory+I/O bus of the of the 386SX itself, not that of the separate hard-disk interface (PATA).

    However, the BIOS in the T1850C will only recognise small-capacity drives with a specific limited set of number of heads/tracks/sectors-per-track (unlike modern machines, it is not smart enough to *ask* the drive for it’s logical geometry). In particular, the Conner CP2084 (80Mb), Toshiba MK-2024FC (80Mb) and Conner CP2124 (120Mb) all work fine in the T1850C – those are the drives that Toshiba provided as standard with the machine. All three of those also work perfectly in other much newer laptops with IDE/PATA bays – for instance, Dell Latitude D630 (2006, Intel Pentium M 750 @ 1600 MHz), and so on, and so on.

    1. Ah, thank you for the clarification. There must have been a bit more to the problem given that the controller would choke on an unpartitioned drive, but I probably could have gotten around that by writing a partition table to the drive externally.

      After being able to run all of the applications I needed on a Toughbook running Win98 DOS mode I (alas) ended up ditching this experiment :/

Leave a Reply