NanTan Early 1990's laptops Guide (386/486/Pentium)

[3lectr1c]

Someone on a discord server I'm in extracted the Nan Tan 9200's boot chime from the BIOS ROM for fun, so here it is

New info I just got from someone else, apparently the boot chime is literally just a file called "QUEST1.WAV" that was included in ESS's audio driver package, lol.

 

[3lectr1c]

Thank you This is literally the first piece of info I have seen on the internet lol. How much do you think a fully functional one with original charger, windows 3.1, all of the paperwork, mouse's and the leather bag would go for?

Forgot to reply, I apologize! Fully working, considering it's a passive matrix unit, I'd say $100-150.

 

[RetroShare]

Hello Nantan friends,

I've bought an FMA86T from the flea market yesterday. Works great except for the fan and the cmos battery.
I tried to open it, but something doesn't let me open it. Anyone here who successfully opened it?
Maybe I have to release the keyboard and the mechanism is the same like the other notebooks. But how?
I saw two clips at the bottom of the keyboard, which I can push back. But no luck getting it out.

 

[sjgallagher2]

Picked up a DataWorld LP320 (a rebranded Nan Tan 8100 as I understand it) from an estate sale, kept in a nice carrying back in a basement for many years, power brick included, but didn't power on.


Did some preliminary maintenance (notably removed the CMOS battery, fortunately not very corroded, board underneath was fine). I was bummed about the power supply not working, it read +15.6V on the male pins, but when plugged in the voltage was floating and it couldn't drive anything. This led me to reverse engineering the whole power input board (though not the separate power converter board that actually derives the +12V, +5V, and -23V rails from the +15V input) in hopes of tracking down the issue. Also I like reverse engineering.

For posterity, here's what I've got so far.

The connector is a mini-DIN 8-pin, made this graphic with the pin numbers:



The cable has three wires (white, red, and black) and a shield with drain wire. The shield doesn't connect anywhere.

Pins	Cable connection	PCB connection
1,3	White wire	Ground? Disconnected on mine
2,5	Red wire	+15.6V IN
4,6	N/C	Ground
7,8	Black	+15.6V IN
Shield	Shield	N/C

The board I investigated:



You can overlay this mirrored and adjusted version of the bottom side with the top side photo above:


Here's the reverse engineered schematic so far, take voltages with a grain of salt since I was bypassing the connector at first and jumping the +15V in after the fuse and twin diodes (at JP2 in the schematic):

Named nets mostly come from the silkscreen, so I don't know what CL, BL, and VR mean, but I named VO1 and VO2 since they were Outputs.

The PCB, mostly there but with some wrong footprints (laziness) and missing connections (that are however included in the schematic):



The comparator section of the circuit seems like mostly power-good/over-voltage/under-voltage alerts. Haven't looked enough at the transistor section. On my board the big power resistor R36 was open, but since J2 doesn't have any connection it shouldn't matter to functionality.



Anyway, fun as this was, the problem ended up being with the PCB power connector not having any connection to ground. Only pins 4 and 6 were grounded, but those didn't have any connection in the cable, very confusing. Also confusing that the +15V was carried on two 22awg wires (red and black) in the cable, while the ground is presumably the single white wire (also 22awg), something else must be going on.

 

[sjgallagher2]

Other notes:

• Matches specs on MacDat, except the BIOS chip which is Phoenix 386 ROM BIOS PLUS (chip marked "PH242 | L0151988", copyright Phoenix 1987, so predates Phoenix BIOS PLUS v4)
• Same PSU model, KTX-8912A, branded DataWorld
• The DC/DC converter I was working on was marked "DC TO DC CONVERTER | CP/N: 8200 | UP/N: UCT-89104 | MADE IN TAIWAN | UNION CIRCLE"
• From the mags:
  • PC Magazine, March 12, 1991 - Dataworld Data LP/320 Fact File
    • "The 19-pound Data LP/320's weaknesses far outweigh any residual strengths, most notably the system's low prices, robust construction, and ability to run a full 80386 processor on batteries. Poort hardware performance makes this a tough recommendation."
  • InfoWorld, Sep. 3, 1990 - Evaluating and Scoring High-End 20-MHz 386-Based Portables

Now I'm driving it with a bench top power supply and it comes to life barely, but emits what seem like slow POST beep codes. Attaching the sound file. Anyone have any ideas what might be the issue?

My blind guess is that it's an Early POST beep that's too slow because of a clock issue or something. But I have no idea. Maybe that's the way these old machines sounded, but I imagined a more typical IBM PC style high pitch beep.

 

[3lectr1c]

Fantastic work on the power supply board!
The POST error it's throwing is 1-1-3: CMOS read/write error. My 8100 did the exact same thing, and the cause was that the battery corrosion had slightly oxidized the pins on the CMOS RAM chip, which is socketed right next to where the battery is located. For me, I just pulled the chip, applied contact cleaner, and then reseated it. That brought it right back to life.

 

[sjgallagher2]

Ah so that's how I interpret that code! thanks so much, I'll look around and see if I can track down the error. I wonder if it's the battery, I powered it by a benchtop supply at +6V but maybe I had it wrong somehow, I need to order a replacement cell already. My only experience with vintage PCs is a Compaq Portable I fixed up a couple years ago, so I'm just using a 1998 edition of Upgrading and Repairing PCs, and the PC Engineer's Reference Book to try and get a feel for things. Otherwise it's repair from first principles your website was a godsend for this laptop, thank you for the work.

 

[3lectr1c]

I'll look around and see if I can track down the error. I wonder if it's the battery, I powered it by a benchtop supply at +6V but maybe I had it wrong somehow, I need to order a replacement cell already.

Have you reseated and contact cleaned the CMOS RAM chip? I've found working on various laptops from this time that 1-1-3 is a favorite error code for them to throw, and can be for various reasons (usually that the laptop in question requires a working cmos battery to boot) but in this case we have a known cause in the bad contact. And I can say for 100% certainty that the 8100V doesn't require a CMOS battery or main battery to be installed for it to work.

 

[sjgallagher2]

Have you reseated and contact cleaned the CMOS RAM chip? I've found working on various laptops from this time that 1-1-3 is a favorite error code for them to throw, and can be for various reasons (usually that the laptop in question requires a working cmos battery to boot) but in this case we have a known cause in the bad contact. And I can say for 100% certainty that the 8100V doesn't require a CMOS battery or main battery to be installed for it to work.

I reseated the BIOS(?) chip, pictured below; didn't have contact cleaner so I made do with IPA and a brush. Still no dice though, will have to try a more proper clean. I assumed that chip was the BIOS ROM, but the POST beep codes still play even when that chip is completely removed, so I guess it's something else? Or the Early stage POST is stored somewhere else. I know you said CMOS RAM but that seems odd to me (if we're talking about the same chip), having a branded RAM chip separate from the rest of the RAM. I'll probably make a new thread for this issue or piggy back off the one you made (that I just found).

 

[3lectr1c]

File:Intel P8242.jpg - Wikimedia Commons

commons.wikimedia.org

It looks like that might actually be the keyboard controller. Weird, I swear I had looked up something related to that chip and found it was CMOS RAM. Dunno.

 

[DeltaDon]

The Intel P8242AH is this: https://www.alldatasheet.com/datasheet-pdf/pdf/66188/INTEL/P8242AH.html