Toshiba T1200 restoration + 3inONEder install

[xjr358]

Hello everyone,
This time I want to share my experience with a T1200 repair/restoration project. I bought the laptop from eBay, it was a bit dusty and came with an original Toshiba power adapter. Thanks to this forum I knew this thing should be recapped there are leaking caps at the PSU and display backlit inverter. So I replaced them and it turned on and booted up from the floppy drive. As the built-in 20Mb MFM hard drive has not shown signs of life I decided to use 3inONEder Compact Flash adapter made by @Conventional Memories - and this is a complete game changer for Toshiba (and some other) laptops of that era. I installed the adapter and after some trials and errors managed to install DOS 5.0 at 1Gb compact flash card. It works like magic and allows you to do a lot of experiments with different OS and software. Huge thanks to @Conventional Memories for this product!
After some cleaning machine turned out lovely, there were a couple of issues that need to be addressed:

1. Main battery to Lithium 18650 conversion
2. CMOS battery replacement to CR2032

 

[xjr358]

T1200's main battery is 7.2V so I decided to use four 18650 elements in 2s2p (two serial + two parallel) with 2s BMS board and it works well. I replaced a built-in 2.4 rechargeable battery with a CR2032 with a plastic cradle connected via a diode to prevent charging voltage from the motherboard (a method that I discovered from the Necroware youtube channel). There's a third battery (the yellow one) for the RAM that allows you to change the main battery without losing the information in the RAM. I don't find anything with the recuired voltage (4.8V) and size so I just removed it from the laptop.

 

[xjr358]

There are questions left - Is the high-pitch noise from the display inverter inevitable? It's not that loud but noticeable. Or it's my exemplar problem?
BTW
You can activate composite Video Out by pressing Fn+PgDwn and it works very well!

 

[neosunrise]

Nice work! Congrats on the successful restoration. Do you know where I can purchase the 3-in-1 card? It’s sold out on eBay.

 

[Conventional Memories]

Awesome, looks great.
Happy to see another T1200 restored and running with it's original PSU, often these get so destroyed by capacitors leaking that they require mayor rework.

The whine from the backlight inverter is normal on these, nothing to be worried about. I think they always were like this.

How that 18650 battery replacement working out? I've been thinking about doing something like this on some of my machines. Although I don't really use them away from the wall it would be a fun project and a good way to use some of the recycled 18650s I've been hoarding.
Is the BMS outputting a constant voltage?
Does the low voltage detection in the T1200 work?

Thanks for the shout out btw happy to see another one of my expansion cards put to good use.

Nice work! Congrats on the successful restoration. Do you know where I can purchase the 3-in-1 card? It’s sold out on eBay.

Thanks for the interest
They are currently all sold out unfortunately. But I'll be selling another batch in a couple of months.
I'll update my website when I have an exact date. https://conventionalmemories.com/wiki

 

[neosunrise]

Awesome, looks great.
Happy to see another T1200 restored and running with it's original PSU, often these get so destroyed by capacitors leaking that they require mayor rework.

The whine from the backlight inverter is normal on these, nothing to be worried about. I think they always were like this.

How that 18650 battery replacement working out? I've been thinking about doing something like this on some of my machines. Although I don't really use them away from the wall it would be a fun project and a good way to use some of the recycled 18650s I've been hoarding.
Is the BMS outputting a constant voltage?
Does the low voltage detection in the T1200 work?

Thanks for the shout out btw happy to see another one of my expansion cards put to good use.


Thanks for the interest
They are currently all sold out unfortunately. But I'll be selling another batch in a couple of months.
I'll update my website when I have an exact date. https://conventionalmemories.com/wiki

That’s great to hear! Just to confirm that it can be used to replace JVC JD-3824? I have an NEC Multispeed HD which looks very similar to OP’s laptop that uses this hard drive. The head is stuck to the platter. I’m hoping to get a modern solution for it.

 

[Conventional Memories]

That’s great to hear! Just to confirm that it can be used to replace JVC JD-3824? I have an NEC Multispeed HD which looks very similar to OP’s laptop that uses this hard drive. The head is stuck to the platter. I’m hoping to get a modern solution for it.

I'm afraid that won't work. These expansion cards are more like ISA cards than direct HDD replacements. They connect to the Toshiba expansion slot which exposes an ISA compatible bus. They don't work with the internal HDD controller but contain the XTide bios.

 

[xjr358]

How that 18650 battery replacement working out? I've been thinking about doing something like this on some of my machines. Although I don't really use them away from the wall it would be a fun project and a good way to use some of the recycled 18650s I've been hoarding.
Is the BMS outputting a constant voltage?
Does the low voltage detection in the T1200 work?

It turned out well - now I can take T1200 to the pub, show it to friends, and play some games. It works from the battery for at least two-three hours, so I don't know if a "low battery" LED works as it should. But what I do know is that when the batteries are charged that indicator near the DC jack is turned green (it's red during the process) which happens (green light) when I just unplug the battery as well. So I guess that the BMS does not provide a constant voltage or constant current (well, it is not supposed to do so). It's pretty dumb and only cuts off the battery when it reaches a predefined voltage (~8.4V for 2 serial-connected batteries, overcharge protection) and protects it from deep discharge), it makes a charge balancing as well that serial-connected batteries able to charge evenly. When the battery is charging, the voltage is around 7.4V, (it's around 12V when the battery is unplugged). I don't worry about the charging current as the ballast resistors in PSU are barely warm.

I know that a constant voltage / constant current (CVCC) charging is better for the 18650 but I think for my use case it's good enough as it is. (maybe an intellectual charger project one day).

 

[Conventional Memories]

It turned out well - now I can take T1200 to the pub, show it to friends, and play some games. It works from the battery for at least two-three hours, so I don't know if a "low battery" LED works as it should. But what I do know is that when the batteries are charged that indicator near the DC jack is turned green (it's red during the process) which happens (green light) when I just unplug the battery as well. So I guess that the BMS does not provide a constant voltage or constant current (well, it is not supposed to do so). It's pretty dumb and only cuts off the battery when it reaches a predefined voltage (~8.4V for 2 serial-connected batteries, overcharge protection) and protects it from deep discharge), it makes a charge balancing as well that serial-connected batteries able to charge evenly. When the battery is charging, the voltage is around 7.4V, (it's around 12V when the battery is unplugged). I don't worry about the charging current as the ballast resistors in PSU are barely warm.

I know that a constant voltage / constant current (CVCC) charging is better for the 18650 but I think for my use case it's good enough as it is. (maybe an intellectual charger project one day).

Awesome, thanks for the info.
I would love to try making one and do some testing. I wonder if the BMS is enough protection with a weird charging setup like this. Will have to do some more research.

 

[xjr358]

Awesome, thanks for the info.
I would love to try making one and do some testing. I wonder if the BMS is enough protection with a weird charging setup like this. Will have to do some more research.

You're very welcome! Please share the results with us if you'll do so!

 

[neosunrise]

I'm afraid that won't work. These expansion cards are more like ISA cards than direct HDD replacements. They connect to the Toshiba expansion slot which exposes an ISA compatible bus. They don't work with the internal HDD controller but contain the XTide bios.

This is what my NEC's motherboard and the HDD controller card look like. The motherboard has a 34-pin connector, not sure if it's ISA compatible. I really hope to get this computer up and running with a working hard drive/CF card.

 

[Conventional Memories]

This is what my NEC's motherboard and the HDD controller card look like. The motherboard has a 34-pin connector, not sure if it's ISA compatible. I really hope to get this computer up and running with a working hard drive/CF card.

you should take a ook at this thread, we were just discussing these hard drives. it might be worth trying to repair it. Absolutely no guarantee that will work with this 20mb drive but it's certainly worth a try if it's dead anyway.

https://forum.vcfed.org/index.php?threads/toshiba-t3100-20-with-a-dead-drive.1246210/#post-1368713

Also take a look at the video posted. THEtechknight made a pinout for the HDD controller from the T3100/20 which looks very similar to your controller. Unfortunately it doesn't look like there is enough there to support an xtide controller.

 

[Conventional Memories]

I have been doing some research on the battery replacement. Mainly because it is a fun project and because it would be really cool to have one of the T1200 unique features up and running, that is it was the first portable computer to have a "sleep" function. It could be unplugged from the wall for up to 10 days and still be ready to pick up where you left off.

Now, I have my doubts a simple BMS would do the trick, Lithium batteries can be very unsafe when overcharged or drained too far.

Some tests on the T1200: (actually did it on two T1200 laptops to make sure everything was consistant)
Charger puts out 11.7v into the NICAD battery (6 cells in series) for a total of 1.95v per cell. this seems a bit high, I would have expected something more like 1.5v per cell. but after some more reading I can confirm that is normal for "fast charging".

The battery low warning happens at 7.20v (blinking LED and beeping)

The automatic sleep happens at 7.1v and suspends everything into ram.


Now lets take a look at the 2s (series) 18650 battery you made.
A normal charge voltage would be 2x 4.25v = 8.5v max
And a safe cutoff voltage for an 18650 is about 2.8v x2 =5.6v

lets start with the cutoff voltage of 5.6v, that seems a lot less than the cutoff from the T1200 of 7.1v (3.55v per 18650)
But if we look at the discharge curve for a typical 18650 we can see that there is only about 15-20% of the capacity left at 3.55v per cell.

So this will work very well I think, it is actually recommended not to discharge the last 15% of 18650 batteries if you want to maximize how long they will last.

Next is the charging current, I have to do some more tests on the T1200 to determine the actual current it charges at. But in general NICAD batteries will be charged at a constant current of about 1/5th or 1/10th their capacity, either way that should be way slower than what 18650 cells are charged at. So this will work fine I think, although it will charge very slowly. I will need to do some more tests on the T1200 to make sure.

Now the last and most important question, How about the charge voltage?
The T1200 puts out 11.7v (cv) and the charge voltage for the 2s 18650s should be 8.5v max, that is 5.85v vs 4.25v per cell. that 5.85v per cell is way to high and will destroy your batteries maybe even in an explosive way.
Now the real question is can a simple BMS take that 11.7v input and regulate it down to 8.5v or lower?
It seems really hard to find good info on this. It certainly doesn't seem like they are designed to do that.
Some 2s BMS modules do list the charge input voltage, they all seem to be 8.4v - 9v
That's not nearly enough to take the 11.7v input.
I have found some that list the input as "8.4v-9v (12v max)", but I have no idea if they are designed to actually take almost 12v for the entire charge.

I would like to do some more reading and testing, but at this point I do not feel that it is actually safe to do this. These BMS boards are only meant to be used as a protection of the batteries and to occasionally balance the cells. But they still require a charge circuit designed for 18650 cells. these don't contain any charging capability.

I will order some of the BMS boards that say "8.4v-9v (12v max)" and look if they can actually produce a safe charging voltage with 11.7v input without overheating.

@xjr358 , what BMS are you using? I would be surprised if it is designed to take the 11.7v

 

[xjr358]

Hi!
Thank you for such a detailed answer! I will read it tomorrow thoroughly and reply. For now - yes, I understand how dangerous these elements could be. And to make things worse look what I did
14500 with an AA cradle + BMS for the hard RAM

 

[xjr358]

Hello, everyone.
Hi @Conventional Memories so what we have RN:

1. The discharge is OK (I can report that the laptop beeps and the "low battery" LED is turning on when the battery is low)
2. Charging current is kind of okayish (though slow)
3. Now the charging voltage - it's higher than it should be WITHOUT the battery! But if you plug the battery it drops to 8.4V so I suppose that we're OK here as well (look at pics in the attachment)

BTW I can report that the trick with the RAM battery works well - it gives only 4.05V when charged (instead of the 4.8V of the Toshiba battery that has been installed) but that is enough to prevent the message "DATA IN HARD RAM WAS LOST" from appearing (for some amount of time). The recipe is: 14500 rechargeable battery + standard AA battery's cradle + 1 battery BMS for the 18650 (yes I know that the charge/discharge current is not optimal but... I feel lucky)

Your Q - "Now the real question is can a simple BMS take that 11.7v input and regulate it down to 8.5v or lower?" - no way it can do that. BMS is just a couple of MOSFETs with a function to limit the floor and ceiling of the level of charge. What BMS did I use? The cheapest one (1.25USD) pic in attch. The good news is that WE DON'T NEED THAT in our case because of these two resistors. We call it a "ballast" resistor in Russian, don't know what should I call it in English coz my English is not much better than my knowledge of electronics. That's the place where the voltage is transformed into heat!

Now the big question:

The Prince of Persia doesn't want to walk slowly - the "Shift" key is not working for that (though I can pick up the sword with it and it works well in DOS)

And a couple of small ones:

1. Is it possible to use a full 1024KB of RAM for emm... RAM not for the 640k + 384k of the "Hard Ram"
2. What brand of Belgium beer do you prefer?

BTW for safety reasons - I managed to keep Toshiba's thermal fuse.

 

[Conventional Memories]

Still not safe at all IMHO, the voltage drops to the voltage of the battery pack but the T1200 PSU is still trying to charge the battery up to 11.7v, it is still pumping current into the pack.
Now depending on what BMS you are using it might cut off the charge at around 4.3v as a safety measure, but that's just as it sounds to act as a safety. That is still not a normal voltage to cut of charging and what a BMS is designed to do.
I don't think I can convince you, but this might just be a warning to others coming across this post, not to try and do this.

Why not just add a couple of diodes and a buck converter? still not ideal, but a lot better.
Or if you want a faster charging solution, why not add a usb-c charger in the battery pack with a couple of diodes?

If you want the T1200 without an extra external plug to do the fast charging, why not tap 12v from the power plug to add a charge controller? all better, and safer solutions that can be done for under €4 in parts.

 

[xjr358]

Why not just add a couple of diodes and a buck converter? still not ideal, but a lot better.
Or if you want a faster charging solution, why not add a usb-c charger in the battery pack with a couple of diodes?

Thank you very much for your time. Please can you elaborate on this? Is it possible to make this setup safer?

 

[xjr358]

I'm not trying to prove something, I need a better solution to my problem with this laptop, and as am I judge by the quality of your 3inOneEnder you're the right guy to ask.
Edited:
If my messages sound wrong please consider there's a possibility that because of my bad english

 

[Conventional Memories]

Thank you very much for your time. Please can you elaborate on this? Is it possible to make this setup safer?

I'm not trying to prove something, I need a better solution to my problem with this laptop, and as am I judge by the quality of your 3inOneEnder you're the right guy to ask.
Edited:
If my messages sound wrong please consider there's a possibility that because of my bad english

Sorry for the delay, Don't worry, no offence taken. Just wanted to make sure nobody tries to replicate this and risks the chanse of starting a fire when leaving it charging overnight.

I think what I would do, (to be clear I am definitely not an expert in any of this) Is to add a diode coming from the the original charge connector from the PSU so that it can only draw power from the battery and not charge it.
And then tap 12v from the power plug (solder wires to the bottom of the pcb), connect this 12v into a proper charge controller for a 2S battery with the propper current rating for your cells. (If you can't find a charger that takes 12v input, you could add a small buck converter.)